1

I N T R O D U C T I O N

Introduction

What do we really know about the Internet? The Internet is

an unregulated band of computers that continuously share
information. This information may contain fundamental
truths, profound thoughts and visions, or integrated deceit
and outright lies. How many people are under the impres-
sion that because they found information on the Web, that
the information is absolute truth?

The only thing that can definitely be said about the

Internet is that it is a network of computers that provides a
vast distribution medium to a countless number of people.
However, tapping into this vast resource base is only as
effective as the communication mechanism between you
and your target audience and the ability of the audience to
receive and process the information. Speakers and televi-
sion screens are useless without a transmitter to translate
and receive radio and television signals. Likewise, the
multitudes of thin-client browsers are rendered useless
without an effective, application-oriented Web server. ■

01.1389-6 Intro 1 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 2 Introduction

The Purpose of This Book

With the integration of the Microsoft Internet Information Server 4.0, the Microsoft Transac-
tion Server, and Active Server Pages, the power of distributing worthwhile applications over
the Web is a reality, which leads to the vision of this book. This book is solely dedicated to
demonstrating the powerful functionality of Active Server Pages to deliver real-world applica-
tions over the Web.

In order to fully understand Web application development, you first have to recognize that this
environment is not the traditional development environment you are used to. Even though the
phrase “it’s just HTML,” echoed from many IS managers’ offices, many of the inherent develop-
ment issues taken for granted in traditional application development are not available in the
stateless world of HTTP. In addition, applications have become more content-centric rather
than process-centric. The new development team now includes content creators, graphic art-
ists, and network engineers, as well as traditional application developers, database program-
mers, and project managers.

Active Server Pages provide a powerful interface to develop programming logic that can be
used to implement and distribute your information through applications across the Web. The
book provides a systematic, hands-on approach to decrease the Web-based learning curve to
implement Active Server Page technology.

Who Should Use This Book

This book is targeted for the existing application developer that has some experience develop-
ing client/server applications, yet provides a basic foundation for newer developers. Remember,
the goal of this book is to provide a knowledge transfer mechanism that enables application
developers to transition their existing experience developing applications into building robust,
scalable, Web-based applications using Active Server Pages. At the same time, this book pro-
vides the foundation to help push beginning- and junior–level application developers into the
more advanced developer roles.

This book does not require developers to have experience developing applications with Active
Server Pages. This book is intended to leverage developers with some Microsoft Visual Basic
skills and database development skills and apply that knowledge to Web application develop-
ment. However, to help those with less experience in Web development and Visual Basic to
quickly come up to speed in developing Web-based applications, Parts I and II of the book are
designed to give an architecture perspective and a primer of VBScript. This primer is intended
not to teach every detail of VBScript, but is designed to show developers the needed program-
ming functionality to rapidly produce ASP applications.

01.1389-6 Intro 2 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 How the Book Is Organized 3

How the Book Is Organized

This book is organized in a manner that demonstrates what it takes to build, develop, and de-
liver Web-based applications. As you might expect, it starts with the basics and proceeds to the
more advanced topics. This organization goal is intended to create an environment that pro-
vides a smooth transition from the highly defined application development tools available today
to the much broader development environment that forms Web-based programming.

This book is separated into six parts, each part building off the preceding to demonstrate the
technical evolution of Active Server Pages. Each part of the book ends with an example chapter
that demonstrates the use of the various objects and tools covered in the preceding chapters.
Even though the content of each part may be technically different, all of the examples focus on
using Active Server scripts to build enterprise Web-based applications.

■ Part I Introduction to Active Server Pages

■ Part II VBScript: The Foundation of Active Server Pages
■ Part III Active Server Objects: The Essential Building Blocks
■ Part IV Active Server Components
■ Part V Database Management with Active Server Pages
■ Part VI Active Server Pages and the Enterprise Solution
■ Appendixes

Part I Introduction to Active Server Pages

Part I focuses on introducing the high-level, conceptual layers needed to understand Web-
based application development. This part of the book describes the differences between static
and dynamic Web sites and how delivering Web-enabled applications requires a host of inte-
grated systems working together.

Part II VBScript: The Foundation of Active Server Pages

Part II introduces the primary scripting language driving Active Server Pages functionality.
VBScript provides the interpretation or processing layer between the requesting browser and
server-side data storage units. Active Server Pages use VBScript to convert coded processing
logic into native HTML. The pure HTML output is then accessible to any HTML-compatible
browser type.

Part III Active Server Objects: The Essential Building Blocks

Part III demonstrates the core functionality of Active Server Pages. Active Server Pages are
comprised of six programmable objects that are used to manipulate, manage, and control the
various aspects of Web processing. This part focuses on identifying the scope and composition
of Active Server Pages applications and how each object can be utilized in delivering your Web-
based applications.

01.1389-6 Intro 3 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 4 Introduction

Part IV Active Server Components

Part IV discovers how components can be used to implement additional functionality on your
Web site and Web server applications. These components can be used to perform application
functionality that might not be made available through native VBScript. Furthermore, these
components provide the basic links to tap into other information stores accessible to the Web
servers. This part of the book closes with the creation of a sample Web site that uses most of
the installed Internet Information Server components.

Part V Database Management with Active Server Pages

Part V focuses on using the ActiveX Data Object to provide a rich set of database functionality.
Active Serve Pages can use the Active Data Object to insert, update, and delete information
from any ODBC-compliant database. This section wraps up by adding database functionality to
an online catalog Web site.

Part VI Active Server Pages and the Enterprise Solution

Part VI demonstrates how Active Server Pages technology is used within a variety of Web-
based systems to help build, manage, and distribute your Web applications. This part of the
book demonstrates using the Microsoft Transaction Server to maximize resource and database
connection-polling features, how to use the Visual InterDev ASP development tool, and how to
manage your site with the Internet Information Server 4.0. Further, this part of the book ex-
plains the role of the Microsoft Message Queue Server and how to build intranet/extranet
capabilities for your sites and applications. This part of the book ends with converting the
sample Web online catalog registration system into a distributed n-tier application database-
driven application.

Conventions Used in This Book

Que has over a decade of experience developing and publishing the most successful com-
puter books available. With that experience, we’ve learned what special features help readers
the most. Look for these special features throughout the book to enhance your learning
experience.

The following font conventions are used in this book to help make reading it easier:

■ Italic type is used to introduce new terms.

■ Screen messages, code listings, and command samples appear in monospace type.
■ Shortcut keys are denoted with underscores.

TIP Tips present advice on a quick or often overlooked procedure. These include shortcuts.

CAUTION
Cautions warn you about potential problems that may cause unexpected results or mistakes to avoid.

01.1389-6 Intro 4 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 Conventions Used in This Book 5

N O T E Notes present interesting or useful information that isn’t necessarily essential to the
discussion. A note provides additional information that may help you avoid problems or
offers advice that relates to the topic. ■

This is an exciting time to be an application developer. The new Web-based application develop-
ment paradigm presents exciting challenges in the world of application development. As we
begin to see how the Web is slowly being integrated in many aspects of our lives, this book is
intended to make designing, developing, and deploying ASP applications a reality.

01.1389-6 Intro 5 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 6 Introduction

01.1389-6 Intro 6 10/26/98, 8:47 AM

Brands3/Stone4 SEU Active Server Pages #1389-6 SRing 11/14/97 Intro lp#3
 P A R T

I
Introduction to Active Server Pages
1 Getting Started with Web Enabling: Creating Dynamic
Applications with Active Server Pages 9

2 Developing and Deploying Active Server Page

Applications 29

02.1389-6 Part I 7 10/26/98, 8:53 AM

Brands SEU Active Server Pages #1389-6 Part I SRing 10/28/97 lp#1
02.1389-6 Part I 8 10/26/98, 8:53 AM

Brands SEU Active Server Pages #1389-6 Part I SRing 10/28/97 lp#1
 1
9

C H A P T E R
Part
I
Ch

Static Content versus Dynamic

Getting Started with Content
Learn the differences between static

Web Enabling: Creating content and dynamic content.

The Changing Internet Applica-

Dynamic Applications tion Architecture

Learn how all the different pieces of
the extended Internet architecture
with Active Server fit together to deliver Web-based
applications.

Pages The Dynamic Content

Revolution
Learn how dynamic content applica-
tions have evolved over time, and
how COM has formed the founda-
tion of dynamic content.
The goal of this chapter is to provide the basic framework
and architectural concepts needed to Web-enable your Active Server Pages and the
applications. To accomplish this, I will identify and discuss ActiveX Platform
the ever-growing number of Web technologies that collec- Discover how Active Server Pages
tively are needed to build and deliver your Web-based and the ActiveX platform are used to
systems. The secondary goal of this chapter is to help extend Web server functionality and
define the impact of COM technology on Web develop- the benefits Active Server Pages
ment, the application developer, training and delivery bring to Web-enabling your applica-
requirements, and infrastructure costs. tions.

First and foremost, I want to clearly define what the term

Active Server Pages means to Web-based application devel-
opment. Active Server Pages is a new technology from
Microsoft that provides the capability for the Web server
to process application logic and then deliver standard
HTML to the client browser. The results can then be
delivered to a variety of client-side Web technologies,
such as standard HTML, ActiveX, Java, browser
plug-ins, and DHTML. Although fairly straightforward,
this simple concept has great ramifications in shifting
traditional applications to the Web. The first direct

03.1389-6 CH01 9 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 10 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

result of using Active Server Pages is that only HTML is sent to the client’s browser. ASPs do
not automatically send ActiveX controls to the browser. By default, ASPs send only ASCII text
to the browser. This lets any browser types running on any operating systems to access the
applications and workflow logic embedded in the ASP scripts.

Active Server Pages give you the ability to deliver more than just HTML. ASP enables you to
deliver HTML, client-side scripting, Web controls, and server-side processing and connectivity
features. ASP scripts can deliver client-side scripts, such as VBScript and JavaScript, to be
executed on the client’s browser. In addition, ASP can also deliver a wide range of Web func-
tionality by acting as a transfer vehicle for ActiveX controls, Java applets, and other Web com-
ponents. In addition, ASPs not only produce dynamic HTML depending on the client request,
but also provide the capability to tap into existing systems, such as databases, document re-
trieval services, mail servers, groupware servers, and other COM-based information servers.
ASPs now act as an HTML interpreter that was once only accessible through native interfaces,
such a Microsoft Exchange client, Lotus Notes client, or a customized Visual Basic application.
With the wide range of functionality, Active Server Pages and the Internet Information Server act
as a medium for porting existing applications to—and building new applications for—the Web. ■

Static versus Dynamic Content

The World Wide Web has progressed from Web pages that couldn’t support images to a global
network capable of supporting full-scale applications. This transition is made possible through
changing static information into dynamic, database-driven applications using logic executed on
the Web server.

Static Content
The World Wide Web was initially created to share textual documents around the world. These
textual documents were static documents that served primarily as online reference materials.
The term static is used to refer to these documents because the requesting user had no ability
to interact with the content delivered from the Web server. These documents were converted
from their original source type into Hypertext Markup Language (HTML). Once in HTML for-
mat, the document could be delivered from a Web server to any Web browser using the
Hypertext Transfer Protocol (HTTP). This static approach to creating and reading Web pages
and Web content is still used in many sites today to display personal information, corporate
profiles, or online reference material that does not frequently change (see Figure 1.1).

Dynamic HTML Content

The billboard approach, or the posting of static information on the Web, suited the needs for
general information dissemination. However, the Web user community started to require
greater functionality from this vast information medium called the Internet. This need for ad-
vanced functionality using dynamic content eventually grew into the need to develop and de-
ploy entire applications that used the World Wide Web as a global network. The term dynamic
describes the process of creating HTML content depending on the information that is sent or
submitted to the Web server. The Web server would process the information and convert the

03.1389-6 CH01 10 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 Static versus Dynamic Content 11

output into HTML. The resultant HTML, tailored for the specific user input and application Part
settings, is then delivered to the browser. Figure 1.2 shows an example of dynamic content at
the Federal Express Web site. The FedEx tracking site uses information entered by the user to
I
track the location of a package. If you know the Airbill tracking number, country destination, Ch
and ship date, the Web application searches the FedEx tracking information storage system
and returns the location of the package.
1

FIG. 1.1
Traditional corporate
Web sites are often
examples of static
content.

FIG. 1.2
Dynamic HTML enables
the HTML content to
change depending on
the information
submitted by the user.

03.1389-6 CH01 11 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 12 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

Initially, sites would process small pieces of information, such as accepting user information
from a customer feedback page. This information could be stored in text files, converted to an
e-mail message, or displayed back to the requesting user for verification. But eventually, these
smaller modules gained more flexibility by performing calculations and reading and writing to
databases until Web-based applications began to emerge. Soon, as the benefits of developing
and deploying applications over the World Wide Web became apparent, the conversion of stan-
dard desktop applications into a Web-based format had begun.

Designing and Deploying Web-Enabled Applications:

A Changing Architecture for Changing Needs
Developing Web-based applications relies on many network and application components work-
ing together to deliver information to the requesting client. In the once simple world of HTTP,
Web browsers extracted information from Web servers, as shown in Figure 1.3. This Web-
based architecture relied on existing network infrastructure and the TCP/IP protocol stack to
broadcast, carry, route, and assemble in-transit data. TCP/IP is not exclusively used with the
World Wide Web; it is also used in other services such as Gopher, File Transfer Protocol
(FTP), and Telnet.

FIG. 1.3 HTTP

The simple data-sharing
FTP
architecture that formed
the premise of the World Gopher
Wide Web.
Browser Internet
Web
Server

However, after some standardization of HTML languages, image specifications, and protocol
stacks, the world slowly began to see the Internet for what it is: a global network of computers.
This band of network computers is owned by a combination of public and private localized
networks. These networks consist of private and public organizations and government organi-
zations that pay and maintain these backbones.

Although there is not a centralized organization unit for controlling the Internet itself, there are
several organizations that propose standards and guidelines on how the Internet should be
used. For example, the Internet Activities Board (IAB) is responsible for reviewing Internet
networking issues, such as architecture and TCP/IP and domain name issues. The World Wide
Web Consortium (W3C) suggests guidelines and standardization for the World Wide Web.

With so many forces influencing how the Internet behaves and operates, utilizing this global
network in the same way you use your own private network presents many challenges. To
transform the global network into a reliable application platform, the original architecture of
the Web must be enhanced to meet many needs that we take for granted when developing
traditional applications. The enhanced architecture to develop and deploy your Web-enabled
applications can be seen in Figure 1.4.

03.1389-6 CH01 12 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 Designing and Deploying Web-Enabled Applications: A Changing Architecture for Changing Needs 13

FIG. 1.4 Part

The evolution of Web Proxy Firewall
architecture to support
Server
Internet
I
the basic needs of a Ch
Web-enabled Browser
application.
Web Certificate
1
Server Saver
Database (Production)
Server
HTTP

Server Programs In-house Servers

ASP Exchange
PERL Lotus Notes
ISAPI Index
CGI FTP/Gopher
Java SMTP

As you can see, not only is there a dramatic shift in the number of components now needed to
create, develop, and deploy applications, but there is also a fundamental shift in the method for
developing applications. For the past several years, the shift to client/server architecture has
crept into IT shops around the world. The client/server world relies on the processing power
found on the client machine to off load server requests and processing loads.

However, the benefits of distributing the processing onto the client in traditional client/server
applications are quickly overshadowed by the amount of manpower needed to deploy and
update applications. As you saw in Figure 1.4, the Web-based architecture represents a shift
back to the server-centric deployment model. This model enables a centralized deployment
and distribution mechanism without relying on individual client configurations. Furthermore,
the server-centric model also enables a central connection point to external resources, such as
in-house data stores like Microsoft Exchange, Lotus Notes, or your own custom-built propri-
etary data systems. Let’s take a closer look at the different components needed in the new Web
architecture paradigm.

The Web Browser

The Web browser provides a graphical, text-based terminal interface to the Web server. This
terminal approach provides an interface between the user and the Web server. The Web
browser is responsible for translating HTML sent by the Web server into a graphical user
interface within the browser.

The choice of Web browser is often directly related to the needs of the targeted user. For ex-
ample, if one design goal of your Web application is to reach the largest user group, your appli-
cation should rely only on HTML standards, which can be used to deliver information to all
browsers. However, if your targeted user group is somewhat more defined, you can design
your solutions to take advantage of the special characteristics of the planned user community.
For example, if you need to build an intranet application in which all desktops are running

03.1389-6 CH01 13 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 14 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

Windows 95 and NT operating systems, you can use ActiveX controls that run in Internet Ex-
plorer. If your application assumes the use of Internet Explore 3.0 or Netscape Navigator 3.0
and up, an HTML-only or HTML and Java applet application can be deployed. In most situa-
tions, you will find that defining your targeted user community will dictate the look and feel of
your applications and Web browser.

The Web browser can also use client-side scripting to perform tasks and operations within the
browser. Scripting languages such as JavaScript, JScript, and VBScript are used to reduce
server processing and network bandwidth by performing actions such as field validation or
calculations within the requesting browser. For example, you can create a mini-online help
system by displaying a help message when the user moves the mouse over a hyperlink.

The Web Server

The Web server has several responsibilities that all center around delivering HTML to the
requesting client browser. The traditional billboard or information-sharing approach, which
initially created the need for the World Wide Web, quickly demanded more functionality from
the user community. The Web servers were soon able to process executable scripts that gave
needed functionality and connectivity to other systems, as you saw in Figure 1.2. The capability
to process and establish these connections to other server-side components is made possible
through program execution on the server from various executable sources. Some of these
sources include the Common Gateway Interface (CGI), ISAPI, PERL, REXX, Java, VBScript,
and JScript. All of these processes are aimed at developing application logic that is processed
on the server. The server-side scripts, either acting as stand-alone applications or embedded in
Active Server Pages, can be used to transform the Web server into a gateway that exposes
information stored in other servers. In particular, the Internet Information Server can access
information from database servers, mail and news servers, or any other COM-based server,
such as Lotus Notes or Microsoft Exchange.

The HTTP-TCP/IP Connection

One of the most amazing features of the Internet is that all the Internet technologies are based
on sending and receiving simple ASCII text. The ASCII text is transferred via TCP/IP. TCP/IP
is responsible for wrapping or packaging data into smaller packages and sending the informa-
tion across the network. These smaller packages each have a destination address that is used
to route and re-route the packages. Therefore, packets can take a variety of approaches to
reaching the destination depending on network loads. When the data packets are received at
the destination, the data is re-assembled back into a complete document. TCP/IP also performs
error checking to ensure that no data packets were lost in transit. TCP itself is a connection-based
protocol, which is responsible for establishing a communication session between a client and
server.

The HTTP service runs as an application on the Web server, and is responsible for managing
the Hypertext Transfer Protocol. HTTP is a client/server protocol that accepts requests and
delivers requests via HTML. However, unlike most traditional client/server applications such
as FTP and Gopher, once HTTP delivers the information to the requesting client, the connec-
tion or persistence is released. This feature was built into the original design of HTTP. HTTP

03.1389-6 CH01 14 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 Designing and Deploying Web-Enabled Applications: A Changing Architecture for Changing Needs 15

was designed to accept and handle large numbers of client requests. To maximize the number Part
of potential clients, HTTP typically releases the connection after the information is delivered by
the server to minimize the number of busy or concurrent connections.
I
Ch
Services such as FTP and Gopher, however, keep the connection between the client and server
open until the session is terminated. If a user has an open connection to the server but is not 1
transmitting any data, the connection is still kept open, possibly preventing other users from
connecting to the server. Fortunately, the inherent drawbacks of the connectionless HTTP
protocol can be avoided (depending on the client configuration) by using the HTTP KeepAlive
request.

The HTTP KeepAlive request maintains a TCP connection between the client and server (if
both the server and client’s browser are configured to use this feature). The importance of the
KeepAlive function arises when you need to ensure that information delivered from the server
is completely transmitted to the requesting browser. When the client receives the complete
data stream, a trigger indicating a completed transmission is returned to the server. The HTTP
KeepAlive also enables multiple requests over one connection. Because every item on a Web
page generates individual server requests, the KeepAlive session will increase performance by
using one concurrent connection to the HTTP server instead of having to spend resources
making separate connections.

By itself, the HTTP service is incapable of directly interfacing with the network interface to
exchange data. The Web server relies on Windows Sockets and the Transportation Layer and
Internet Protocol (IP) to provide the communication layer between the HTTP service and the
network interface, as shown in Figure 1.5.

FIG. 1.5
The Internet Service HTTP FTP Gopher
Suite relies on Windows
Sockets to translate
data from the network Windows Sockets
interface to the HTTP,
FTP, and Gopher service
layers.
Transportation and
Internet Layer
(TCP/IP)

Network
Internet
Interface

The Database Server

The database server plays a vital role in Internet application development. The database server
can be used to store, search, and retrieve information that is stored in a database. This same
database that distributes information to Web users can also be accessed and maintained from

03.1389-6 CH01 15 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 16 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

within your corporate walls. For example, a database can be used for an online class registra-
tion system where users can search the database for classes they are interested in and register
for classes via the Web. The administration department can then access this same information
to process the new class registration from either a traditional application built with a tool such
as Visual Basic, C++, Delphi, or PowerBuilder, or it can access the information via a browser
interface.

When using the HTTP server as a connection utility to database servers, there is a slight shift
in architecture that differs from traditional application development. For example, suppose you
are creating an online classified section for your local newspaper and you want to create a
search mechanism that searches for all vehicles of a particular type. From an architecture
perspective, this application will have three components: the requesting browser, the Web
server, and the database server (see Figure 1.6). The browser is responsible for submitting
query requests and displaying the results from the database. The Web server is responsible for
accepting the query from the browser, creating a connection to the database, querying the
database, formatting the results into HTML, and delivering the HTML to the requesting
browser. The database server is responsible for accepting requests from the Web server and
delivering results back to the Web server.

FIG. 1.6 Connection Pooling

When accessing
database information
from a Web browser, the Internet HTTP ODBC OBMS
Web server acts as the
client to the database Web Database
server. Browser Server

Web
Server

Notice that a database connection is not made directly from the browser to the database server.
Instead, the database connection is established only between the Web server and the database
server. As a result, the Web server—not the Web browser—acts as the client to the database
server. This point is particularly important when you create and administer permissions to
your database server. The Internet Information Server automatically creates an NT user ac-
count named IUSR_computername to accept anonymous requests from browsers. In order for
you to create a connection from the Web server to the database server, the IUSR_computername
user account must have permissions on the database server.

Proxy Servers
Proxy servers are designed to mimic the functionality of the browser cache, but instead of being
applied at the browser level, they are applied at the network level. This temporary cache stores
Web page information when a user first downloads it. This information can then be accessed by
additional users from the proxy server, instead of users retrieving information from the site itself.

Although proxy servers increase the apparent speed of the Internet connection, they can cause
problems when developing applications for the Web. For example, because the proxy server is

03.1389-6 CH01 16 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 Designing and Deploying Web-Enabled Applications: A Changing Architecture for Changing Needs 17

designed to act as a replacement for the Web server, how do you ensure that the user is receiv- Part
ing the results from the Web server and not the proxy server cache? Furthermore, not all
proxy servers support the different security authentication methods, such as NT Challenge/
I
Response (NTLM) password authentication. Ch

1
Firewalls
A firewall is a combination of hardware and/or software that is designed to keep private net-
works secure. Although firewalls are intended to keep internal networks free from hackers,
often these same firewalls prevent you from delivering applications to your users. For example,
depending on the security permissions assigned by the firewall administrators, firewalls can
prevent cookies, ActiveX controls, and Java applets from reaching the requesting browser. If
users of your Web-based applications are reporting irregular behavior, remember to check
possible firewall and proxy server limitations.

Certificate Servers
As the popularity of the Internet explodes, people are finding new ways to exploit this global
network of computers. Companies are using the Internet as an extension of their own private
networks by offering access to internal information for their employees or customers through
the public Internet. This capability to access private intranet information via an Internet con-
nection is often referred to as an extranet, which requires explicit security configuration and
management.
However, protecting the private network is only half the security concern. Protective measures
must also be in place to ensure user security and privacy rights. For example, what assurance
does a user have that the online catalog site he is about to place a credit card order with is
really the company the site portrays it to be? Unfortunately, this is one of the largest disadvan-
tages of having an unregulated global network with no central administration unit. Anybody
can place information on the Web, with no preventative measures to ensure that this informa-
tion is accurate or true.

The certificate server was developed to help ensure data integrity for your Web sites or applica-
tions. Data integrity is composed of two parts. The first part, client and server authentication, is
often overlooked. On the public Internet, third-party vendors offer authentication services to
verify the identity and validity of the Web server and Web browser. However, if you are building
an application for your intranet or extranet, you do not have to rely on these external sources
for authentication. The certificate server helps to ease this burden of proof for client and
server authentication by enabling you to issue digital certificates. These digital certificates can
then be administered and managed by the certificate server to regulate entrance to a particular
site or application.

After user and server identification has been properly established, the second part of maintaining
data integrity over the Internet is to ensure data protection while information is being trans-
ferred from the Web server to the client browser. The certificate server uses the digital certifi-
cates with client and server encryption schemes, such as Secure Socket Layers (SSL) and
Private Communication Technology (PCT) protocols, to ensure data integrity.

03.1389-6 CH01 17 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 18 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

External Servers
One of the most exciting aspects of moving to a server-centric application deployment model is
the capability to use the HTTP server as a gateway to other application servers. These other
application servers can consist of pre-built application-based software, such as Microsoft Ex-
change, the Index Server, or the News Server. However, you are not limited to these canned
applications. You can build your own application server to meet your specific application needs.
This flexible method for extending the reach of your Web server into other applications pre-
sents exciting opportunities for Web-based application development. Now these servers, which
were once accessible only via a proprietary client interface, can be accessed by any browser
anywhere in the world.

The Evolution of Active Server Pages

The concept of processing application logic on the Web server and connecting to external
servers from the Web server is not a new idea specific to Active Server Pages. The capability to
create dynamic HTML by sending information to an executable file existed before the emer-
gence of Active Server Pages through the Common Gateway Interface.

Common Gateway Interface

The Common Gateway Interface (CGI) was one of the first methods used to create dynamic
HTML. CGI enables direct communications between the HTTP server and executable scripts.
Programming in CGI provided a standard communication and processing mechanism between
the requesting client browser, the gateway program, and the HTTP server. The CGI programs
help create a standard interface with the HTTP server to eliminate having to learn the specifics
of Hypertext Transfer Protocol.

CGI programs are usually written in a scripting language such as the Practical Extraction and
Report Language (PERL). The early PERL scripts were created to run in UNIX because the
early HTTP servers only existed on the UNIX platform. However, with the emergence of HTTP
servers for NT, the Internet Information Server supports PERL 5.0 scripts. To initiate a CGI
executable, simply reference the name of the executable script and pass any required param-
eters as shown below:
<A Href = “/myCGIscripts/CalcInterest.exe?Principal=300000”> Calculate Interest
using CGI</A>

In this example, the script named CalcInterest.exe is created and passed an argument named
Principal with the value of 300000.

When a PERL script is called on the Web server, the Web server treats the PERL script as a
separate executable. This executable program is not limited to just one script; it can consist of
multiple scripts running on separate machines. However, because the hosting server treats
CGI applications as separate executables, a new process is created for each instance in which
the CGI application is called. Creating new processes on the server is a very expensive re-
source task and can cause significant resource drain and performance issues, especially when

03.1389-6 CH01 18 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 The Evolution of Active Server Pages 19

considering scalability issues as your Web sites and applications grow. Furthermore, CGI Part
applications suffer from the inability to share information across applications. This is because
each new CGI process is created within its own memory space and cannot dynamically share
I
information with other memory spaces of other instantiated CGI programs. Ch

1
Internet Server Application Programming Interface
The Internet Server Application Programming Interface (ISAPI) builds on the lessons learned
from the shortcomings found in CGI applications. ISAPI shares the same functional aspects of
CGI programming but differs from traditional CGI programming in the way the script is ex-
ecuted. ISAPI relies on loading scripts into the HTTP server’s memory space to reduce the
resource drain required to create a new process, as shown in Figure 1.7. The significance of
using a shared memory space is that all the resources made available to the HTTP service are
now made available to the ISAPI applications.

FIG. 1.7 User 1 User 2

Web Server
ISAPI applications are
loaded in the same HTTP FTP
CGI CGI
Gopher
memory space as the Programs Programs
HTTP server.

Windows Sockets

Transportation and Internet Layer

Internet Network Interface

However, as in most situations, often the greatest strength of a product is also its greatest
weakness. Careful and thorough coding and testing must be used when creating an ISAPI
application. Because the ISAPI application shares the same memory space as the HTTP server,
if the ISAPI application crashes, you can also bring down the HTTP server. If the ISAPI applica-
tion terminates without cleaning up the corrupted memory space, runaway processes or
memory leaks could result.

ISAPI is implemented in one of two ways: either as an ISAPI application or as an ISAPI filter.
ISAPI applications are used to encapsulate the functionality found in traditional desktop appli-
cations, but now load that logic into the Web-based arena. To reference an ISAPI application,
use the same methodology used to reference the CGI executable by calling the ISAPI DLL and
passing the appropriate filters, as shown in the following code:
<A Href = “/myCGIscripts/CalcInterest.DLL?Principal=300000”> Calculate Interest
using ISAPI</A>

03.1389-6 CH01 19 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 20 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

ISAPI filters are used to process information being sent to the HTTP server and post-process
the HTTP server responses. The filter acts as an analytical tool to monitor information being
sent to and being generated from the HTTP server and acts as another layer protecting the
HTTP server from the network layer (see Figure 1.8).

FIG. 1.8
ISAPI filters wrap the HTTP ISAPI
FTP Gopher
HTTP server with Service Application
conditioners for pre-
and post-processing
capabilities. Windows Sockets

Transportation and Internet Layer (TCP/IP)

Internet Network Interface

This inline filtering capability can provide extensive functionality to your Web server. For ex-
ample, you can use ISAPI filters to provide public or customizable encryption, authentication,
and compression schemes. Furthermore, you can use ISAPI filters to initiate authentication
methods, create advance-logging features, or scan incoming requests to thwart possible secu-
rity attacks. With the Internet Information Server 4.0, ISAPI applications and filters can be
applied not only globally across the entire server, but also to specific applications residing on
the server.

The HTML Object Model

Up to this point, I have focused largely on the various servers and server components needed
to create dynamic content. However, to fully consider the entire Internet architecture, you also
have to consider how the browser translates and organizes the HTML sent to it from the Web
server.

Just as programming languages adhere to a defined specification identifying the structure of

the language, each Web browser follows its own internal model or blueprint that defines how
the browser operates. The object model usually represents a hierarchical organization chart that
identifies the different relationships between objects and collections for the particular browser
system. For example, Figure 1.9 illustrates the Internet Explorer object model.

In most situations when you design Web-enabled applications, you only have to consider the
specific browser’s object model when you develop client-side scripts. These scripts, interpreted
within the browser, provide advanced functionality not found in native HTML. Client-side
scripting languages, such as JavaScript, VBScript, and JScript, are used to manipulate the con-
tent delivered by the server. The scripts manage the content by manipulating the object model

03.1389-6 CH01 20 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 The Evolution of Active Server Pages 21

of the browser. Typically, this involves controlling the object model as a response to a particu- Part
lar user event. This client-side programming functionality can be as complex as you need it to
be. For example, the HTML object model is used when the user navigates to a different frame
I
page, as shown in the following example: Ch
<A Href = ”/frames/readme.htm” TARGET = “frmain”>Terms and Conditions</A> 1
FIG. 1.9
Browser object models HTTP Service
define how the browser
controls and organizes
the HTML delivered by ISAPI Filters
the Web server.

Windows Sockets

Transportation and Internet Layer (TCP/IP)

Internet Network Interface

When user clicks the Terms and Conditions link, the HTML model is responsible for display-
ing the readme.html in the frame window named frmain. The client-side scripts work the ob-
ject model to manipulate the information. The HTML object model can be used to read and
extract information from or about a page. For example, the HTML object model exposes the
hierarchical navigation to determine how many frames are located on a page, as shown in the
following code:
Window.Document.Frames.Count

Furthermore, client-side scripting can be used manipulate the HTML objects on a page. For
example, the following code illustrates a JavaScript function, named SetValue, which is used to
set the value of a check box named chkReturnEmail:
<SCRIPT LANGUAGE = “JavaScript”>
Function SetValue(){
If (Document.FrmData.ChkReturnEmail.Checked){
Document.FrmData.ChkReturnEmail.Value = “True”
}
Else{
Document.FrmData.ChkReturnEmail.Value = “False”
}
}
</SCRIPT>

This script uses the HTML object model to determine whether a check box named
chkReturnEmail is selected. The HTML object model is used to set the hierarchical relation-
ship between objects on the page. For example, an HTML check box named chkReturnEmail is

03.1389-6 CH01 21 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 22 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

located on a HTML form named frmData. The JavaScript function uses the HTML hierarchy
to check if check box is selected. If the check box is selected, the script function sets the value
of the check box to True. This value would then be sent to the Web server when the page is
submitted.

Keep in mind that the object models for different browsers differ slightly. As a result, your
Web-based applications should be thoroughly tested to ensure compatibility across browsers
and client-side scripting engines.

Dynamic HTML Object Model

As the evolution of HTML continues, the Dynamic HTML object model expands the functional-
ity found in the standard HTML object model. The Dynamic HTML object model now enables
almost complete control over all aspects of Web pages by expanding properties of the docu-
ment object, as shown in Figure 1.10.

FIG. 1.10 Window

The new Dynamic HTML
Vacation
object model enables
greater control of pages. Flames

History

Navigator

Location

Scripts

Document

Forms

Link

Anchors

This new and improved object model embraces using stylesheets, content positioning, and
downloadable fonts to control the look and feel of your sites and applications. Stylesheets en-
able Web page authors to control the visual characteristics of their sites from a single template.
The visual style components for an entire site, such as alignment and font attributes, can be
managed from a single template file. New content-positioning features eliminate the limitations
of positioning items that existed in standard HTML. In standard HTML, you could not layer
items, such as text or images, on top of each other. Dynamic HTML gives you the capability to
layer items on an Web page. In addition, Dynamic HTML also enables the capability to down-
load fonts that are required for the design or display effect for your site. Downloadable fonts
eliminate the dependence of having to rely on fonts that have been previously installed on the
client’s machine.

03.1389-6 CH01 22 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 The Evolution of Active Server Pages 23

Dynamic HTML is intended to give greater flexibility in programming functionality and control- Part
ling and regulating how the site is displayed. In order to use these new features of Dynamic
HTML, a Dynamic HTML-compatible browser is required. Internet Explorer 4.0 and Netscape
I
Communicator 4.0 browsers support the Dynamic HTML standard. However, regardless of Ch
what client interprets it, the HTML is still delivered from functionality to process application
logic on the Web server.
1

Active Server Pages and The Active Server Model:

A COM-Based Evolution
Active, active, active. Everything released from Microsoft has the tag “Active” attached to it:
Active Server Pages, Active servers, ActiveX controls, the ActiveX framework, ActiveX Control
Pad, and Active scripting. The goal for this section is to define the implications that the “Active”
tag carries and what it means for applications that you want to port to the Web.

The Active revolution can trace its origins back to the development of the Common Object
Model (COM). COM was a standard developed by Microsoft to create a standard communica-
tion mechanism between components. This binary standard would enable non-vendor specific
components to interface with each other. When components had a communication standard,
cross-vendor components should share their property methods and events with other compo-
nents. This process of activating another component’s properties, methods, and events is
known as OLE automation.

An example of OLE automation is the capability to embed an Excel spreadsheet in a Word

document. However, the OLE automation offers great flexibility in application development.
You can create your own OLE servers in COM-based tools that expose their methods to calling
programs. These OLE servers can be interfaced from native Win32 clients, such as Visual
Basic, Delphi, and PowerBuilder applications, or from middle-tier components, such as the
Web server. OLE automation servers can be created as stand-alone executables (.exe) or they
can exist as process Dynamic Link Libraries. As stand-alone executables, the OLE servers
perform in a similar manner to CGI scripts, where the application is created as a new process
in its own memory space. You can also compile your OLE server as a DLL to create ISAPI
application and filters.

The reach of COM automation originally enabled OLE controls, or OCX controls, to decrease
development time by adding extra functionality to the native application development environ-
ment, such as Visual Basic or Delphi. These controls represent pre-canned functionality that
would otherwise have to be coded by hand. For example, a grid OCX control could be added to
a Visual Basic form to retrieve, filter, and display information from a database into a grid layout.
This drag-and-drop database functionality would have taken a considerable amount of time if
that functionality had to be coded by hand. These OLE controls have been recompiled and
optimized as ActiveX controls to add that same functionality to Web pages.

Distributed COM (DCOM) is a Microsoft standard that enables COM objects to communicate
across machine boundaries regardless of network configurations. This distributed component
communication mechanism is the foundation that provides the Internet Information Server and

03.1389-6 CH01 23 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 24 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

Active Server Pages the flexibility to communicate with other COM-based systems. For
example, this functionality enables you to expose database information from any ODBC-
compliant data source, access mainframe data via the SNA server, and extract information
from COM-based servers, such as Lotus Notes, Exchange, or your own customized server.

Active Server Pages represent the capability to process application logic on the Web server.
The ASP scripting engine can process any script-compliant language, such as VBScript and
JScript. These script-compliant languages are referred to as ActiveX scripts because they can
be processed on the Web server. The ActiveX scripting engine is actually a COM component
itself.

Furthermore, the ActiveX scripting engine can also process requests made by non-ActiveX
scripting languages, such as PERL, LISP, and REXX scripts. The process that makes this pos-
sible is using COM to provide an OLE wrapper around these scripts. This wrapper acts as an
interpreter between the non-ActiveX scripts and the ActiveX scripting language.

Furthermore, ActiveX documents refer to the same process of converting OLE controls to
ActiveX controls, but instead of being applied at the individual component level, COM technol-
ogy is applied at the application level. ActiveX documents refer to the capability to load entire
applications as a COM object within the Internet Explorer.

As you can see, COM provides ASPs with great functionality to access various information
stores on the server side and deliver that information—as HTML—to the client. However, the
flexibility is not necessarily guaranteed when delivering solutions via client-side ActiveX con-
trols or ActiveX documents. Keep in mind that in order to load the ActiveX controls or ActiveX
documents, the requesting browser must be COM compatible, meaning the ActiveX controls
and documents must be compatible with both the browser and the browser’s operating system.
Using client-side ActiveX controls may not be the best solution to reach the maximum of users
on the Internet, but they do provide great value for intranet/extranet applications. Intranet/
extranet applications are unique because, in most situations, the user group’s development
platform is standardized and well defined. This standardization enables you to tailor your solu-
tions to the user group’s profile instead of trying to deliver a solution to the lowest common
denominator.

The Composition of Active Server Pages

As you can see, most of the Internet Information Server’s expanded connectivity features are a
result of the processing of logic in Active Server Pages. Active Server Pages are text-based files
comprised of a combination of HTML tags and Active Server scripts. The Active Server scripts,
whether written in VBScript, JScript, or your own script-compliant language, are interpreted by
the Active Server engine. The Active Server scripts usually contain variables, operators, and
statements to control the application logic processed by the server. For more detailed informa-
tion about the specifics of an ASP application, please refer to Chapter 10, “The Composition of
an Active Server Application.”

03.1389-6 CH01 24 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 The Evolution of Active Server Pages 25

ASP Delimiters Active Server scripts are distinguished on the page from HTML tags (> and < Part
symbols) and normal content by using the <% and %> delimiters. The delimiters can either be
grouped to contain multiple lines of script code or can be embedded within HTML tags. For
I
example, the following example demonstrates assembling a group of VBScript Active Server Ch
scripts into a subroutine:
1
<%
Sub CalcNewBalanceRef(ByRef Balance)
Balance = FormatCurrency(Balance+Balance*22.5/100/12) ‘=== Calculate
Balance
Response.Write Balance ‘=== Write balance
to page
End Sub
%>

Alternatively, you can integrate HTML tags with Active Server tags, as shown below:
<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<BODY><Title>Date Printed: <% = Now %> </Title> <!--Prints Current Time
and Date
<% If Session(“NewUser”) Then %> <!--Checks if this is a
new user
<P>Welcome to the online reservation system</P> <!--Prints generic
message
<% Else %> <!--User is
registered
<P>Welcome back, <% Session(“UserName”) %>! </P> <!--Creates custom
message
<% End If>
</BODY>
</HTML>

This example uses several inline delimiters to create dynamic content when a user enters the
page. Inline delimiters are used write the current system time to the HTML title tag. The Ac-
tive Server script then checks session variables to determine whether the user is a new user or
if he has been to the site before. If the user is a new user, a generic Welcome message is dis-
played. However, if the user has been to the site before, a customized message displaying the
user name is shown on the page.

Setting the ASP Scripting Language There are two different ways to set the scripting lan-
guage that will process the ActiveX scripting logic. The first method is to rely on the default
scripting language of the Internet Information Server, VBScript. However, relying on the de-
fault scripting language is not always a good practice. As Web sites and applications grow, you
may have to move your application to other servers where the default scripting language is not
the same as the one you delivered your scripts in. To prevent errors because of an incorrect
scripting language trying to interpret your scripts, use the LANGUAGE attribute to explicitly
define the proper scripting interpreter, as shown here:
<%@ LANGUAGE=”VBSCRIPT” %>

03.1389-6 CH01 25 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 26 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

If JScript was the desired scripting language, use JScript with the LANGUAGE tag, as shown
here:
<%@ LANGUAGE = JScript %>

Alternatively, server-side scripting can be implemented outside the Active Server delimiters by
using the <SCRIPT> tag and the RUNAT attribute. For example, the following code creates a
function called dbNull that is used to eliminate NULLs when writing information from a database.
<SCRIPT RUNAT=Server LANGUAGE=”VBScript”>
<!--
Function dbNull(vDBText)
Dim Ret ‘=== Initializes variable
If IsNull(vDBText) Then ‘=== Test Null condition
Ret = ”” ‘=== set variable to zero-length string
Else
Ret = Trim(vDBText) ‘=== Trim and blank spaces
End If
dbNull = Ret ‘=== Return result back to function
End Function
-->
</SCRIPT>

Keep in mind that an Active Server Page does not have to contain any HTML to use the Active
Server Page functionality. For example, the following code sample automatically routes the
requesting browser to another page at /newlocation/default.asp:
<% Response.Redirect “/newlocation/default.asp” ‘=== Routes browsers to new
location %>

Automatically routing users to a new page location (if your site/application has moved or has
changed filenames) is a common technique.

You can build from this concept and create Active Server Pages that do not display any ML
content but are just used to process application logic and workflow. For example, in the follow-
ing script, if a new user enters a site, he is automatically routed to the Registar.asp. If the user
is not a new member, he sees the Member.asp.
<%
If Session(“NewUser”) Then ‘=== Check to see if the user is a new user
Response.Redirect “/Register.asp” ‘=== Automatically directs to register site
Else
Response.Redirect “/Member.asp” ‘=== Directs user to the members site
End If
%>

Variables, Operators, and Statements In order to process application logic and workflow,
each scripting language has its own specific syntax that is used to define and set variables, use
operators for comparing items, and use statements to help define and organize the code.

03.1389-6 CH01 26 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 From Here… 27

Active Server Components and Objects The scripting variables, operators, and statements Part
can be used to tap into special Active Server tools that add programming functionality to the
ActiveX Server. These tools consist of Active Server objects and Active Server components.
I
Active Server objects are built-in objects that collectively represent the functionality of the ActiveX Ch
Server. Six individual objects are used to dissect the different roles and responsibilities of the
server into manageable and programmable aspects. For example, the Request object is respon-
1
sible for retrieving or accepting information from the browser. Another Active Server object,
the Response object, is responsible for sending information to the client. The Session object is
responsible for managing information for a specific user session. The Server object is respon-
sible for the administrative functionality of the server. The Application object is used to manage
all information in the ASP application. The ObjectContext object is used manage transaction
processing for components managed by the Microsoft Transaction Server. The objects will be
covered in greater detail in Part III, “Active Server Objects: The Essential Building Blocks.”

Active Server components run on the server to provide extra functionality that extends the
reach of the Active Server Page beyond the Web server. For example, you can use Active
Server components to create a database connection; create, send, and manage e-mail; deter-
mine the requesting browser type; and much more. Furthermore, you are not limited to only
the components that ship with the Internet Information Server; you can build and distribute
your own components. For more information on the components that ship with the IIS 4.0, see
Chapter 17, “Implementing Existing Server-Side Components.” For more information on build-
ing your own components, please see Chapter 18, “Creating Your Own Active Server Compo-
nents.” Chapter 28, “Working with the Microsoft Transaction Server,” illustrates how to use the
Transaction Server to manage your components.

Server Side Includes Your Active Server Pages can also use Server Side Includes (SSI). SSI
enables you to insert information from a text file into your Active Server Pages before the page
is interpreted. Server Side Includes are very useful for adding information that is common
across multiple pages, such as header, toolbar, or footer information. For example, includes are
used to display footer information such as copyright information or contact information. Fur-
thermore, Server Side Includes are particularly useful as you start developing vast libraries of
scripting functions that can easily be inserted into your Active Server Pages.

From Here…
In this chapter, I have covered the basic principles needed to Web-enable your site or applica-
tion. Having an overview of the ever-expanding Web architecture is critical when considering
building and deploying secure, robust, and scalable Internet-based solutions. This chapter has
discussed the migration from static to dynamic content, as well as shown how the evolution of
COM has transformed OLE and OCX technology from client-side ActiveX controls to extend-
ing this functionality at the server level.

03.1389-6 CH01 27 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 28 Chapter 1 Getting Started with Web Enabling: Creating Dynamic Applications with Active Server Pages

Active Server Pages provide the ability to process server-side logic and can also deliver rich
client-side functionality by delivering HTML, client-side JavaScript, JScript, or VBScript, Java
Applets, ActiveX controls, and other Web components. In addition, ASP scripts do not necessar-
ily have to display content to the browser but can also be used to process and control logic and
workflow on the server.

At this point, you have seen the general architecture needed to develop and deploy Web-based
applications. Chapter 2 will expand on this role to identify the tools needed to make the rollout
of your Web applications possible. After that, Part II of the book will demonstrate using
VBScript to perform the server-side processing of ASP. Parts III and IV will cover the details of
ASP objects and components. Part V will demonstrate how to tap into database functionality
using Active Server Pages. Part VI will cover how ASP integrates with other existing systems,
such as the Microsoft Transaction Server, the Microsoft Messaging Queue, and the Internet
Information Server.

For additional information, please see the following chapters:

■ Chapter 4, “Working with Operators,” demonstrates how to use variables to start

building ActiveX scripts.
■ Chapter 10, “The Composition of an Active Server Application,” will illustrate the
different pieces of an ASP application.
■ Chapters 17 and 18, “Implementing Existing Server-Side Components” and “Creating
Your Own Active Server Components,” will teach you how to use server-side VBScript to
provide connections to pre-built and customizable servers.
■ Chapter 21, “Working with ADO’s Connection Object,” describes how to establish a
database connection.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.
■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,”
demonstrates how to build ActiveX components to create an n-tier distributed application
for the Amazing DataWidget WareHouse Catalog.

03.1389-6 CH01 28 10/26/98, 8:55 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH01 lp#1
 2
29

C H A P T E R

Part
I
Developing and Active Server Pages and
Component Architecture Ch
Discover how Active Server Pages,
2
Deploying Active Server component architecture, and the
Transaction Server all work to-
gether to build scalable, distributed,
Page Applications Web-based applications.

The Internet Information

Server
Learn how the advances in Internet
Information Server help manage
Active Server Page sites and appli-
cations.
Now that the various components that comprise dynamic
Web-enabled applications have been identified in Chapter Active Server Tools
1, this chapter is dedicated to discussing how ASP tech- Discover the variety of tools you can
nologies are used to create, develop, manage, and deploy use to create, develop, and deploy
your sites and applications. Active Server Pages use the Active Server Pages and Active
Internet Information Server as a gateway to communicate Server scripts.
with other systems, such as SNA, mail, and database
servers. This chapter introduces you to the various ASP Tying It Together
tools needed to move from theory into the tangible pro- Learn how Active Server Pages can
grammatic tools to meet your specific business needs: The be used with your current informa-
Microsoft Management Console, which is used to admin- tion system to deliver information to
any HTML-compliant browser.
ister your applications; Visual InterDev, which provides an
integrated development environment for creating Active
Server Pages; FrontPage, which is used as a graphical
user interface tool to add content and create templates
that can be manipulated by your Active Server Pages; and
the Script Wizard and Script Debugger, which can be used
to create and debug client and server scripts. ■

04.1389-6 CH02 29 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 30 Chapter 2 Developing and Deploying Active Server Page Applications

Planning Active Server Web Sites and Applications

Successfully developing and deploying dynamic Web sites and applications depends on the
strength and flexibility of your Web server to deliver content via server-side scripting. These
two areas have been the focus of Microsoft’s Internet vision with its release of the Internet
Information Server 4.0 (IIS 4.0 or just IIS). Internet Information Server 4.0 provides a transac-
tional-based Web server that is tightly integrated with the NT operating system. The advances
in IIS 4.0 provide advantages that can be separated into two camps: improvement in HTTP-
related service areas and the additional functionality in the managing and developing applica-
tion functionality. The advancements in the HTTP services area enable Internet Information
Server to manage multiple Web sites, tailor site- or application-specific settings, and enable
HTTP 1.1 support. Advances made in the application development side include transactional-
based applications, process isolations, SSL support, Active Data Object (ADO), and new devel-
opment tools.

The remainder of this chapter will focus on how to take advantage of all the features of Active
Server Pages and Internet Information Server. Using IIS and ASP, you can build dynamic,
secure, transactional-based applications by following these basic ASP guidelines:

1. Create Active Server Pages to provide a browser-independent transportation vehicle for

embedded workflow, business logic, and transaction processing.
2. Create COM-based components to process core application logic within the Microsoft
Transaction Server.
3. Configure the IIS site using the Microsoft Management Console to provide application-
specific settings, such as Secure Socket Layer, client authentication, or process isolation.

Why Use Components?

In Chapter 1, I briefly mentioned that you can use server-side components to add functionality
to your Web server applications. The value of this concept should not be underestimated. Com-
ponent architecture is a direct result of the ideals driving Object Oriented Programming (OOP).
OOP focuses on developing a series of reusable components that collectively can be used to
minimize application development time and distribution problems while capitalizing on the
benefits of code reuse. By separating an application into smaller, self-contained modules, indi-
vidual components can be created, tested, and implemented more quickly without worrying
how it is going to effect the rest of the application. In addition, applications can be developed
faster because as your library of components grows, entire applications will simply become a
collection of components.

Component architecture also enables applications to be easily updated and quickly re-deployed
because only individual modules have to recompiled and tested. This is unlike traditional appli-
cation development, in which the entire application would have to be recompiled, re-tested and
re-deployed. This new architecture model is shown in Figure 2.1.

04.1389-6 CH02 30 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 Planning Active Server Web Sites and Applications 31

FIG. 2.1 Traditional Component Based Applications

Component-based Application
architecture separates "1 executable contains Application Reporting Component
complex applications all logic" Calls
Authentication Component
into smaller, more
manageable parts. Validation Component

Part
One of the greatest advantages of using component architecture is scalability. In traditional I
application development, the entire application is embedded within one executable file. If an
Ch
application exhausts system resources of the host machine, the entire application would have
to be moved onto a more powerful machine. Component architecture enables you to distribute 2
individual components onto other systems to meet processing needs. This component distribu-
tion process not only enables you to plan for scaling and load balancing concerns, but also
enables you to maximize your existing investment in hardware.

For example, assume that you build a component that performs a search against a database.
This database searching component can be called from a native Win32 client, such as a Visual
Basic or Delphi application, and from a HTTP request initiated by any Web browser, as shown
in Figure 2.2.

FIG. 2.2
COM components 32-Bit
enable any Web client Windows
access to the same Application
information and
application logic as
native Windows
applications.
Application Components
Reporting

Validation

Web Authentication
Server
Web tts and
Browser Active
Server
Pages

In theory, component architecture brings many benefits to application development, but how
can you transform this theory into tangible benefits to your application or organization? COM-
based components can be built using tools from most software vendors today, including
Microsoft, Borland, Powersoft/Sybase, Oracle, IBM, and Micro Focus. COM objects can be
compiled as in-process dynamic link libraries or stand-alone executables. (The details of how to
create a COM component are discussed in Chapter 18, “Creating Your Own Active Server
Components.”)

04.1389-6 CH02 31 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 32 Chapter 2 Developing and Deploying Active Server Page Applications

Once the COM object is created, the object has to be registered. If the object is to be run on
the local system, the REGSVR32.EXE server registration utility should be used to register the
component. If you are installing the component to run on the remote system, use the
DCOMCNFG.EXE utility. The DCOMCNFG utility, as shown in Figure 2.3, enables you to configure
the component’s properties and permissions. Both of these utilities can be found in the Win-
dows System directory.

FIG. 2.3
The DCOMCNFG utility
enables remote
components to be
configured on remote
machines.

Alternatively, you can install and manage your components by using the Microsoft Transaction
Server, as shown in Figure 2.4. The Transaction Server provides complete component manage-
ment ranging from resource and pooling management to transaction process and monitoring
capability.

FIG. 2.4
The Transaction Server
manages components
and gives the capability
to accept or reject
changes made by
components.

Why Use Transactions?

Transactions take the functionality found in individual components and create a logical unit of
work from a specific series of component interactions. The capability to automatically roll back
the changes made by the components is particularly useful in application development. For

04.1389-6 CH02 32 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 Planning Active Server Web Sites and Applications 33

example, consider a simple game between two rival teams. As the two teams play, penalties and
goals are tracked. If one team scores a goal, that goal is only valid if the scoring team did not
also cause a penalty. If the scoring team was responsible for the penalty, the results of the play
are nullified and the score is rolled back to the previous score. If no infractions occur, the score
is counted and play continues. The Transaction Server acts as the referee who initiates the
start of the game. When play starts, the Transaction Server issues a BeginTrans command.
The BeginTrans command marks a checkpoint where violations will cause game play to be Part
returned back to. If a goal is scored and the scoring team did not cause a penalty, the Transac-
tion Server issues a CommitTrans statement to count the goal. However, if the scoring team was I
guilty of creating the violation, the Transaction Server issues a RollbackTrans to void the goal Ch
and restart play at the last valid BeginTrans statement.
2
This same methodology can be applied to using components that perform your own unique
business rules. For example, consider a situation in which a sales ordering system checks and
holds an item in inventory until the credit card payment process has been successfully com-
pleted. You can use your favorite COM development tool to build two components. The first
component reserves an item from the inventory database. The second component processes
the credit card order. You can use transaction processing to abort the reservation of the item if
the credit card is rejected. If the credit card order is successfully processed, the transaction, or
in this case the reservation, can be accepted and the changes permanently made to the inven-
tory.

The capability to manage the logical grouping of work by one or more components is handled
by the Transaction Server. The Transaction Server, as you saw in Figure 2.4, is responsible for
managing component transactions as well as component resource and management issues.
The Transaction Server utilizes Just in Time ( JIT ) activation to optimize system resources
consumed by components. The Transaction Server only activates a reference to a component
when it is needed before creating a new object, but it also enables the object reference to exist
when the object is not in use or is deactivated. In addition, now the Transaction Server elimi-
nates the need to code component security, threading, process handling, and concurrency
issues into your components. This integrated component management reduces development
cost while providing a scalable environment for your components. For more information about
the details of the Microsoft Transaction Server, see Chapter 28 “Working with the Microsoft
Transaction Server.”

One of the exciting new features of IIS is that transaction processing is built directly into Active
Server Pages. Now you can call the transaction SetAbort or SetComplete methods directly
within an Active Server Page to provide rollback or acceptance of changes made by compo-
nents. In previous version of IIS, transaction processes were only available within the compo-
nent itself, meaning the success or failure logic was built into the component, not the calling
Active Server Page. However, now transactions can be controlled from the Active Server Page.
For the details of Active Server Page transaction processing, see Chapter 12, “Controlling the
Server with the Server Object and Using Transaction Processing with the ObjectContext
Object.”

04.1389-6 CH02 33 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 34 Chapter 2 Developing and Deploying Active Server Page Applications

Creating Active Server Pages

After the various components have been built for your site, the next step is to create the Active
Server Pages that you will use to process application logic on the Web server. Active Server
Pages are simply text files, just like HTML files. You can create Active Server Pages in an
ANSCI editor, such as Notepad or VI. However, you can also use Microsoft Visual InterDev to
create and manage your Web-based applications. In conjunction with Visual InterDev, you can
use any HTML graphical user interface, such as FrontPage, to help generate the HTML. In-
stead of writing HTML by hand, you can use the generated HTML as a template that can be
manipulated by your Active Server code. For more information on using Visual InterDev,
please refer to Chapter 27, “Visual InterDev: The Active Server Pages Integrated Development
Environment.”

Visual InterDev
Microsoft’s Visual InterDev is a development environment in which you can create, edit, de-
ploy, and manage Active Server Pages. Visual InterDev combines a rich set of database connec-
tivity tools, wizards, and design-time controls to increase the functionality and decrease the
development time to build Active Server Applications. The Visual InterDev Development Envi-
ronment can be seen in Figure 2.5.

FIG. 2.5
Visual InterDev provides
an integrated develop-
ment environment for
Active Server Page
scripts and applications.

The functionality of Visual InterDev can be categorized into four sections:

■ File and source code management

■ Database connectivity and live design-time access

04.1389-6 CH02 34 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 Creating Active Server Pages 35

■ Active Server Page functionality

■ Client-side functionality

For more information about Visual InterDev, please see Chapter 27.

FrontPage
FrontPage and other GUI HTML tools add value to creating Active Server Pages by adding the Part
visual components that are missing from Visual InterDev. At the time this book went to press,
Visual InterDev does not perform visual drag-and-drop placement and does not provide easy
I
access to the control’s properties, events, and methods directly onto Active Server Pages. Ch
Developers have grown accustomed to this capability that is found in tools such as Visual Basic 2
or HAHTSite from HAHT software. Visual InterDev does enable you to launch HTML GUI
creation tools, such as FrontPage, to generate the visual aspects of a Web page. The page can
then be used as stand-alone HTML page to host static content or can be used to generate a
skeleton or template page to properly place and control dynamic content from ASP scripts, as
shown in Figure 2.6.

FIG. 2.6
Use the visual aspects
of GUI HTML editors to
control page layout.

This technique of using GUI tools to quickly generate HTML can save a lot of time and frustra-
tion spent on getting complicated HTML page layout properly adjusted. After the page is cre-
ated, you can edit the HTML source code to create the dynamic content on the page while
relying on the HTML tags to quickly generate the look and feel of the page.

04.1389-6 CH02 35 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 36 Chapter 2 Developing and Deploying Active Server Page Applications

ActiveX Scripting Languages: Linking It All Together

Using Visual InterDev is a good environment for generating Active Server scripts, but eventu-
ally you will have to get down to the details of script code. Active Server Pages are the combi-
nation of ActiveX scripting and HTML tags. The processing of server-side scripts is what gives
Internet Information Server its application-based functionality. The script processing also en-
ables Internet Information Server to act as an interpreter between remotely distributed brows-
ers and proprietary in-house systems. The Internet Information Server comes with scripting
engines to process VBScript and JScript. The default scripting engine for the Internet Informa-
tion Server is VBScript.

Keep in mind that scripting languages are not fully functional programming languages like C
or C++. Scripting languages are interpreted, not compiled, and they cannot be used to create
stand-alone applications. These scripting languages rely on a host to interpret the code. Al-
though you might consider this a weakness, this dependence on host-based processing enables
the scripts to be portable to a variety of environments. This flexibility is what enables these
scripting languages to be interpreted on both the Web clients and Web servers. Although the
functionality is generally the same between the scripting languages, deciding what scripting
language to use is a function of leveraging your existing language skill set in client- and server-
side processing and delivering that functionality to the largest user base.

VBScript
VBScript was initially created by Microsoft to be a lightweight scripting language to interpret
user events triggered within the Internet Explorer browser. VBScript is actually created from
Visual Basic for Applications (VBA), a pure subset of Visual Basic. VBScript can be used to
create references to control HTML intrinsic objects, ActiveX (formally known as OLE) automa-
tion objects, ActiveX controls, and Java applets.

JavaScript
JavaScript is also a lightweight, interpreted scripting language that provides the same function-
ality as its VBScript counterpart. The syntax of the JavaScript language is similar to C. Java-
Script is a product of Netscape Communications Corporation and Sun Microsystems, Inc.

JScript
JScript is Microsoft’s version of JavaScript, and it was designed to lend OLE-based functionality
found within VBScript to the JavaScript programming structure. This is not to imply one script-
ing language is stronger or weaker than the other; each scripting language has its own
strengths and weakness. The trick is to determine which scripting language best meets your
application and targeted users needs.

N O T E Even though JScript is a modeled after JavaScript, there are incompatibilities between the
two object models. These slight differences could generate runtime scripting errors. To
prevent errors, remember to thoroughly test your scripts before deploying your applications. ■

04.1389-6 CH02 36 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 Scripting Tools 37

The JScript and JavaScript object models are similar, but variations between the models exist.
As a result, the slight varieties can generate runtime errors when executing your JavaScript.

Scripting Tools
Because of their portability, scripting languages are finding their way into more and more
applications. For example, VBScript, once limited to existing only in Internet Explorer browser Part
clients, not only provides the core scripting functionality driving Active Server Pages but is I
now being integrated into Microsoft Office applications, such as Microsoft Outlook. With the
rising importance being placed on scripting languages, script development environments are Ch
beginning to emerge. 2
Microsoft Script Wizard and the Script Debugger
Microsoft currently uses two separate tools to handle creating and debugging scripts. Someday
you can expect these two tools to be merged into one interface, but at the time of this writing,
the Script Wizard and Script Debugger are located as separate tools. The Script Wizard pro-
vides a tree view-based navigation methodology to create, edit, and delete scripts on a page.
The Script Wizard, as shown in Figure 2.7, scans the current page for programmable objects,
and lists the associated properties, methods, and events for each object, which enables you to
associate client actions with a Web-based event. For example, you can determine when the user
clicks a button to display an alert window.

FIG. 2.7
The Script Wizard
enables you to
associate and code
client- and server-side
scripts.

The Script Wizard can be found in both FrontPage and Visual InterDev. However, there are
definite areas where this interface can be improved. For example, the Script Wizard does not
provide a what-you-see-is-what-you-get (WYSIWYG) environment to easily set the properties
and associate events.

04.1389-6 CH02 37 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 38 Chapter 2 Developing and Deploying Active Server Page Applications

The Microsoft Script Debugger provides debugging capabilities for both client- and server-side
VBScript and JScript. The Script Debugger is similar to the debugging environment found in
Visual Basic (see Figure 2.8). This environment enables you to set breakpoints, retrieve and
assign property and object values, and provide different stepping mechanisms to systematically
debug your scripts. For a closer view at the Script Debugger, see Chapter 9, “Using the Script
Debugger.”

FIG. 2.8
The Script Debugger
provides client and
server debugging
features for VBScript
and JScript.

Visual JavaScript
Visual JavaScript is Netscape’s JavaScript development environment, as shown in Figure 2.9.
Visual JavaScript has a GUI interface that provides HTML creation and editing functionality,
drag-and-drop graphical representation of code items on an HTML page, a one button deploy-
ment mechanism, and an inline debugger.

The Visual JavaScript utilities are server-side JavaScript and JavaScript components that supply
application logic processing and database functionality. These JavaScript components rely on
Common Object Request Broker Architecture (CORBA) and Internet Inter-Orb Protocol
(IIOP) combination standard.

For more information about the Visual JavaScript tools, visit Netscape at http://
www.netscape.com.

04.1389-6 CH02 38 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 The Internet Information Server 39

FIG. 2.9
Visual JavaScript from
Netscape is a complete
JavaScript client- and
server-side scripting
development environ-
ment.
Part
I
Ch

The Internet Information Server

The Internet Information Server 4.0 is the foundation you need to deliver and configure your
Active Server applications, and it provides a rich set of HTTP features, including the extensive
site and server management features found in the Microsoft Management Console (MMC). IIS
4.0 also provides Secure Socket Layer 3.0 and client authentication to help ensure that your
content is delivered solely to its intended audience. In addition, the IIS supports HTTP 1.1
standards, which give the latest functionality of the Hypertext Transfer Protocol and server-
side Java.

The Microsoft Management Console

IIS 3.0 uses the Internet Service Manager to administer the Internet-based services such as
HTTP, FTP, and Gopher servers. However, with the release of Internet Information Server 4.0,
the Internet Service Manager has been converted into a generic networking interface called
the Microsoft Management Console (MMC).

The MMC is now a centralized networking administration tool that not only manages the
Internet-based services, such as the HTTP, FTP, and Gopher services, but also can be used to
manage network services and other BackOffice products, such as SMTP and News services,
as shown in Figure 2.10.

Within the Internet Information Server 4.0, you can configure multiple Web sites/applications
from within the Microsoft Management Console. Furthermore, the MMC enables each site to
have its own unique application settings. For example, Figure 2.11 demonstrates implementing

04.1389-6 CH02 39 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 40 Chapter 2 Developing and Deploying Active Server Page Applications

an encryption ISAPI filter that will only be applied to that specific site or application instead of
all the sites on the HTTP server.

FIG. 2.10
Networking interfaces
are now managed
through the Microsoft
Management Console.

FIG. 2.11
IIS 4.0 enables to you
manage multiple Web
sites, each with their
own application
settings.

This feature is particularly useful when you consider the administration costs needed to man-
age large-scale intranets. Now you can use the MMC to create local Web site administrators or
operators, as shown in Figure 2.12.

These operators can now control the specifics of their sites without having to bother the server
administrator. The assigned Web operators can remotely manage the site/application through
an HTML version of the Internet Service Manager, as shown in Figure 2.13.

04.1389-6 CH02 40 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 The Internet Information Server 41

FIG. 2.12
IIS 4.0 reduces
webmaster administra-
tion by allowing local
operators to administer
their own sites based
on NT user security
accounts. Part
I
Ch

FIG. 2.13
The MMC can transfer
administration tasks to
local site operators.

For more information on the MMC, please see Chapter 31, “Using Internet Information Server
4.0 to Manage Your Web Applications.”

Secure Socket Layer 3.0 and Client Authentication

The need to safely deliver information between the Web client and server is paramount when
developing Web-enabled applications. To that end, Internet Information Server 4.0 supports
Secure Socket Layer 3.0, which provides a secure channel between the browser and the server.

04.1389-6 CH02 41 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 42 Chapter 2 Developing and Deploying Active Server Page Applications

In conjunction with SSL 3.0, the Internet Information Server also provides its own certificate
server. Now you can issue your own certificates to provide client authentication for your Active
Server Pages and ISAPI applications.

HTTP 1.1 Support

The Internet Information Server also supports the new HTTP 1.1 specifications. The HTTP 1.1
standard provides many advantages over HTTP 1.0. However, in order to utilize these new
improvements, both the requesting browser client and Web server must be HTTP 1.1 compli-
ant. For more information about detecting the browser type and version of browsers visiting
your site or application, please refer to the Browser Capabilities Component discussion in
Chapter 17, “Implementing Existing Server-Side Components.”

HTTP 1.1 has four new features that help the Web server deliver and manage content. The first
two features, persistent connections and pipelining, change the fundamental communication
mechanisms between the Web client and Web server. In traditional HTTP 1.0, when a browser
requests a page from the Web server, a connection to the Web server is established via TCP.
Then the Web server would accept the request, process the request, and respond to the re-
quest usually by delivering information to the user. After the information was delivered to the
requesting client, the connection between the Web client and server was released. The over-
head associated with creating and releasing connections requires a significant amount of
server resources. The real problem was that a connection is made and released for every item
in an HTML page. For example, if an HTML page contains six images, a minimum of seven
individual connections (including the initial request for the page) are requested, established,
and released from the server. Multiply the amount of requests per page times the amount of
hits that your site receives, and the resource load can be easily visualized. Further notice that
only one request per connection was allowed.

HTTP 1.1 helps resolve the need to establish multiple connections, which are limited to only
one request per connection, by using persistent connections and pipelining. Persistent connec-
tions enable the established communication channel between the browser and the server to be
kept open for a set period of time, instead of closing after the server issues its first response.
Pipelining adds to this persistent connection functionality by enabling multiple requests to be
transmitted over the same connection before receiving any responses from the host.

HTTP 1.1 also supports the HTTP PUT and HTTP DELETE commands to help publish and manage
files via HTTP. The HTTP PUT command enables information to be uploaded to a server. As a
complement to this feature, the HTTP DELETE command provides the capability to remove con-
tent from the Web server.

The Internet Information Server accepts header information from a requesting browser that
identifies the HT TP level of the requesting client. If the browser is HTTP 1.1-compliant, the
server responds by sending an HTTP 1.1 response. If the browser is 1.0-compliant, the Web
server uses HTTP 1.0 to respond to the client.

04.1389-6 CH02 42 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 Extending Active Server Pages into Your Existing Network 43

Server-Side Java
IIS 4.0 provides an updated Java Virtual Machine that was initially shipped in IIS 3.0. The Java
Virtual Machine is responsible for interpreting the Java pseudo code. The communication
between the Java Virtual Machine (VM) and Internet Information Server is made possible via
DCOM. Remember, DCOM was designed to provide a communication specification that en-
ables component communication regardless of networking protocol or machine boundaries.
DCOM is also the communication vehicle that enables any ActiveX components, whether ex- Part
ecuted on the browser’s client or Web server, to communicate. I
Ch

Extending Active Server Pages into Your Existing 2

Network
At this point, I have covered the complete spectrum of architecture considerations and applica-
tion development tools. However, I want to emphasize the benefits of implementing an Active
Server Page architecture. Consider the typical network scheme illustrated in Figure 2.14.

FIG. 2.14 IP
Network
Applying Active Server-
based component
architecture to your HTTP
Server
network schema. Local Win32
clients Reporting

Validation

Internet Firewall
Transaction
Server
Remote
Web Browser Reporting
Validation

Database
Server
Local Web
Browsers

Main Frame
Data

In-House
Servers
Terminal • Exchange
• News
• SMTP

04.1389-6 CH02 43 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 44 Chapter 2 Developing and Deploying Active Server Page Applications

The IP-based network running a variety of mainframe terminals and workstations running on
Windows 3.1, Window 95, and NT provides access to database servers and host-based informa-
tion stored within a mainframe. In this network architecture, there are various ways that clients
can access information. The following situations are used to demonstrate the different types of
data access architecture and how Active Server Pages can be implemented to access the same
information on any Web browser.

Terminal Access
Host-based systems are information systems in which the application logic and information
storage is stored within a remote hosting environment. This type of system removes all pro-
cessing responsibility from the client and uses the client simply as a display mechanism. In
Figure 2.14, the terminal gains access to the host application by establishing a session to a
mainframe of the network. The mainframe provides all the processing of application workflow
and delivers that information to the terminal.

Client/Server
In the traditional client/server world, the processing power of the application is distributed
between the client’s machine and the server machine. The client, typically a Windows-based
application, will request information from a database server. After the data is received, this
information can be processed on the client’s machine. The server plays a split role in this envi-
ronment. Typically, the database server is used to store information and also shares some of
the processing power by using applying logic embedded in triggers and stored procedures.
The client/server configuration starts with replacing the dumb terminal with a machine ca-
pable of localized information storage and processing power. The installed applications on this
machine are responsible for controlling the user workflow. The application can also make calls
directly to server to retrieve information. The server receives, processes, and returns the
information to the calling client. The client application can use the information within the appli-
cation to complete its desired task.

N-Tier Client/Server
The n-tier client/server model expands on the traditional client/server model by introducing
a middle tier. This middle tier, often called the business tier, houses the business rules of the
application. This business rule tier moves some of the processing logic contained in the client
application and the database server into third, or middle, tier. The shift of the business rules
into this middle tier enables the application logic to be separated into smaller components.
These components can then be managed by the Transaction Server to increase reliability,
programmability, and scalability of the application. In addition, centralized updates to the appli-
cation can be made at the middle tier to reduce deployment costs to remote clients. As you can
see in Figure 2.14, the Win32 client application would call a component that is managed by the
Transaction Server. This component would then complete its intended purpose. In this situa-
tion, the component retrieves information from the database and returns the result back to the
calling Win32 application.

04.1389-6 CH02 44 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 From Here… 45

Web Access
The Web browser architecture blends the basic functionality of the terminal emulator and
client/server architecture together to access in-house information stores. The Web browser
is the equivalent of a graphical terminal emulator that uses the HTTP server as a connection
point to the network. Once it is connected to the network, the Web server acts as a client to
other servers on the network. (This situation is also demonstrated in Figure 2.14, where the
Web client is connected to an HTTP server.) Once connected to the HTTP server via Intranet Part
or Internet access, the HTTP server is responsible for processing application workflow issues, I
making connections to servers, and translating the information in HTML. This processing
Ch
power is implemented using Active Server Pages. For example, the HTTP server delivers a
page to the client’s browser requesting information that will be used for a database lookup. If 2
the user enters the appropriate information in the required fields, the Active Server Page can
extract the information and establish a connection to the database server. The database server
can then process and deliver the information back to the Web server. The Web server and
Active Server Pages then format the information and deliver it back to the browser.

N-Tier Web Access

Web-based systems merge the terminal emulator and client/server architecture schemas to
provide a flexible application architecture. This architecture can also enjoy the benefits of
component architecture by enabling Web clients and native Win32 clients to access COM-
based components. The components, managed by the Transaction Server, do not care what
systems are activating them (security permission dependent, of course). Therefore, the Web
server and the Win32 client application receive the same treatment and functionality from the
middle tier component layer.

For example, assume that a component was created to be responsible for running a specific
database report. This database report component can be called by both an Active Server Page
via the HTTP server or from a 32-bit Windows application to retrieve the same report information.

This type of approach brings real benefits to delivering information solutions. In most situa-
tions, you will need to implement this component-based approach to meet the advanced func-
tionality needs of the in-house application administration staff and provide access to the same
information to any Web browser in the world.

From Here…
In this chapter, you have learned how to apply the Web-based architecture to your existing
architecture and what tools are needed to deploy Web-based systems. The convergence of
application architecture and application development tools center around delivering functional-
ity via the Internet Information Server and Active Server Pages. Together, the stage is set to
provide an exciting time in application development, where the shift to component-based archi-
tecture with thin-client browsing capabilities will reap tangible benefits for your application or
organizational needs.

04.1389-6 CH02 45 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 46 Chapter 2 Developing and Deploying Active Server Page Applications

The Internet Information Server shares half the responsibility by:

■ Processing Active Server scripts, CGI, WinCGI, and the Java Virtual Machine on the
server.
■ Managing multiple Web sites and applications within the same server.
■ Supporting Secure Socket Layer 3.0 client authentication, integrated Certificate Server,
and HTTP 1.1 standards.
■ Supporting process isolation and component load and unloading capability to eliminate
having to stop and restart the entire Web site.

Active Server Pages, on the other hand, extend the functionality of the IIS Web server by:

■ Utilizing transactional-based scripts for thread and process management of components.

■ Providing server-side script debugging capabilities.
■ Running each application within its own process to minimize HTTP service failures.
■ Accessing database connectivity through the ActiveX Data object to any ODBC-compli-
ant database.

Internet Information Server and Active Server Pages use the scripting languages that are the
glue that hold everything together. Whether the scripts are processed at the Web server or
within the browser, scripts provide the valuable interactions between Web objects, such as
HTML, ActiveX, or Java-based objects.

At this point, you should be ready to get into the details of Active Server Pages. Looking for-
ward, the rest of the book focuses on implementing Active Sever pages separating the ASP
functionality into five modules:

■ Part II, “VBScript: The Foundation of Active Server Pages,” focuses on learning how
VBScript is used to power the application logic driving your Web-based applications.
■ Part III, “Active Server Objects: The Essential Building Blocks,” illustrates the six built-
in objects that collectively form the foundation to ASP applications.
■ Part IV, “Active Server Components,” guides you through the installable components the
ship with the Internet Information Server.
■ Part V, “Database Management with Active Server Pages,” provides a foundation on how
to access your database via Active Server Pages.
■ Part VI, “Active Server Pages and the Enterprise Solution,” walks you through how to
tap into the functionality of other systems through Active Server Pages.

04.1389-6 CH02 46 10/26/98, 9:02 AM

Brands03/Stone3 SEU Active server pages #1389-6 SRing 11/13/97 CH02 lp#3
 P A R T

II
VBScript: The Foundation of Active Server
Pages
3 VBScript Programming Basics: Variables and Scope 49

4 Working with Operators 69

5 Controlling Program Flow with VBScript 89

6 Working with Functions, Subroutines, and Dialog

Boxes 111

7 Datatype Conversion Features 127

8 Putting It All Together with VBScript 143

9 Using the Script Debugger 161

05.1389-6 Part II 47 10/26/98, 9:04 AM

Brands SEU Active Server Pages #1389-6 Part II SRing 10/28/97 lp#1
05.1389-6 Part II 48 10/26/98, 9:04 AM

Brands SEU Active Server Pages #1389-6 Part II SRing 10/28/97 lp#1
 3
49

C H A P T E R

Introducing VBScript
VBScript Programming Learn how VBScript provides the
foundation of Active Server Pages

Basics: Variables and and why it is essential for Web-

enabling your applications. Part
II
Scope Scripting the Web Client and
Web Server with VBScript Ch
VBScript is the glue that holds Web 3
browser and Web server compo-
nents together. Learn how VBScript
provides the interaction between
ActiveX controls and the HTML
object model on the Web browser
The core functionality of Active Server Pages is centered on and how VBScript provides process-
ing power and communication links
the server-side processing of application logic on the Web
server. Active Server Pages process script-compliant between the Web server and various
server-side objects.
languages to provide a tight integration of accessing
server-side data components with application logic to VBScript Datatypes
deliver standard HTML to any browser. This powerful
Master how VBScript stores vari-
combination is made possible through the execution of
able information and how the vari-
scripting languages processed using Microsoft Internet
ous subtypes and conversion func-
Information Server. VBScript, one of many possible tions are used to harness the flex-
server–side scripting languages, provides the communica- ibility of the Variant datatype.
tion mechanism essential to interfacing with COM-based
servers, such as Microsoft Exchange Server, Microsoft Variable Declaration
Index Server, Database Servers, or Lotus Notes Domino Discover how to declare variables,
Servers. and identify scope issues within
VBScript to run on Active Server
Furthermore, Active Server Pages and VBScript enable
Pages or the Web client.
you to access your own customized ActiveX servers that
are compiled as in-process Dynamic Link Libraries
(DLLs). These ActiveX servers are installed as compo-
nents (formally known as OLE servers) to meet your
specific application needs. These ActiveX components can
be written with various tools such as Visual C++, Visual

06.1389-6 CH03 49 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 50 Chapter 3 VBScript Programming Basics: Variables and Scope

Basic, and Delphi to include such functionality as database access, internal messaging, and
workflow systems. This extension of your existing network and application servers is solidly
integrated within the Web server through the server-side processing of VBScript. VBScript
acts as translator to provide the seamless extension of your existing network and application
infrastructure to any Web client. ■

What Is VBScript?
VBScript is a lightweight scripting language that provides programming functionality based on
the Visual Basic programming language. VBScript is natively executed on the Internet Ex-
plorer browser and can be executed in other browsers through plug-in technologies. VBScript
is also the default scripting language of the Internet Information Server 3.0 and later. The use
of scripting languages, such as VBScript and its JScript counterpart, is particularly interesting
because the script source code is actually embedded as text within the Web page. This capabil-
ity to deliver text-based programmatic functionality is vastly different from traditional applica-
tions in which client functionality requires executing compiled source code.

As Microsoft Office became the office automation industry standard, VBA and VB became an
integral part of every Microsoft Office product. The VBScript language is derived from the
Visual Basic for Applications and Visual Basic programming languages. VBA was originally
created to supply non-programmers, such as managers or sales personnel, with the ability to
add programming functionality into their office documents without learning the specifics of a
programming language. VBA was then developed to provide the capability to add programming
logic into all office products. At the same time, the emergence of browser-based technology
also shared the need for a lightweight programming interface. VBScript was created to meet
these needs.

Because of its VBA/VB-based background, VBScript enjoys a large developer base, allowing
users to be well-positioned to create intelligent Web pages by adding script to be executed on
the Web client or used to create the actual HTML page sent to the browser. The script-embedded
pages are loaded in the client’s browser and used to execute logic to based on a user’s actions.
These scripted Web pages can now take advantage of functionality not found in native HTML
by enabling VBScript to manage and manipulate the HTML object model.

Traditionally, scripting languages perform programming tasks on the Web browser. Client-side
scripting adds value by embedding script processing in the Web page itself. For example, client-
side tasks can include field validation, client-side image mapping, data manipulation, and user
dialog boxes. Client-side scripting also plays a vital role in offloading processing power from
the Web server and reducing the amount of network traffic by eliminating the amount of
round-trip requests to the server. For example, you can use client-side scripting to ensure that
the user has entered only numbers and prompt the user for correct information without requir-
ing any processing from the Web server.

This transferring of responsibility is critical to maximize Web server performance, especially in

server-centric Web-based applications. The server-centric view merges client/server and host-
based architecture to enable a centralized deployment mechanism that takes advantage of

06.1389-6 CH03 50 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 VBScript’s History 51

processing power on the requesting client. When you consider that your Internet applications
can have potentially thousands of concurrent connections, distributing processing can signifi-
cantly reduce the Web server workload.

Scripting can also be used for more than the client-side validation of HTML intrinsic objects,
such as single and multiline text boxes, option buttons, check boxes, and drop-down lists.
Developers can use VBScript to attach, control, and manipulate ActiveX controls and Java
applets within Web pages. Because VBScript is a subset of VBA, many Microsoft Office prod-
ucts use native VBScript code directly in Office forms. This tight integration of VBScript into
the BackOffice suite enables you to minimize code development as applications are ported to
the Web.

Until now, VBScript was a useful, lightweight scripting language that could manipulate, control,
and process objects in an HTML Web page. However, VBScript could only process ActiveX Part
controls, Java applets, and scripting logic on the Microsoft Internet Explorer. With the release
of IIS 3.0 and Active Server Pages, VBScript has now extended its reach into the Web server. II
VBScript now provides unlimited programming flexibility at the Web-server level without rely- Ch
ing on a specific browser type. The server-side processing of VBScript within Active Server
Pages allows for the dynamic HTML creation needed to customize, create, and deliver user-
3
defined Web pages. It is important to note that the Web pages that are delivered by Active
Server Pages are pure HTML and contain nothing proprietary to Microsoft’s Internet Explorer.
This ability is needed to deliver basic application features we take for granted in traditional
application development. The flexibility to respond to user-defined requests is essential for
creating applications. Creating customized HTML pages is made possible through Active
Server scripting and the combination of server-side scripting and Active Server Pages. Fur-
thermore, this server-side programming does not rely on a server proprietary development
tool that you have to train or find from a limited developer base. Now the same Visual Basic
programming skills can be directly applicable to programming on the Web server.

VBScript’s History
VBScript was initially developed to extend client-side HTML functionality based on an event-
driven programming model. VBScript 1.0 was initially released as a part of the Internet Ex-
plorer 2.0 browser, and provided a large amount of programming flexibility found in its native
VBA and Visual Basic programming models. The largest difference between VBScript and
Visual Basic is that VBScript prevents any file access. This is because one of VBScript’s pri-
mary design objectives was to provide a safe scripting language that would prevent malicious
intent from harming underlying browser subsystems. With the release of the Internet Informa-
tion Server 3.0 and Active Server Pages, VBScript can now also be executed on the Web server.

VBScript 2.0 expands programming flexibility found in the initial release of VBScript by allowing
two main programming features. The first expanded flexibility of VBScript 2.0 is the capability
to connect to ActiveX (OLE) automation servers from the Web server level. This connection
feature provides the interface to database servers, application servers, and a multitude of Ac-
tive Server components. All these connections will be discussed in Parts II, III, and IV.

06.1389-6 CH03 51 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 52 Chapter 3 VBScript Programming Basics: Variables and Scope

N O T E Remember: Using the ActiveX Server, a.k.a. IIS 3.0, does not limit your user base to only
ActiveX-compliant browsers. The ActiveX Server uses server-side scripting to provide
transparent data flow from proprietary server-side systems to any browser. ■

Extending Functionality on the Web Client and

Web Server
As the advantages of thin-client architecture continued to gain momentum, the need to provide
programming capabilities at the Web server became increasingly clear if delivering applica-
tions over the Web were to become a reality. With Microsoft’s introduction of the Active Plat-
form, VBScript was transformed from a limited Internet Explorer user base to providing vast
programming extensibility on the server.

Communication Between Objects and Events

Web developers can use VBScript to add application logic to native HTML that would tradition-
ally require server-side processing. VBScript enables the management of application workflow
and object management based on manipulating the HTML object model. An object model repre-
sents a hierarchical blueprint of all objects, methods, events, and operators that exist for a
given language. The HTML object model is what makes the interaction between all objects on
an HTML page possible within the browser.

Manipulating Client-Side ActiveX Controls VBScript can manipulate the HTML object
model to act as the communication mechanism that enables objects, such as intrinsic HTML
controls, ActiveX controls, and Java applets, to interact and communicate with each other. This
interaction is important because it allows the different objects to share their respective proper-
ties, events, and methods. In Figure 3.1, for example, an HTML button initiates a VBScript
subroutine that processes the value of a radio button group and displays the result to the user.
Listing 3.1 shows the source code used to build this example.

FIG. 3.1
VBScript provides
interaction between
HTML, ActiveX, and Java
objects in the browser.

06.1389-6 CH03 52 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Extending Functionality on the Web Client and Web Server 53

Listing 3.1 3_EXT_FUNCT_CLIENTCONTROLS.ASP—Using VBScript to

manipulate HTML controls.
<HTML>
<HEAD>
<TITLE>Client Controls</TITLE>
</HEAD>
<body bgcolor=”#FFFFFF”>
<SCRIPT language=”VBScript”><!--
Sub checkradio()
for i=0 to 2
if document.frmbrowser.r1.item(i).checked then
MsgBox “You have selected “ &
document.frmbrowser.r1.item(i).value
End if
next Part
end sub
--></SCRIPT> II
<form method=”POST” name=”frmbrowser” onsubmit>
Ch
<div align=”center”><center><table border=”1" cellspacing=”1"
width=”40%” bordercolor=”#808080">
<TR>
3
<TD width=”100%”><table border=”0" cellpadding=”2"
width=”100%”>
<TR>
<TD align=”center” colspan=”2" width=”100%”
bgcolor=”#000080"><font color=”#FFFFFF”
face=”Arial Rounded MT Bold”><strong>What is
your favorite browser?</strong></font></TD>
</TR>
<TR>
<TD align=”right”><input type=”radio” checked
name=”r1" value=”Internet Explorer”></TD>
<TD>Internet Explorer </TD>
</TR>
<TR>
<TD align=”right”><input type=”radio”
name=”r1" value=”Netscape”></TD>
<TD>Netscape Navigator </TD>
</TR>
<TR>
<TD align=”right”><input type=”radio”
name=”r1" value=”Mosaic”></TD>
<td>Mosaic </TD>
</TR>
<TR>
<TD align=”center” colspan=”2" width=”100%”><input
type=”button” name=”Your Choice”
value=”Make Choice” language=”VBScript”
onclick=”call checkradio()”></TD>
</TR>
</TABLE>
</TD>
</TR>

continues

06.1389-6 CH03 53 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 54 Chapter 3 VBScript Programming Basics: Variables and Scope

Listing 3.1 Continued

</TABLE>
</center></div>
</form>
</body>
</HTML>

You can further extend client-side scripting by using VBScript to communicate with ActiveX
controls that have been downloaded and registered on the client’s machine. For example,
Figure 3.2 demonstrates an HTML button that uses VBScript to acquire the date selected from
an ActiveX Calendar Control.

FIG. 3.2
VBScript extracts
information from ActiveX
controls on the browser.

Manipulating Server-Side ActiveX Controls VBScript’s communication functionality isn’t just

limited to client-side controls; it also applies to the Web server components as well. The same
interaction that existed between the client-side HTML buttons and VBScript’s searching event
and the ActiveX control in the preceding examples also applies to VBScript and non-visual
Active Server Page components. Figure 3.3 illustrates server-side VBScript scripting accessing
the Active Server Page Browser Component. The Browser Component, as discussed in Chap-
ter 17, “Implementing Existing Server-Side Components,” is used by the Internet Information
Server to determine what type of browser is requesting information from the Web page.

Listing 3.2 shows the VBScript source code that creates an instance of the Active Server Page
Browser Component on the Web server and then displays the browser’s properties on the
HTML page.

06.1389-6 CH03 54 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Extending Functionality on the Web Client and Web Server 55

FIG. 3.3
VBScript is used on the
Web server to control
an Active Server Page
Browser Component to
identify the requesting
browser type.

Listing 3.2 EXT_FUNCT_WEBSERVER.ASP—Controlling ASP components Part

and displaying the results to the browser. II
<%@ language=“VBScript” %> Ch
<HTML>
<head>
<TITLE>Server Component</title>
3
</head>
<body bgcolor=”#FFFFFF”>
<center>
<% Set bc = Server.CreateObject(“MSWC.BrowserType”) %>
<p><font size=”4">
Hello <%= bc.browser %> <%= “ “ &bc.version %> user!
</font></p>
</center>
</BODY>
</HTML>

Differences Between VBA, VB, and VBScript

If you have coding experience in Visual Basic or VBA, you’re probably interested in learning
about the differences between the environments. The most dramatic difference is the develop-
ment environment. As with initial releases of HTML, VBScript, JScript, and JavaScript, all the
code had to be written by hand in an ASCII-based text editor. This was intentionally done to
enable any text editor anywhere in the world to be capable of creating HTML tags. However,
the release of visual design tools, such as the ActiveX Control Pad, FrontPage98, Visual
InterDev, and the Script Wizard, enables you to create and manage HTML and HTML objects
in a GUI-like environment.

I use the term “GUI-like” because these emerging environments will need to mature several
versions before they achieve the functionality found in GUI application development tools
today. In addition, the current tool set is a variety of single-point products that are independent
of one another, making development, testing, and deployment cumbersome. For example, prior
to the ActiveX Control Pad, manipulating ActiveX controls was an extremely cumbersome

06.1389-6 CH03 55 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 56 Chapter 3 VBScript Programming Basics: Variables and Scope

process available only via a text editor. However, the ActiveX Control Pad provides a visual
environment similar to Visual Basic, and enables ActiveX controls to be visually placed and the
objects’ properties manipulated. The ActiveX Control Pad used another Microsoft product, the
Script Wizard, to help develop VBScript and JavaScript code.

Although the two products by themselves have good merit, using the two in conjunction
proved to be awkward. For example, traditional Visual Basic will expose a control’s default
event when you double-click the object. However, if you double-click an object in the ActiveX
Control Pad, its properties are displayed. In order to place code in that event, such as a button’s
OnClick event, the Script Wizard would have to be started first. Then the specific object would
have to be selected from a list displaying all objects on the page. Afterward, the specific event
would have to be selected in order to place code in that event.

The Script Wizard, as seen in Figure 3.4, is a step in the right direction, but there’s still room
for improvement in both the design-time and runtime components. For example, the Script
Wizard can create VBScript but there are no built-in debugging features. Traditionally, to test
your scripts, you have to exit the Script Wizard, save the new script, and refresh your browser
to see if it performs as needed.

FIG. 3.4
The Script Wizard
provides a GUI-like
VBScript and JavaScript
development interface.

To further add to the scripting frustration, the Script Wizard lacks many of the features that
developers have grown accustomed to over the years. The largest of the missing components
is the lack of a runtime debugger directly within the Script Wizard. Microsoft has released a
beta of the Script Debugger for Internet Explorer that provides inline-debugging capabilities.
The Script Debugger, as demonstrated in Chapter 9, “Using the Script Debugger,” enables you
to debug client- and server-side scripts by using breakpoints, step-through, colored syntax, and
a debug window to check variables or initiate new code. But, again, this acts as a separate
product. The Script Debugger can be downloaded from the Microsoft at http://
www.microsoft.com/workshop/prog/scriptie/.

06.1389-6 CH03 56 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Extending Functionality on the Web Client and Web Server 57

Server- and Client-Side Syntax

You probably noticed in Listings 3.1 and 3.2 that the VBScript client-side syntax varies from the
VBScript server-side code. The browser-based VBScript code is neatly contained inside the
<SCRIPT></SCRIPT> HTML tags, whereas the VBScript server code is dispersed throughout
the Active Server Page using the <% %> tags. However, you can also use the <SCRIPT> tag to run
VBScript at the server by specifying the RUNAT.

The Web Browser Client Syntax The <SCRIPT> tag enables various scripting engines, such as
VBScript, JavaScript, or even your own scripting language, to process the embedded HTML
code. You can embed scripting languages into HTML by using the following syntax:
<SCRIPT LANGUAGE=scripting language>

where scripting language is any ActiveX scripting language.

Part
Each scripting language—VBScript, JScript, and JavaScript—has a defined object model. The
specific differences between the three standard scripting languages is beyond the scope of this
II
chapter, but can be investigated further in Que’s Special Edition Using JavaScript, Special Edi- Ch
tion Using JScript, and Special Edition Using VBScript. 3
For example, the following code demonstrates using client-side VBScript to write text to a Web
page. The example uses the OnLoad event of the page to trigger the LoopWrite subroutine.
<SCRIPT LANGUAGE=”VBScript”>
<!--
Sub LoopWrite()
Dim I
I = 1
Do While I < 8
Document.Write “<p> <font size =” & i & “>” & “ Welcome”
I = I + 1
Loop
End sub
-->
</SCRIPT>

<HEAD>
<TITLE>The Do While Loop</title>
</HEAD>
<BODY bgcolor=”#FFFFFF” onload = LoopWrite()>
</BODY>
</HTML>

Placement of <SCRIPT> Tags When an HTML page contains scripting information, the Web
browser parses and compiles the Web page as the browser receives the information from the
Web server. Scripts are instantiated in the order that they appear in the HTML. However, prob-
lems arise in the browser world because Internet Explorer supports both JavaScript and
VBScript. Consequently, each scripting language requires its own separate interpreter to pro-
vide communication between the browser and the embedded script. Because each scripting
language is unique, a scripting error will occur if the default JavaScript interpreter is initiated
and realizes that it’s trying to interpret VBScript code.

06.1389-6 CH03 57 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 58 Chapter 3 VBScript Programming Basics: Variables and Scope

CAUTION
In previous versions of Internet Explorer, if a block of <FORM> code is placed ahead of a block of <SCRIPT>
code, an Object doesn’t support the property or method error may result.

To prevent errors from occurring, make sure any <SCRIPT> tag is placed before (remember
from top down) an object event. To avoid this problem, all <SCRIPT> tags should be defined
within the <HEAD>...</HEAD> tags. Using the <HEAD> location ensures any script code that may
be called by a form element has already been loaded and compiled by the proper scripting
interpreter.

IIS 3.0 Server Syntax VBScript is the default scripting language of the Active Server. As
mentioned previously, server-side VBScript uses the <%...%> and the <SCRIPT></SCRIPT> tag
delimiters within Active Server Pages. With the exception of Input and MsgBox, the Active
Server processes any valid VBScript code within these delimiters. Because the Active Server
Pages are processed at the server, these traditional user interface dialog boxes are ignored.
Furthermore, you’re not limited to using VBScript as your Active Server scripting engine.
Notice the syntax similarities between client and server scripting language selection:
<%@ LANGUAGE = Scripting Language %>

where Scripting Language is any ActiveX scripting language.

You can also use the <SCRIPT> tags to process script at the server. This syntax is similar to the
client-side scripting, but the RUNAT tag specifies the location to process this script.
<SCRIPT Language = “VBScript” Runat =”Server”>
Function ConvertNull(varTemp)
If IsNull(varTemp) Then
ConvertNull=””
Else
ConvertNull = Trim(vartemp)
End If
End Function
</SCRIPT>

This code example is called to convert values extracted from a database into empty strings or
the database value.

Security
VBScript was initially designed to be a safe, interpreted language to be executed on the Inter-
net Explorer browser. Designed to minimize possible security violations to the client’s file and
subsystems, VBScript prevents any file input/output and prevents access to the client’s operating
system. To further prevent destructive intentions, VBScript uses only one datatype, Variant.

06.1389-6 CH03 58 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Extending Functionality on the Web Client and Web Server 59

Internet Explorer Browser Security The Internet Explorer browser has additional security
features built into the administrative functions of the browser. I will not cover all the security
configuration issues that are available in the Internet Explorer browser, but I will highlight the
topics that relate to VBScript. The Security tab of the View | Options menu in Internet Explorer
enables the browser to control whether ActiveX controls, plug-ins, ActiveX scripts, Java, and
Active content can be used in the browser. In the example demonstrating the use of VBScript
to control a client-side ActiveX control, if the Safety Level of the browser were set to High, then
unsafe objects would not be loaded on the browser, as shown in Figure 3.5.

FIG. 3.5
Internet Explorer
enables three levels of
security for Active
Part
content.
II
Ch

After the security settings are set, Internet Explorer manages the browser security concerns.
For example, if your security settings are set on high, Internet Explorer warns you of the po-
tential conflicts and will not display a potentially dangerous control, as shown in Figure 3.6.

FIG. 3.6
The Internet Explorer
prevents unsafe objects
from being loaded onto
the browser.

TIP Using VBScript to reference an unsafe ActiveX control while security is set to High in Internet Explorer
could cause an Object required error. To resolve the problem, change the security settings to
Medium and accept all the warnings.

Server Security Server-side scripting security is based on the NT File System (NTFS) archi-
tecture and the Internet Information Server. NTFS is responsible for access control to your site
by enabling the administrator to set the individual user permissions for the site. These permis-
sion settings not only dictate who has access to files on Web server, but also enable you to
provide different access levels to the files. For example, everybody can have read access to
files in a directory, but only the administrator can write files into that directory. To help moni-
tor these security settings, NTFS enables you to provide auditing capacities to monitor the file-
level security settings. This automatic security record-keeping is essential for tracking possible
security infractions on your system.

06.1389-6 CH03 59 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 60 Chapter 3 VBScript Programming Basics: Variables and Scope

The Internet Information Server provides various client authentication methods to verify the
identity of the requesting user. In addition, protecting the data while in transit is made possible
through Secure Sockets Layer (SSL). SSL is responsible for encrypting the data as it is trans-
ferred between the client and Web server. The Internet Information Server also provides the
capability to limit the execute permissions on each Web site.

VBScript Variable Datatypes

The use of programming variables is the basis of any programming language. A variable acts as
a temporary storage location for your application to access and manipulate data. Variables are
identified by a name and contain information stored in memory as a value.

Variant
While Visual Basic supports many datatypes, VBScript only supports one datatype, Variant.
This datatype is unique because it represents all the datatypes found native in Visual Basic,
such as Integer, String, Double, and Currency. The Variant datatype’s power is that the char-
acteristics of the variable change depending on what type of information it stores. However,
this automatic conversion capability is countered by the performance loss needed to identify
the variable type.

Subtypes
The Variant datatype consists of smaller units called subtypes. Subtypes identify how the data is
stored within the Variant. They’re centered on the String, Numeric, Byte, Boolean, Date,
Empty, Null, Error, and Object datatypes (see Table 3.1). The specialized subtypes exist to
provide optimized functions for extended functionality of the specific datatype, as well as small,
lightweight functions optimized for speed.

Table 3.1 The VBScript 2.0 Subtypes

Category Description
Empty Value is 0 for numeric variables or a zero-length string (“”) for
string variables.
Null Variant intentionally contains no valid data.
Boolean True or False.

Byte Integer in the range 0 to 255.

Integer Integer in the range -32,768 to 32,767.

06.1389-6 CH03 60 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 VBScript Variable Datatypes 61

Category Description
Currency -922,337,203,685,477.5808 to 922,337,203,685,477.5807.
Long Integer in the range -2,147,483,648 to 2,147,483,647.
Single Single-precision, floating-point number in the range -3.402823E38
to -1.401298E-45 for negative values; 1.401298E-45 to 3.402823E38
for positive values.
Double Double-precision, floating-point number in the range -
1.79769313486232E308 to -4.94065645841247E-324 for negative
values; 4.94065645841247E-324 to 1.79769313486232E308 for
positive values.
Date (Time) Number that represents a date between January 1, 100 to
December 31, 9999. Part

String Variable-length string that can be up to approximately 2 billion II

characters in length. Ch
Object Object. 3
Error Error number.

As mentioned previously, the subtypes of Variant provide for a wide range of conversion func-
tions used to manipulate the subtypes, such as date/time values, string expressions, and cur-
rency values, as shown in Table 3.2.

Table 3.2 Conversion Functions in VBScript 2.0

Category Converts to
CByte Byte
CCur Currency
CDate Date
CDbl Double
CInt Integer
CLng Long
CSng Single
CStr String

For more information about the details of the conversion functions, please see Chapter 7,
“Datatype Conversion Features.”

06.1389-6 CH03 61 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 62 Chapter 3 VBScript Programming Basics: Variables and Scope

Identifying the Datatype

With the various subtypes available, at some point it might be necessary to determine what
variable subtype you’re dealing with. The VarType function returns a numeric value that identi-
fies the subtype. Table 3.3 displays the results of the VarType function.

Table 3.3 Identifying the Datatype

Subtype Return Value
Empty 0

Null 1

Integer 2

Long 3

Single 4

Double 5

Currency 6

Date (Time) 7

String 8

Automation Object 9

Error 10

Boolean 11

Variant 12

Non-Automation Object 13

Byte 17

Array 8192

Using Variables
Now that the different characteristics of the Variant datatype and the conversion functions
have been briefly covered, the length of time that the variable is available is critical in applica-
tion development. This section will discuss scope or lifetime of variables in memory, how to
create a variable, and the limitations and restrictions of using VBScript variables.

Variable Scope
Variables in VBScript can exist at two levels or scopes: either at the script or procedure levels.
The term scope refers to the amount of time that the variables can be referenced in memory.

06.1389-6 CH03 62 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Using Variables 63

The amount of time, often referred to as the variable lifetime, depends on the level at which the
variables are declared.

■ Script-level code is contained outside all Function or Sub modules.

■ Procedure-level code is contained within a Function or Sub module.
Procedure-level code is a subset of script-level code. Therefore, any variable declaration made
inside one function or subroutine can’t be referenced in a separate script or subroutine. To
share variables across procedures, you have to use a script-level variable. (Script-level variables
are the equivalent of global variables in Visual Basic.) For example, the code in Listing 3.3
demonstrates scoping issues with script and procedure-level variables.

Listing 3.3 SCOPE_AND_LIFETIME.ASP—Using scope to determine a

Part
variable’s lifetime.
<SCRIPT language=”VBScript”>
II
<!-- Ch
Dim vTotalDays <!-- Script Level Variable: Longer Scope -->
VtotalDays = Request.QueryString(“Days”) 3
Sub GetVehicleInfo(vDays)
Dim vSubTotal <!-- Procedure Level Variable: Limited Scope -->
if Len(Trim(frmGetData.txtVehicleType.value)) = 0 then
vMsgStr = “Please Select A Vehicle”
vVehicleInfo = “”
vBillRate = 0
Else
vMsgStr = “”
vVehicleInfo = frmGetUserInfo.value
vBillRate = RS(“Rate”)
End if
VSubTotal = (vBillRate*vTotalDays)
MsgBox “Your Total is “ & vSubTotal
End Sub
Sub SaveInfo()
RS(“SubTotal”)= vSubTotal <!-- Out of Scope: Stores a Null -->
RS(“TotalDays”)= vTotalDays <!-- In Scope: Stores the Value -->
End Sub
-->
</SCRIPT>

This example illustrates how variable scoping issues can cause logical errors in your applica-
tion. The two VBScript functions are used to help calculate and save a subtotal for a car reser-
vation system. The subroutine GetVehicleInfo is used to calculate the subtotal of the rented
vehicle and display a message box to the user. The SaveInfo function is used to write the sub-
total and the total amount of days the vehicle was used into a database. After running the script,
the developer noticed that the vehicle subtotal was not being saved to the database. The root of
this problem arises from the improper use of variable scoping. The variable vTotalDays is
properly stored because it is declared as a script level variable and can be accessed by any
subroutine in the script. However, the variable vSubtotal was not declared as a procedure-level

06.1389-6 CH03 63 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 64 Chapter 3 VBScript Programming Basics: Variables and Scope

variable and cannot be shared among procedures. Even though the vSubtotal variable is cre-
ated and calculated in the GetVehicleInfo subroutine, it cannot be seen in the SaveInfo sub-
routine. As a result, when the subtotal is stored into the database, a NULL value is stored.

Declaring Variables
As previously mentioned, variables contain two characteristics: a name and a value. Variables
can be created either explicitly by declaring a variable name, or implicitly by simply referring to
a variable name. Explicit declaration refers to the process of creating or instantiating a variable
by directly calling the variable name with either the Dim, Public, or Private keywords. The
Dim, Public, and Private keywords are used to define the length of time the variable can be
referenced in the application. Implicit declaration refers to the process of creating a variable on-
the-fly by referring to the variable name. In both cases, when a variable is declared or initial-
ized, the default value of the variable depends on the assumed datatype. A numeric variable is
initialized to 0, whereas a string variable is initialized to a zero-length string (“”).

Dim The Dim statement is used to explicitly declare a variable. You will notice that in
VBScript, as in Visual Basic, you do not have to declare a variable as a specific datatype. This
change in syntax is a result of the existence of only one datatype—Variant. Dim’s syntax is as
follows:
Dim varname[([subscripts])]

where varname is the name of the variable and subscripts represents the number of declared
dimensions.

The Dim statement can also be used to store information in arrays. An array is treated as one
variable, but it is used to contain a collection of similar information. This collection is accessed
by iterating through the array starting at the lower bound of zero. Arrays can also be multi-
dimensional, allowing up to 60 dimensions. For example, the following code creates a static
array with 21 sequential indexes or storage locations that represents the items ordered from an
online shopping catalog. Because arrays are zero-based, 21 sublocations of the variable
ItemsOrdered are actually stored. The second and third lines of code populate the first and
second indexes in the array with a value from a text box on the Order Form page.
Dim ItemsOrdered(20)
ItemsOrdered (0) = frmOrderForm.TxtOrderItemNum.Value
ItemsOrdered (1) = frmOrderForm. TxtOrderItemNum.Value

ReDim The ReDim function is used when manipulating dynamic arrays. Arrays can exist in
two forms: static or dynamic. A static array sets the number of indexes when the array is de-
clared. The previous example demonstrated the use of a static array. Dynamic arrays allow the
number of indexes to vary at runtime. Dynamic arrays are created by using empty parentheses
() when initially declaring the variable.

Working with dynamic arrays is a little more difficult than working with static arrays. Dynamic
arrays are used when the allocation of storage space is necessary, allowing you to add memory

06.1389-6 CH03 64 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 Using Variables 65

space to your array. To accomplish this, use the ReDim command to properly allocate memory
space. The ReDim command is used to redefine the array size to be larger or smaller as needed.
If you want to increase the size of the array and keep its contents, use the Preserve keyword.
For example, the following code creates a dynamic array call RentedVehicles and resizes the
array to six elements. The last line of code expands the number of elements in the array to 11,
while also preserving any data in the first six elements.
Dim RentedVehicles()
ReDim RentedVehicles(5)
RentedVehicles(0) = frmOrderForm.TxtSelectedVehicle.Value
ReDim Preserve RentedVehicles(10)

TIP You first must ReDim a dynamic array before you can assign values to it.

Part
Public The Public declaration keyword is new to VBScript 2.0. Public is implemented at II
script level to declare public variables that have scope throughout the existence of the process-
Ch
ing of the page. The Public declaration can also include static and dynamic array declaration.
The Public syntax is as follows: 3
Public varname[([subscripts])]

where varname is the name of the variable and subscripts represents the number of declared
dimensions.

Private The Private statement, also new to VBScript 2.0, is the opposite of the Public state-
ment. Private variables are available only to the script level in which they are declared and can
be used with static and dynamic array processing. Private’s syntax is as follows:
Private varname[([subscripts])]

where varname is the name of the variable and subscripts represents the number of declared
dimensions.

Table 3.4 illustrates identifying variable scope by the declaration type.

Table 3.4 Determining Variable Scoping with Declarations in VBScript

Declaration Script Scope
Dim Script level, procedure level
ReDim Procedure level, procedure level
Public Script level
Private Script level

06.1389-6 CH03 65 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 66 Chapter 3 VBScript Programming Basics: Variables and Scope

TIP When a variable is first declared, the default value of the variable depends on the assumed datatype. A
numeric variable is initialized to 0, whereas a string variable is initialized to a zero-length string (“”).

Option Explicit The Option Explicit statement is used to prevent variables from being
implicitly created within VBScript. To use the Option Explicit statement, use the following
syntax:
Option Explicit

The Option Explicit statement must be declared before any script is processed by the script-
ing engine.

The code in Listing 3.3 demonstrated the difference between script- and procedure-level pro-
cessing. The variable VtotalDays was explicitly declared using the Dim keyword, whereas the
variables vMsgStr, vVehicleInfo, and vBillRate were implicitly declared by simply referenc-
ing the variable name.

Declaration Limitations
Even though VBScript is a text-based system, the VBScript interpreter follows its own defined
programming model. The programming model defines a blueprint of conventions that the
VBScript language needs to follow to operate. The rules are usually manifest in syntax, opera-
tors, and naming conventions. Variable names in VBScript also follow a standard convention.
The following is a list of VBScript naming conventions.

■ Variables cannot contain an embedded period.

■ Variables must be fewer than 255 characters.
■ Variable must be unique in the scope.
■ Variables must begin with an alphabetic character.
■ VBScript must have fewer than 127 procedure-level variables instantiated.
■ VBScript must have fewer than 127 script-level variables instantiated.

TIP A multidimensional, multi-indexed array is viewed as only one variable in VBScript.

From Here…
Variable declaration, datatypes, and scoping issues form the foundation of the VBScript pro-
gramming language. To use this knowledge, you next have to look at what types of operations
you can perform on variables. The following chapters provide this information:

■ Chapter 4, “Working with Operators,” demonstrates how to use variables to start

building to ActiveX scripts.
■ Chapter 7, “Datatype Conversion Features,” illustrates the subtype conversion features
available to manipulate your datatypes.

06.1389-6 CH03 66 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 From Here… 67

■ Chapter 20, “Database Access with ADO,” guides you through an overview of accessing
and managing your database connections.
■ Chapter 21, “Working with ADO’s Connection Object,” describes how to establish a
database connection.
■ Chapter 23, “Working with ADO’s Recordset Object,” teaches you how to use database
cursors and manipulate the resultant database query information.

Part
II
Ch

06.1389-6 CH03 67 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 68 Chapter 3 VBScript Programming Basics: Variables and Scope

06.1389-6 CH03 68 10/26/98, 9:06 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH03 lp#4
 4
69

C H A P T E R

Arithmetic Operators
Working with Operators Learn how to use the arithmetic
operators to manipulate your nu-
meric variables and mathematical
functions.

Comparison Operators
Examine the comparison operators
VBScript provides a variety of programming operators to needed to compare and contrast
variables, datatypes, and objects.
help manipulate variables in your Active Server Pages.
Most programming languages break down their object Logical Operators Part
models into logical groups based on functionality. The
ability to manipulate objects and variables form the foun-
Discover how logical operators
extend the comparison operators’
II
dation of workflow control in an application. This chapter functionality by applying compari- Ch
discusses the use of arithmetic, comparison, logical, and son operators to Boolean expres-
concatenation operators in VBScript 2.0 and explains how sions.
4
VBScript prioritizes these operators. ■
Concatenation Operators
Learn how to properly use the
numeric and string concatenation
operators without having to experi-
ence the programming troubles
from automatic datatype conversion.

Order of Operations
Guard against programming logic
errors in your ActiveX scripting by
learning how VBScript systemati-
cally processes operators.

07.1389-6 CH04 69 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 70 Chapter 4 Working with Operators

Arithmetic Operators
The VBScript arithmetic operators perform various mathematical calculations. Most of the
arithmetic operators have the following syntax:
myResult = myExp1 Operator myExp2

where myExp1 and myExp2 are numeric expressions and Operator is a mathematical operator.

The arithmetic operators supported by VBScript are listed in Table 4.1.

Table 4.1 Arithmetic Operators

Operator Symbol Description
Addition + Sums two numbers.
Subtraction - Finds the difference between two numbers or sets
the negative value of a number.
Multiplication * Multiplies two numbers.
Division / Divides two numbers.
Integer Division \ Divides two numbers and returns an integer result.
Exponentiation ^ Raises a number to the power of an exponent.

Addition
The addition operator (+) is used to sum numeric expressions. It can also be used to append or
concatenate strings together, so great care must be used with the addition operator. The syntax
for the addition operator is:
myResult = myExp1 + myExp2

where myExp1 and myExp2 can be variables of any subtype. Remember, there is only one
datatype in VBScript: Variant. Adding two variables may produce unexpected results because
of the Variant’s capability to change subtypes automatically based on the variable’s datatype.
To prevent unexpected results, consider the following operation rules for the addition operator,
as shown in Table 4.2.

Table 4.2 Rules of Addition Operator

If expressions are... Then operator will...
Both Numeric Add
Both Strings Concatenate
String and Numeric Add

07.1389-6 CH04 70 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Arithmetic Operators 71

Consider a situation where the addition of two numbers is required. The following code with
variables myVar1 and myVar2 (both of Variant datatype) illustrates the unexpected results that
may occur when using the addition operator.
<%
myVar1 = 30: myVar2 = “5”
myResults1 = myVar1 + myVar2 ‘=== Returns 35
myResults2 = myVar2 + myVar2 ‘=== Returns “55”
myResults3 = myVar2 + myVar1 ‘=== Returns 35
myResults4 = myVar1 + myVar1 ‘=== Returns 60
%>

You can see from this example that, without a proper understanding of the Variant datatype
and the behavior of the addition operator, a VBScript programmer could become frustrated
quickly with unexpected results! In the above example, the variables are first created and ini-
tialized. Both of these variables are of the Variant datatype. However, VBScript has already
assigned an automatic subtype to each. The variable myVar1 is of subtype Integer because it
was assigned an integer value. The variable myVar2 is of subtype String because it was as-
signed a string value. Then, in the third line of code, myVar1 and myVar2 are added together.
Following the rules in Table 4.2, adding a string value and a numeric value together results in
the two values being summed. In this situation VBScript automatically converts myVar2 to a
numeric value and adds it to myVar1, and then assigns the sum to myResults1. The next line of Part
code concatenates the two string variables to produce the literal string “55” for the result. The II
fifth line of code produces the same results as the third line. VBScript converts myVar2 to a
Ch
numeric value and adds it to myVar1. The last line simply adds two numeric values together.

To help prevent some of these datatype issues, VBScript offers conversion functions to explic-
4
itly convert a variable to a desired subtype. The CStr function converts a variable to a String
subtype and the CInt function converts a variable to an Integer subtype. The conversion func-
tions will be discussed in greater depth in Chapter 7, “Datatype Conversion Features,” and are
demonstrated in the following script:
<%
dim myVar1, myVar2, myResults
myVar1=5: myVar2=10
myResults = CStr(myVar1) + CStr(myVar2) ‘=== Returns “510”
myResults = CStr(myVar1) + CInt(myVar2) ‘=== Returns 15
myResults = CInt(myVar1) + CStr(myVar2) ‘=== Returns 15
%>

Remember, in order to protect yourself from potential programming errors, always be sure to
use the conversion functions to ensure your addition operator performs as expected with the
proper datatypes.

TIP To prevent unexpected string concatenation when using the addition operator, use conversion functions
to ensure proper data subtypes.

07.1389-6 CH04 71 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 72 Chapter 4 Working with Operators

Subtraction
The subtraction (-) operator is used to determine the difference between two numbers or is
used to negate a positive number. The dual functionality of the subtraction operator requires
two different syntaxes. To determine the difference between numbers, use the following syntax:
myResult = number1 - number2

where number1 and number2 are numeric subtypes.

To provide the unary negation, or the negative value of an expression, use the following syntax:
myResult = -(expression)

where expression is a numeric subtype.

The subtraction operator is often referred to as the negation operator because is negates a
positive number. If the result of the negation operator is empty, the result is treated by other
numeric expressions as if it were zero (0).

Multiplication
The multiplication operator (*) is used to return the results of two numbers multiplied to-
gether. The multiplication operator uses the following syntax:
myResult = number1 * number2

where number1 and number2 are numeric subtypes. If either number1 or number2 is NULL, the
data subtype of the result variable will be NULL. If number1 or number2 is Empty, the value of the
empty variable will be converted to zero.

Division
The division operator (/) is used to divide one numeric expression into another numeric ex-
pression. The syntax for the division operator is:
myResult = number1 / number2

where number1 and number2 are numeric expressions and myResult returns a floating-point
datatype.

The rules for the multiplication operator also apply to the division operator. If either number1 or
number2 is NULL, the resulting data subtype is NULL. If either numeric expression is Empty, that
numeric expression is treated as a zero.

Exponentiation
The exponential operator (^) is used to raise a number to the power of an exponent. The expo-
nential operation is performed using the following syntax:
myResult = number ^ exponent

where number and exponent are any numeric value.

07.1389-6 CH04 72 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Arithmetic Operators 73

Keep in mind that number can be negative only if exponent is an integer value. Also, if the
number or exponent variables are NULL, then the resulting data subtype will also be NULL.

Integer Division Operator

The integer division operator (\) is used similarly to the division operator (/) but is specifically
designed to return an integer result. The integer division operator uses the following syntax:
myResult = number1 \ number2

where number1 and number2 are numeric expressions that are rounded to either a Byte, Integer,
or Long subtype before the division. Furthermore, if either numeric expression is NULL, the
result is NULL; and if either expression is Empty, the numeric expression is treated as a zero.

TIP Use the TypeName function to determine the subtype of a variable.

For example, consider the following script:

<%
myExp1 = NULL ‘=== Assigns a NULL
myExp2 = 2000.86 ‘=== Assigns a Value Part
myResult = myExp1 \ myExp2 ‘=== Performs Integer Division but is
➥assigned a NULL value II
Response.Write myResult &”<BR>” ‘=== Write nothing to page
%> Ch

This script would not return any values because the NULL value assigned to the variable myExp1 4
produces a NULL result. This assignment of NULL often occurs when extracting information from
a database. To prevent the cascading of the NULL datatype, use the VarType character as shown
below:
<%@ LANGUAGE=”VBSCRIPT” %>

<SCRIPT LANGUAGE = “VBScript” RUNAT = SERVER>

Function TrimNulls(vValue)
Select Case TypeName(vValue)
Case “Null”
ret = 0
Case Else
ret = vValue
End Select
TrimNulls = ret
End Function
</SCRIPT>

<HTML>
<HEAD>
<BODY>
myExp1 = NULL ‘=== Assigns NULL
myExp2 = 2000.86 ‘=== Assign value
myResult = TrimNulls(myExp1) \ TrimNulls(myExp2) ‘=== Trims out the NULL
➥values

07.1389-6 CH04 73 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 74 Chapter 4 Working with Operators

Response.Write “myExp1: “ &TypeName(myExp1) &”<BR>” ‘=== Displays NULL

Response.Write “myExp2: “ &TypeName(myExp2) &”<BR>” ‘=== Displays Double
Response.Write “myResult: “ &TypeName(myResult) &”<BR>” ‘=== Displays Long
Response.Write myResult &”<BR><HR>” ‘=== Returns 0
</BODY>
</HTML>

In the code example, the function TrimNulls was used to place NULL values with 0 values. As a
result, the ASP script returns a 0 for the variable myResults.

Modulus Operator
The modulus operator (MOD) is used to return only the remainder when two numbers are di-
vided. The modulus operator uses the following syntax:
myResult = number1 MOD number2

where number1 and number2 can be any numeric expression. The MOD operator divides number1
by number2 to return only the remainder portion of the result. For example, the expression E =
21 Mod 8.37 returns the value 5 of a type Long. Using the regular division operator on the
expression F = 21 / 8.37 returns 2.5089605734767 of type Double.

In addition to supporting basic mathematical functionality, VBScript also supports several

trigonometric and logarithmic functions. Table 4.3 lists these mathematical functions.

Table 4.3 Mathematical Functions Supported in VBScript

Function Description
Sin Determines the sine of an angle.
Cos Determines the cosine of an angle.
Tan Determines the tangent of an angle.
ATan Determines the arctangent of a number.
Exp Determines the base of natural logarithms raised to a power.
Sqr Determines the square root of a number.
Log Determines the natural logarithm of a number.

The extended mathematical functions use the following syntax:

myResult = Function (expression)

For example, to find the cosine of a 30-degree angle, the following code would be used:
<% MyAngle = Cos(30) %>

07.1389-6 CH04 74 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Comparison Operators 75

Comparison Operators
Comparison operators analyze the relationship between expressions. Remember that expres-
sions can be any combination of keywords, operators, variables, or constants that form a string,
number, or object. VBScript uses different comparison operator syntax when comparing ex-
pressions and when comparing objects. If you are comparing expressions, the following syntax
is used:
myResult = myExp1 (operator) myExp2

where myExp1 and myExp2 contain any legal expression, and the comparison operator repre-
sents a symbol that defines a relationship between the two expressions. The comparison opera-
tors used in VBScript are illustrated in Table 4.4.

Table 4.4 Comparison Operators

Operator Symbol Description
Equality = Is equal to
Inequality <> Is not equal to
Part
Greater Than > Is greater than

Lesser Than < Is less than

II
Ch
Greater Than or Equal To >= Is greater than or equal to
Less Than or Equal To <= Is less than or equal to
4
Object Equivalence IS Object comparison

Comparison operators give you the ability to determine the relationship between two expres-
sions. Traditionally, developers expect these comparisons to evaluate to either True or False.
But it is important to consider a third possibility: NULL. If either expression in the comparison is
a NULL datatype, the comparison will return a NULL value. This is important to keep in mind
when designing application logic. Programmers should design an application flow that responds
to the possibilities of a comparison evaluating to True, False, or NULL. Table 4.5 represents the
control logic for the traditional comparison operators, excluding the IS comparison operator.

Table 4.5 The Comparison Operator Results

Operator Returns True if Returns False if Returns NULL if
< myExp1 < myExp2 myExp1 >= myExp2 myExp1 or myExp2 = NULL

<= myExp1 <= myExp2 myExp1 > myExp2 myExp1 or myExp2 = NULL

> myExp1 > myExp2 myExp1 <= myExp2 myExp1 or myExp2 = NULL

continues

07.1389-6 CH04 75 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 76 Chapter 4 Working with Operators

Table 4.5 Continued

Operator Returns True if Returns False if Returns NULL if
>= myExp1 >= myExp2 myExp1 < myExp2 myExp1 or myExp2 = NULL

= myExp1 = myExp2 myExp1 <> myExp2 myExp1 or myExp2 = NULL

<> myExp1 <> myExp2 myExp1 = myExp2 myExp1 or myExp2 = NULL

Comparison operators act like arithmetic operators in that they also change behavior based on
the datatypes of the operands. You can help prevent unexpected results by identifying the
datatype by using the TypeName function or by using string conversion utilities, such as CStr
and CLng. (These formatting and conversion functions are discussed in further detail in Chap-
ter 7.) For example, if the variable vNumber = 11\3 (the integer division function), then
TypeName(vNumber) returns Integer.

Use Table 4.6 to serve as a guideline of expected logic performed based on operand subtype.

Table 4.6 Evaluating Expressions Using VBScript Comparison Logic

Expression Performed logic
If both are numeric Numeric comparison.
If both are strings Perform a string comparison.
If one expression is The numeric expression is less than the string
numeric and the other expression.
is a string
One expression is Empty Perform a numeric comparison, using 0 as the
and the other is numeric Empty expression.

One expression is Empty Perform a string comparison, using a zero-length

and the other is a string string (“”) as the Empty expression.
Both expressions are Empty The expressions are equal.

The second type of comparison operator is the object comparison operator. The IS operator is
used to compare object references. If you are comparing objects, the following syntax is used:
myResult = object1 IS object2

where object1 and object2 represent objects and the use of the keyword IS determines a
relationship between objects.

When comparing objects, there is no NULL possibility. The relationship is either True or False.
If the objects are the same, then the IS function returns True. If the objects are not the same,
the function returns False.

07.1389-6 CH04 76 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Comparison Operators 77

Equality Comparison
The equality operator is the most commonly used comparison operator. The equality compari-
son is used to determine whether a condition is equal to a value, and in most situations, pro-
vides the coding logic to react to a decision. An example of an equality comparison would be a
test to determine the color of a shirt. This test would be implemented by comparing the value
of the variable myShirt to determine whether the color of a shirt was blue. To perform the
equality comparison, use the following syntax:
Expression = Condition

where Expression and Condition are any valid VBScript expression. To test whether the color
of the shirt is Blue, you would then use the equality (=) comparison operator in any type of
decision-making process:
<% If myShirt = “Blue” Then %> ‘=== If then loop with equality comparison
operator

or
<% Do While myShirt = “Blue” %> ‘=== Do while loop with comparison operator

What’s interesting about the equality comparison is that the comparison can be made either
implicitly or explicitly. An implicit equality comparison is used to evaluate whether a specific Part
expression is True or is often used to determine if larger categories of values are True. For II
example, you may not be interested in testing for a specific time value, but you might want
Ch
to know whether the current variable actually contains a date value, as shown in the following
script: 4
<%
timeStamp = Now
If IsDate(timeStamp) Then
Response.Write “The time is “ & timeStamp
End If
%>

The script accomplishes this by using the IsDate function to test whether the VBScript vari-
able timeStamp contains a date value. Because the NOW function returns the existing system’s
time and date, the ISDATE function in the first line would return True. In this situation, after
successfully testing the IsDate function, the next line of script writes the current date and time
to the page.

An explicit equality comparison verifies that an evaluated expression and a variable’s value are
equal. For example, the following example is used to see what fruit is on sale today and set the
prices accordingly. If the variable vFruit =”Apple”, then the following comparison would fail.
<%
If vFruit = “banana” Then ‘=== Does ‘Apple’ = ‘Banana’, no
vStr = “Sale Today” ‘=== If true then set additional variable
vPrice = vSalePercent * vProductCost
Else ‘=== Comparison is not equal, so this is

07.1389-6 CH04 77 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 78 Chapter 4 Working with Operators

➥executed
vStr = “Special Today”
vPrice = vSpecialPercent * vProductCost
End If
%>

Inequality Comparison
The inequality operator (<>) provides the opposite functionality of the equality comparison,
ensuring that an evaluated statement is not equal to a variable’s value. This type of comparison
is often used when you need to test for a range of values instead of a specific value. For ex-
ample, if you need to test for all color shirts not equal to ‘Blue’, then use the inequality opera-
tor to search for all values not equal to ‘Blue’. This inequality comparison is demonstrated in
the following example:
<%
If myShirt <> ‘Blue’ Then
MsgStr = “You look better in blue” ‘=== This line is not executed
End If
%>

Greater Than Comparison

The greater than (>) comparison determines whether one expression is larger than another
expression. For example, if the variable vAcctBalance =”200”, the following logic would fail
and the value of vChargePercent would be equal to 0.25636.
<%
If vAcctBalance > “500” Then
vChargePercent = 0.23953
Else
vChargePercent = 0.25636
End If
%>

Less Than Comparison

The less than (<) comparison operator, opposite in functionality to the greater than operator,
determines whether the expression on the left side of the operator is smaller or less than the
expression on the right side. For example, the following comparison will return True if the
current value of the vGasTank variable is “4.5”.
<%
If vGagTank < 6.5 Then
vMsg = “Better get gas soon!”
End If
%>

07.1389-6 CH04 78 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Logical Operators 79

Object Comparison
Object comparison (IS) operator can only be used to compare object types and cannot compare
other objects properties. The following example demonstrates how the IS operator is used to
compare objects. The Set statement is first used to create replica objects, MySecondObject and
MyThirdObject, from the object MyObject. Next, the Set statement is used to create another
object, YourSecondObject, from the object YourObject.
<%
Dim MyObject, MySecondObject, MyThirdObject, YourObject, YourSecondObject,
➥MyResult
Set MySecondObject = MyObject ‘=== Assign object references.
Set MyThirdObject = MyObject
Set YourSecondObject = YourObject
MyResult = MySecondObject IS MyThirdObject ‘=== Returns True.
MyResult = YourSecondObject IS MyThirdObject ‘=== Returns False.
%>

Remember that the Set command is used to declare and reference objects. The Dim command
is used to declare and reference variables.

Logical Operators Part

Logical operators help to expand the functionality found in comparison operators by using a II
different evaluation approach. The logical operator focuses on the Boolean method of deter- Ch
mining whether a process returns True or False (see Table 4.7).
4
Table 4.7 Logical Operators
Operator Symbol Description
Conjunction AND Performs a logical conjunction.
Negation NOT Performs logical negation.
Disjunction OR Performs a logical disjunction.
Exclusion XOR Performs a logical exclusion.
Equivalence EQV Performs a logical equivalence.
Implication IMP Performs a logical implication.

The VBScript operators work behind the scenes performing a bitwise comparison of bits that
represent numeric expressions. Comparing the two expressions is simply a straightforward
comparison of bits (either a 0 [zero] or 1 [one]), as demonstrated in Table 4.8. It’s important to
note that these logical operators operate on Boolean expressions. Therefore, in evaluating the
expression myVarB = Not (yourVarA), where yourVarA = ‘Valid’, the NOT operator would
return a ‘Type mismatch’ error.

07.1389-6 CH04 79 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 80 Chapter 4 Working with Operators

Table 4.8 Bitwise Comparison: The Foundation of Logical Operators

Bit Expression1 Bit Expression2 Result
0 0 0
0 1 1
1 0 1
1 1 1

Negation (NOT)
The negation comparison (NOT) is used to perform logical negation on an expression. This
logical comparison is similar to the negatory comparison operator. The NOT operator uses the
following syntax:
myResult = Not myExpression

where myExpression is any expression. The NOT operator actually performs its operations by
inverting the bit value of the variable. Therefore, if the variable myExpression is equal to True,
the result of the negation operator would be False. The negation operator is used to return one
of three results: True, False, or NULL, as shown in Table 4.9.

Table 4.9 The Results of the Negation (NOT ) Operator

Expression Result
True False

False True

NULL NULL

Remember that a NULL value represents a variable that contains no valid data. Consequently,
any operator that performs a function on this variable will reproduce a corresponding NULL
result. Therefore, it is very important to declare your datatypes within your Active scripts. Also,
keep in mind that an empty string (“”) is not equivalent to a NULL string, as seen in the code
listing below.
<%
MyVar = “”
myResult = IsNull(MyVar) ‘=== Returns False.
%>

<%
MyVar = NULL
MyResult = IsNull(MyVar) ‘=== Returns True.
%>

07.1389-6 CH04 80 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Logical Operators 81

A NULL string is a Variant datatype that contains no valid data, whereas an Empty string is a
Variant datatype that contains data, but of length zero. An Empty string is most often the result
of an un-initialized or unassigned variable.

Conjunction (AND)
The logical conjunction (AND) is used to group two or more comparisons together and evaluate
the group of comparisons together as one comparison. The AND conjunction requires that all
comparisons in the group evaluate to True for the conjunction to evaluate to True. If any one
comparison evaluates to False, then the conjunction evaluates to False. For example, the code
listed below will fail because myVarA is less than myVarC.
<%
Dim myVarA, myVarB, myVarC
myVarA = 40
myVarB = 25
myVarC = 55
If (myVarA > myVarB) AND (myVarA > myVarC) Then
‘=== This statement will not be executed
End If
%>

The previous example demonstrates that both expressions, (myVarA > myVarB) and (myVarA > Part
myVarC) must evaluate to True. However, even though the first expression is True, the AND II
operator fails because the second expression returns False. The behavior of the logical AND
Ch
operator can be seen in Table 4.10.
4
Table 4.10 The Logical AND Operation Conditions and Results
Expression1 Expression2 Result
True True True

True False False

True NULL NULL

False True False

False False False

False NULL False

NULL True NULL

NULL False False

NULL NULL NULL

07.1389-6 CH04 81 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 82 Chapter 4 Working with Operators

Disjunction (OR)
The logical disjunction (OR) operator is used to test whether any of a series of expressions
evaluates to True. The OR operator requires the following syntax:
myResult = expression1 OR expression2

where expression1 or expression2 is any valid VBScript expression.

The OR operator provides different programming flexibility than the logical AND operator. The
OR operator will evaluate to True if any of the comparisons evaluate to True, whereas the AND
operator will evaluate to True only if all the comparisons evaluate to True.
<%
Dim A, B, C, D, MyResult
A = 100: B = 80: C = 60: D = Null ‘=== Initialize variables.
MyResult = A > B Or B > C ‘=== Returns True.
MyResult = B > A Or B > C ‘=== Returns True.
MyResult = A > B Or B > D ‘=== Returns True.
MyResult = B > D Or B > A ‘=== Returns Null.
%>

Table 4.11 provides the conditions and results of the OR logical operator.

Table 4.11 The Logical OR Conditions and Results

Expression1 Expression2 Results
True True True

True False True

True NULL True

False True True

False False False

False NULL NULL

NULL True True

NULL False NULL

NULL NULL NULL

Exclusion (XOR)
The exclusion operator (XOR) is used to test whether two expressions differ in results. The
exclusionary operator requires the following syntax:
myResult = expression1 XOR expression2

where expression1 and expression2 are valid VBScript expressions. The best way to see how
the exclusionary operator works is demonstrated in the following example. The first and third

07.1389-6 CH04 82 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Logical Operators 83

exclusionary comparisons return a False value because both expressions individually return
the same value. Essentially, the XOR operator is testing whether the two expressions return
different values. The second exclusionary comparison returns a True value because the two
individual comparisons returned different values, one True and the other False.
<%
Dim A, B, C, D, MyResult
A = 100: B = 80: C = 60: D = Null ‘=== Initialize variables.
MyResult = A > B XOR B > C ‘=== Returns False.
MyResult = B > A XOR B > C ‘=== Returns True.
MyResult = B > A XOR C > B ‘=== Returns False.
MyResult = B > D XOR A > B ‘=== Returns NULL.
%>

A more complete display of the XOR results, depending on the expressions, is illustrated in
Table 4.12.

Table 4.12 The XOR Logical Operator Conditions and Results

Expression1 Expression2 Results
True True False
Part
True False True
II
False True True
Ch
False False False
4
Equivalence (EQV)
The equivalence operator (EQV) is used to perform a logical equivalence of two expressions.
The equivalence operator requires the following syntax:
myResult = expression1 EQV expression2

The equivalence operator is use to determine whether the two expressions produce similar
results. For example, consider the following script:
<%
Dim A, B, C, D, MyResult
A = 100: B = 80: C = 60: D = Null ‘=== Initialize variables.
MyResult = A > B EQV B > C ‘=== Returns True.
MyResult = B > A EQV B > C ‘=== Returns False.
MyResult = A > B EQV B > D ‘=== Returns NULL.
%>

The evaluation of the first EQV expression in the statement A > B EQV B > C will return True
because both expressions produce similar results. The second EQV expression, B > A EQV B >
C, returns False because the two expressions produce dissimilar results. The third expression
returns NULL, illustrating the cascading effect of using NULL and VBScript operators. The EQV
logical operator results are demonstrated in Table 4.13.

07.1389-6 CH04 83 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 84 Chapter 4 Working with Operators

Table 4.13 The EQV Logical Operator Conditions and Results

Expression1 Expression2 Results
True True False

True False True

False True True

False False False

Concatenation Operators
There are two operators in VBScript that are used to link or join strings together (see Table
4.14). The ampersand (&) operator is the true concatenation operator. The addition (+) operator
can also be used to join strings. However, the addition operator, as described above, is prima-
rily designed for adding numeric datatypes and can result in unpredictable results depending
on the variable datatype.

Table 4.14 Concatenation Operators

Operator Symbol Description
String Concatenation & Used to concatenate strings.
String and Numeric Operations + Used for string concatenation and
numeric addition.

& Operator
The ampersand (&)operator is used to concatenate or join strings expressions together. The &
operator requires the following syntax:
myResult = expression1 & expression2

where expression1 and expression2 are valid VBScript string expressions. A common use of
the ampersand operator is to collect information from different objects associated with Web
pages and construct a resulting string. This resulting string is often used to construct URLs
with user-defined variables or to create SQL strings for database access. For example, the
following code uses the concatenation operator to construct a string that will be used to access
product information from a database.
<%
Dim sqlStr
sqlStr = “SELECT * “
sqlStr = sqlStr & “FROM Products” ‘=== Generates the string “SELECT * FROM
➥Products”
%>

07.1389-6 CH04 84 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 Operator Precedence 85

You can also use this operator to create a string that uses values pulled from any object within
the script’s scope. For example, the following code takes the value in a textbox named
txtFirstName on a form named frmLogin and concatenates it with the text “Hello” to create a
string that holds “Hello” and the user name.
<%
Dim MyStr
MyStr = “Hello” & Trim(frmLogin.txtFirstName.value) ‘=== Returns “Hello “
➥and user name.
%>

The ampersand operator also has special rules regarding data subtypes. If the concatenated
expression is not a string, it will be converted to a String subtype. If either of the expressions
is a NULL, then the resulting concatenated string is a zero-length string (“”). However, if both
expressions are NULL, the ampersand returns a NULL value.

+ Operator
The addition (+) operator can also be used to append strings, but using this operator for string
manipulation is not recommended because the + operator performs two functions: the addition
of numeric values and the concatenating of string values. The ambiguity occurs when the +
operator is functioning on expressions of different subtypes. The addition operator requires the Part
following operator syntax: II
myResult = expression1 + expression2 Ch
where expression1 and expression2 are valid VBScript expressions. 4
If both expressions are numeric, the + operator will execute its numeric summarization fea-
tures. If both expressions are strings, the + operator will concatenate the strings. If one expres-
sion is a Numeric and the other expression is a String subtype, then the + operator will add
them together.
<%
dim vVar, vStr
vVar = 30: vStr = “5”
myResults = vVar + vVar ‘=== Returns 60
myResults = vStr + vStr ‘=== Returns 55
myResults = vStr + vVar ‘=== Returns 35
%>

Operator Precedence
Operator precedence is used to describe the order in which different operators are evaluated in
a combination of keywords, operators, and variables. VBScript is similar to Visual Basic in
terms of the order in which VBScript systematically processes expressions.

Expressions are evaluated based on the category of the operator, as shown in the following list.
Arithmetic operators are processed first, followed by the comparison operators, and then logi-
cal operators are evaluated last.

07.1389-6 CH04 85 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 86 Chapter 4 Working with Operators

1. Arithmetic operators
2. Comparison operators
3. Logical operators

TIP Concatenation are evaluated after the arithmetic operators and before the comparison operators.

The precedence of arithmetic and logical operators within their own operator group is defined
in Table 4.15. The arithmetic operators are executed first from the exponential operator to the
subtraction operator. The logical operators’ execution priorities range from the logical negation
operator to the logical implication operator.

Table 4.15 The Arithmetic and Logical Operators’ Execution Hierarchy from
Highest to Lowest
Description Symbol Description Symbol
Exponential ^ Negation NOT

Unary Negation - Conjunction AND

Multiplication * Disjunction OR

Division / Exclusion XOR

Integer Division \ Equivalence EQV

Modulus Arithmetic MOD Implication IMP

Addition +

Subtraction -

Unlike the hierarchical execution approach of the arithmetic and logical operators, the com-
parison operators all share the same level of importance. The comparison operators are
executed in order of appearance from left to right. A similar left-to-right rule also applies to
situations where multiplication and division operators or addition and subtraction operators
occur together.

To add programming control to the use of operators, the use of parentheses overrides the
default operator precedence described above. All operations within parentheses take prece-
dence over operations outside parentheses and nested parentheses. However, all operations
within parentheses follow the default operator precedence. For example, consider the following
script:
<%
Response.Write (5 + 10 * 20 - 5) &”<BR>” Returns 200
Response.Write (5 + 10 * (20 - 5)) &”<BR>” Returns 155
%>

07.1389-6 CH04 86 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 From Here… 87

The first line of the preceding script first performs the multiplication of 10 * 20. Then the
addition operator adds 5 to the total and the subtraction operator subtracts 5 to return 200 as
the result. However, the second line of code first executes the expression in the parentheses
(20-5). Afterward, that result is multiplied by 10 and then 5 is added to the result to return 155.

From Here…
In this chapter, you have learned the various operators that provide the different types of
comparison logic in VBScript. The arithmetic operators are used to manipulate your numeric
expressions and mathematical functions. The comparison operators compare and contrast
variables, datatypes, and objects. VBScript uses logical operators to apply comparison opera-
tors to Boolean expressions and concatenation operators to append strings expressions. Fi-
nally, you saw that VBScript also has a defined order of operations that dictates in what order
operators will be processed.

At this point, you are ready to learn how VBScript uses these operators with decision operators
to help control program and application workflow in your ASP scripts.

■ Chapter 5, “Controlling Program Flow with VBScript,” demonstrates how to use logical
and looping operators to build your applications. Part
■ Chapter 6, “Working with Functions, Subroutines, and Dialog Boxes,” illustrates event- II
driven programming through the use of reusable code modules.
Ch
■ Chapter 7, “Datatype Conversion Features,” describes the datatype conversion functions
in VBScript that ensure proper application execution. 4
■ Chapter 8, “Putting It All Together with VBScript,” demonstrates how to use VBScript to
perform calculations and display the results using Active Server Pages, and illustrates
how to use client-side validation on the Internet Explorer browser.
■ Chapter 9, “ Using the Script Debugger,” describes how to use the Microsoft Script
Debugger to provide inline-debugging capabilities for your client- and server-side scripts.

07.1389-6 CH04 87 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 88 Chapter 4 Working with Operators

07.1389-6 CH04 88 10/26/98, 9:07 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH04 lp#4
 5
89

C H A P T E R

Making a Decision
Controlling Program Learn what types of programmatic
control mechanisms exist for con-

Flow with VBScript trolling application workflow in

VBScript and how to implement the
conditional logic in your application.

Using Looping Logic

Discover the different looping
structures available in VBScript and
how they are used to repeat a pro-
Now that the basics of VBScript variables and operators cess until a condition is met.
have been covered, this chapter will cover how to control
Stopping the Process Flow
program execution flow in VBScript. In Chapter 3, you
Investigate how to programmatically
learned how to store and retrieve values from variables.
interrupt your script execution
Chapter 4 illustrated how to use operators to compare,
using the Exit statement and how to
evaluate, and manipulate variables. The next step is to handle error trapping in VBScript.
investigate the various control structures in VBScript that Part
are used to create the fundamental logic driving your Error Trapping II
application. This program execution control comes in two Learn how to use VBScript’s Err Ch
forms: decision structures and looping structures. Deci- object to handle runtime errors and
sion structures execute conditional statements and, as a error trapping in your ASP and 5
result of a test, decision branches of the appropriate result client-side scripts.
statements. Looping structures cyclically process a section
of code until the conditional statement is satisfied. ■

08.1389-6 CH05 89 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 90 Chapter 5 Controlling Program Flow with VBScript

Using Decision Structures

Making a decision in VBScript is possible through three basic decision structures: If...Then,
If...Then...Else, and Select Case decision trees. These structures test conditions and
perform actions depending on the logic contained in the code. In most situations, the decision
statements are largely responsible for making decisions that drive the business rules for your
applications. The If...Then statement is used to test a condition and execute a block of code if
the condition is valid. The If...Then...Else statement performs the same functionality as the
If...Then statement, but it gives you the option of executing a second block of code if the condi-
tion is not satisfied. Finally, the Select Case statement enables you to test for multiple conditions
rather than limit yourself to simple true/false situations. The Select Case statement lets you test
for multiple conditions and execute blocks of code based on the value of an expression.

The If...Then Condition

The most commonly used decision-based control structure in VBScript is the If...Then
statement. The If...Then statement is used to test a Boolean (True or False) condition and
execute script based on the results. If the evaluated condition is True, the immediate script
following the Then statement is executed. The If...Then statement uses the following syntax:
IF condition THEN
statement
End If

where condition is an evaluated expression and statement represents a block of code that is
executed if condition is True. Once the statement is finished processing or the condition is
returned False, program execution is returned to the code line immediately following the End
If statement.

The condition must be an expression that can be evaluated as either True or False. The re-
sultant True or False transcription enables the scripting logic to present a straightforward
approach to controlling logic and program workflow. For example, the following situation deter-
mines whether the conditional statement of myVar > 50 is True. If so, the condition returns True
and the string message is assigned. However, if myVar is less or equal to 50, the expression evalu-
ates to False and programmatic control skips to the next statement after the End If statement.
<%
Dim myVar
myVar = 100 ‘=== Initialize variables
if (myVar > 50) Then ‘=== Condition evaluates True
strMsgStr = “Do you want to play again?” ‘=== Executed line
End if
%>

Besides testing the Boolean result of a comparison operator as described above, you can also
test an expression that directly returns True or False. For example, consider the following
situation that demonstrates using the IsDate function to execute a multi-line script block that
writes the number of days left until the year 2000. If the variable myDate is not a valid date, such
as “00/00/1999”, the resultant statements are not executed. For more information about the
VBScript’s evaluation functions, please see Chapter 7, “Datatype Conversion Features.”

08.1389-6 CH05 90 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Using Decision Structures 91

<%
dim myDate
myDate = now ‘=== Initialize
➥to today’s date
If IsDate(myDate) Then ‘=== Is myDate
➥a Date Sub Type
Dim myDays
myDays = DateDiff(“d”,myDate,”1/1/2000") ‘=== Find days
➥until year 2000
Response.Write myDays &” days left until the year 2000!” ‘=== write
➥results to page
End if
%>

You can also expand the testing conditions of the If...Then statements to contain other
If...Then statements. Nesting conditional statements is accomplished by placing one
If...Then statement in the result block of another If...Then statement. Nesting conditional
statements enables you to incorporate multiple dependent conditions in your code. For ex-
ample, the following code uses nested conditions to remind you to check your credit card
balance on New Years day in the year 2000. The second nested loop executes if you set the
variable myDate to any valid date greater than or equal to “1/1/2000”.
<%
myDate = now ‘=== Initialize to today’s
➥date
if IsDate(myDate) then ‘=== Is it a valid date?

If (DateDiff(“d”,”1/1/2000",myDate) >= 0) then ‘=== Is the current date

➥past 1/1/2000?
Response.Write “Have you checked you credit card balance today?”
End if Part

End if
II
%> Ch
This previous example first ensures that the variable myDate is an actual date. If the variable is 5
a date, then the second conditional If...Then statement is used to only write the warning
message if the date is 1/1/2000.

This single result approach of the If...Then statement to applying application logic has limited
use. In most situations, you often need to implement an either/or approach to application
workflow. This branching capability is demonstrated with the If...Then...Else statement.

The If...Then...Else Statement

The If...Then...Else statement expands the traditional If...Then statement by adding a
failure condition using the Else statement. This Else condition enables the decision logic to
present a failure option within the original If condition. This internal routing separates script-
ing code that would otherwise have to be placed in a second If...Then statement. The
If...Then...Else statement requires the following syntax:

If condition Then
statements

08.1389-6 CH05 91 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 92 Chapter 5 Controlling Program Flow with VBScript

Else
elsestatements
End if

where condition is a Boolean expression or function, statements is the code that is executed
if the condition statement evaluates to True, and elsestatements is the code that is executed if
the condition evaluates to False.

The If...Then...Else presents two different directions the application logic can follow. First,
you can combine the previous illustrated nested If...Then examples into one conditional
If...Then...Else condition. For example, the following example uses the If...Then...Else
statement to determine the amount of days between the current date and the year 2000. If
there are still days left before the year 2000, the script will present a friendly reminder showing
the amount of days left until the year 2000. Once the year 2000 arrives, the script will remind the
user to check his credit card balance.
<%
dim myDays
myDays = DateDiff(“d”, Now, “1/1/2000”) ‘=== Find amount of days until
year 2000
If myDays > 0 Then ‘=== Still time left
Response.Write myDays &” days left until the year 2000!”
Else ‘=== Gulp!
Response.Write “Have you checked you credit card balance today?”
End if
%>

Notice that the creation of the variable myDays is not necessary for the conditional statement.
The statement If (DateDiff(“d”, Now, “1/1/2000”) > 0) then could replace the statement
if myDays > 0. However, because you want to use the results of the DateDiff function later in
the code block, fewer resources are consumed by storing the variable once instead of requiring
additional processor cycles to recalculate the DateDiff function for the display message.

You can further expand the If...Then...Else functionality to enable multiple test conditions
by adding the ElseIf statement. Use the following syntax to provide multiple test condition:
If condition Then
statements
ElseIf condition-n Then
elseifstatements
Else
elsestatements
End If

Now you can test conditional statements for more than two conditions at a time. For example,
the following code illustrates assigning the tax form helper documentation according to the
number of exemptions an employee selects.
<%
Dim myExemptions, myTaxHelperID ‘=== Initialize variable
myExemptions = 2 ‘=== Set variable
If myExemptions = 0 Then
myTaxHelperID = “Publication 1678”

08.1389-6 CH05 92 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Using Decision Structures 93

ElseIf myExemptions = 1 then

myTaxHelperID = “Publication 1713”
ElseIf myExemptions = 2 then ‘=== This row selected
myTaxHelperID = “Publication 1779”
Else
myTaxHelperID = “Publication 1648”
End if
Response.Write myTaxHelperID ‘=== Write to page the selected result
%>

In the previous example, the code evaluated in the ElseIf condition evaluates until the condi-
tion returns True. When the condition returns True, the code following the Then statement is
executed. If the ElseIf condition evaluates to False, then the next conditional statement is
evaluated. For example, in this situation, the first two Else conditions are evaluated to be
False. However, the third condition returns True and the resultant Then conditional block is
executed. In this example, the variable myTaxHelperID is set to “Publication 1779”. If none of
the conditions in the entire If...Then...Else blocks evaluate to True and an Else condition
exists, the Else statement will be executed. The Else statement acts as the default code block
for the entire If...Then...Else block. If the Else statement does not exist, then no code is
executed.

TIP Always include an Else statement to act as the default statement in case no conditions evaluate to
True. This will minimize potential shortcomings in programming logic when all the evaluation
conditions fail.

The value of the Else condition should not be underestimated; it plays a critical role in deter-
mining workflow. For example, the following code sets the variable myExemptions by accepting
Part
a value from a hyperlink using the Active Server Page Response object. This code segment is
responsible for evaluating the input and setting the Web page to redirect the browser. For II
example, in the previous code example, the variable myExemptions was set to 2. In this case, Ch
this code would set the variable myTaxHelperID equal to Publication 1779. Then the variable
gotoURL would create the string ‘Publication 1779.asp’ and automatically direct the user to
5
that page, assuming it exists.
<%
Dim myExemptions, myTaxHelperID ‘=== Initialize variable
myExemptions = Response. QueryString(“taxExemption”) ‘=== Retrieve value from
➥the URL string
If myExemptions = 0 Then
myTaxHelperID = “Publication 1678”
ElseIf myExemptions = 1 then
myTaxHelperID = “Publication 1713”
ElseIf myExemptions = 2 then
myTaxHelperID = “Publication 1779”
Else
myTaxHelperID = “Retry”
End if
Dim gotoURL ‘=== Initiate variable
gotoURL = myTaxHelperID &”.asp” ‘=== Create the page
➥location to browse

08.1389-6 CH05 93 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 94 Chapter 5 Controlling Program Flow with VBScript

Response.Redirect gotoURL ‘=== Send user to

➥correct page
%>

If the calling page sends in a value of null or a zero-length string for the ‘taxExemption
variable’, the Else condition would be evaluated to create a retry.asp page. This page would
give the user another chance to select from the list of correct forms. However, it you elected
not to use the Else condition, the variable myTaxHelperID would never be set. Consequently, an
HTTP error would result trying to redirect the browser to a page named ‘.asp’. For more
information about the Response object, see Chapter 14, “Speaking to the Web Client: The
Response Collection.”

Although the use of nested ElseIf statements enables the selection of one execution block
from multiple execution blocks, doing so requires that many separate conditions be evaluated.
The Select Case statement has been designed to provide a more effective mechanism for
accomplishing this.

The Select Case Statement

The Select Case statement is designed to improve the efficiency of code that needs to accom-
plish extended If...Then...ElseIf statements. The Select Case statement is used to select a
code block by evaluating the test condition once and then comparing the result with multiple
conditions. The If...Then...Else statements, on the other hand, require conditional testing
for every conditional statement. The syntax of the Select Case statement is
Select Case testexpression
Case expressionlist
statements . . .
Case Else expressionlist
elsestatements
End Select

where testexpression is any numeric or string expression and expressionlist is one or

more of the possible matching expressions.

The logical sequence of the Select Case statement is similar to the If...Then...Else state-
ments where both conditions enable the case for a failure condition. The Else clause in the
Select Case statement is used when the conditional statement does not match any expression
in the expression list. For example, the previous If...Then...ElseIf example can be rewrit-
ten into a Select Case statement.
<%
Dim myExemptions, myTaxHelperID ‘=== Initialize variables
myExemptions = 2 ‘=== Set variable
Select Case myExemptions ‘=== Evaluate condition
Case 0
myTaxHelperID = “Publication 1678”
Case 1
myTaxHelperID = “Publication 1713”
Case 2
myTaxHelperID = “Publication 1779” ‘=== Evaluates line
Case Else
myTaxHelperID = “Publication 1648”

08.1389-6 CH05 94 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Using Decision Structures 95

End Select
Response.Write myTaxHelperID ‘=== Write results
%>

Likewise, you can convert the If...Then...Else example into a Select Case statement to
minimize processing power required to route the browser to the appropriate page depending
on the input.
<%
Dim myExemptions, myTaxHelperID ‘=== Initialize variable
myExemptions = Response. QueryString(“taxExemption”) ‘=== Retrieve value from
➥the URL string
Select Case myExemptions ‘=== Evaluate condition
Case 0
myTaxHelperID = “Publication 1678”
Case 1
myTaxHelperID = “Publication 1713”
Case 2
myTaxHelperID = “Publication 1779”
Case Else
myTaxHelperID = “Retry”
End Select
Dim gotoURL ‘=== Initiate variable
gotoURL = myTaxHelperID &”.asp” ‘=== Create the page
➥location to browse
Response.Redirect gotoURL ‘=== Send user to correct
➥page
%>

By using the Select Case features, you can turn this decision operator into a very robust tool
in developing your applications. For example, the Select Case statement can be used to imple-
Part
ment a client-side validation method to ensure that data is entered correctly on a form.
Function FieldValidate()
II
Dim ret Ch
Title = “Validation Errors”
Select Case True 5
Case ((Len(Document.frmOrder.txtFirstName.value)) < 1)
MsgBox “Please enter your first name”, 48 ,title

Case ((Len(Document.frmOrder.txtLastName.value)) < 1)

MsgBox “Please enter your last name”, 48, title

Case ((Len(Document.frmOrder.Txtamount.value)) < 1)

MsgBox “Please enter an amount greater than zero”, 48, title
Case ((Len(Document.frmOrder.Txtwidgetname.value)) < 1)
MsgBox “Please enter a Widget Name”, 48, Title
Case else
FieldValidate = True
End Select
End Function

In this example, the Select Case statement searches through the case criteria looking for the
first expression that returns True. Each condition is used to identify a required field. For ex-
ample, Figure 5.1 demonstrates a form that requires values in all text box fields.

08.1389-6 CH05 95 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 96 Chapter 5 Controlling Program Flow with VBScript

FIG. 5.1
The Data Widgets
Incorporated HTML order
form requires values in
all text boxes.

The Select Case statement is used in this example to determine whether one of the required
fields on the form is blank. The Select Case will test each expression to evaluate to True.
When a True statement is found, a message box is displayed to alert the user of the required
field.

For more information about using user dialog boxes such as the message box, please see
Chapter 6, “Working with Functions, Subroutines, and Dialog Boxes.”

Implementing Looping Logic

Looping structures in VBScript provide the capability to cyclically repeat code. The repeating
process can occur a set number of times or until a condition is met. VBScript provides four
looping structures to help control your application’s programming logic: the For...Next,
Do...Loop, While...Wend, and the For Each...Next loops.

The For...Next Statement

The For...Next loop repeats a section of code a specific number of times. This is particularly
useful if you know the number of cycles or iterations required. The syntax for the For...Next
loop is the following:
For counter = start To end Step stepsize
statements
Exit For
statements
Next

08.1389-6 CH05 96 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Implementing Looping Logic 97

where counter is a numeric variable that maintains the current iteration number, and start
and end provide the starting and ending counter positions.

CAUTION
Notice that the Next operator does not require the counter variable as in Visual Basic. If you add the
counter variable, VBScript will generate an Expected end of statement runtime error.

The For...Next loop also enables you to specify the step size between iterations. The step size
is added to the counter to determine the value of the next counter. In addition, the step size can
also be positive or negative. A positive step size will iterate from the starting position and finish
when the counter is greater than or equal to the end counter. The following example uses a
For...Next loop with a counter set to start at 65 and end at 122 to print every third ANSI char-
acter value.
<%
Dim myCounter
%>
<%
For myCounter = 65 to 122 step 3
Response.Write chr(myCounter) ‘=== Writes ‘A D G J M P S V Y etc...
Next
%>

If the step size is negative, it also starts at the beginning position and adds the negative step
size to the position to determine the next counter value. For the negative step size, the
For...Next loop will end when the counter is less than or equal to the ending value. The follow-
ing example cycles through a negative step size to display the lowercase values of the alphabet. Part

<% II
For myCounter = 122 to 97 step -1
Ch
Response.Write chr(myCounter) ‘=== Displays results:
zyxwvutsrqponmlkjihgfedcba
Next
5
%>

Keep in mind that the Next operator in VBScript differs from Next operator in native Visual
Basic. Native Visual Basic requires a certain counter variable to be specified in the Next state-
ment. However, VBScript does not require the counter variable. In the Next operator, a state-
ment automatically increments the counter variable designated in the For statement. If you add
the counter variable, VBScript will generate an Expected end of statement runtime error.
Furthermore, if you nest For...Next loops, VBScript prevents naming the counter variables
the same name. For example, the following code generates the VBScript runtime error
Invalid ‘For’ loop control variable.

<%
Dim myCounter ‘=== Initiates Variable
For myCounter =1 to 3 ‘=== Starts first FOR loop
For myCounter = 1 to 10 ‘=== Generate Run Time Error
If (myCounter = 4) Then
Exit For

08.1389-6 CH05 97 10/26/98, 9:08 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 98 Chapter 5 Controlling Program Flow with VBScript

Else
Response.Write chr((100+ myCounter))
End If
Next
Response.Write myCounter
Next
%>

You can also use this For...Next logic to step from a collection of the same objects on the Web
client. For example, Figure 5.2 illustrates VBScript that enables you to select your favorite
browser type.

FIG. 5.2
Client-side example of
using a For...Next
loop to cycle through
controls on HTML page.

The For...Next loop can now be applied to cycle through the controls and display what button
was displayed by the browser:
<SCRIPT LANGUAGE=”VBScript”>
<!--
Sub checkradio()
For I = 0 to 2
If document.frmbrowser.r1.item(i).checked then
MsgBox “You have selected “ & document.frmbrowser.r1.item(i).value
End if
Next
End Sub
-->
</SCRIPT>

When the Make Choice button is pressed, the previous code is executed to loop through the
radio box searching for the selected item. After the radio box that is checked is found, a mes-
sage box is displayed with the result.

08.1389-6 CH05 98 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Implementing Looping Logic 99

The Do...Loop Statement

The Do...Loop is also used to execute cyclical code, but unlike the For...Next loop, it contin-
ues to perform an operation indefinitely until a condition is met. The Do...Loop offers two
conditional operators to evaluate the conditions: the While and Until operators. Both operators
continue processing until the evaluated function returns True.

The Do...While/Until loops can be implemented in two variations:

Do {While | Until} condition
statements
Exit Do
statements
Loop

Or alternatively:
Do
Statements
Exit Do
statements
Loop {While | Until} condition

The difference between the two syntaxes is the location of the conditional statement. In the
first syntax structure, the conditional statement is located in the same execution line as the Do
key word. In the second Do...Loop syntax, the conditional statement is located at the end of
the loop code. This change in location gives programmers the flexibility to test the validity of
the condition first and then process the code block, or to process the code block first and then
evaluate the condition. The following example demonstrates the use of the Do...Loop:
<% Part
Dim myCounter, myMaxValue
myMaxValue = 100 II
Do While myCounter < myMaxValue ‘=== Inner loop.
Ch
myCounter = myCounter + 1 ‘=== Increment Counter.
If myCounter = 50 Then
Exit Do
‘===
‘===
Test condition
Exit loop.
5
End If
Loop
%>

The Do...Loop can also contain other nested Do...Loops. The following example cycles
through the first six months of the year and then cycles through the number of days in each
month. (This example uses some functions that have not been covered yet, but will be covered
in upcoming chapters.)
<%
Dim myMonthCount, myTotalMonths, myYear
myMonthCount = 1: myTotalMonths = 6: myYear = 1997 ‘=== Initialize
➥variables
Do While myMonthCount <= myTotalMonths ‘=== Start outer
➥month loop
Dim myFirstDate, mySecondDate, iDayCount
Response.Write MonthName(myMonthCount) ‘=== Write month
myFirstDate = (myMonthCount &”/01/” &myYear) ‘=== Sets first
➥month

08.1389-6 CH05 99 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 100 Chapter 5 Controlling Program Flow with VBScript

mySecondDate = (myMonthCount+1 &”/01/” &myYear) ‘=== Sets second

➥month
myDayCount = DateDiff(“d”, myFirstDate, mySecondDate)+1 ‘=== determines
➥days per month
iDayCount = 1 ‘=== Initialize
➥counter
Do While iDayCount < myDayCount ‘=== Inner day
➥counter
Response.Write “ “ &iDayCount ‘=== Write results
iDayCount = iDayCount + 1 ‘=== Iterate day
Loop ‘=== Inner loop
Response.Write “<BR>” ‘=== Start new line
➥on new month
myMonthCount = myMonthCount + 1 ‘=== Iterate to a
➥new month
Loop ‘=== Outer loop
%>

This example demonstrates using multiple Do...Loops to cycle through a six-month period of
time and write the amount of days in each month to the Web page. Two loops are needed for
this function: one to cycle through the amount of months, the other loop is needed to cycle
through the amount of days in each month. This dual looping approach is commonly used
when dealing with groups of similar objects. For example, nested Do loops would be needed to
display all credit card transactions for all users on a corporate charge card.

The While...Wend Statement

Another looping operator that is similar in behavior to the Do...Loop is the While...Wend
statement. The While...Wend loop repeats an operation while a condition is True. The
While...Wend statement uses the following syntax:

While condition
statements
Wend

where condition is a numeric or string expression that can be evaluated to True or False, and
statements represents a code block that is executed when the condition evaluates to True.

When the While condition is True, all statements are processed until the Wend statement is
reached. At that point, the program control is shifted back to the While condition and the condi-
tion is evaluated again. The cycle continues until the condition returns False. At that point, the
statement following the Wend statement is executed. For instance, the following example uses
the While...Wend statement to increment the counter variable by one when the value of the
counter is less than 100. When the value of the While condition returns False, the loop is ex-
ited and a string message 100 counters were set is written to the page.
<%
Dim myCounter
myCounter = 0 ‘=== Initialize variable.
While myCounter < 100 ‘=== Evaluate the condition.
myCounter = myCounter + 1 ‘=== Increment Counter.
Wend ‘=== End While loop when myCounter > 99.
Response.Write myCounter &”counters were set” ‘=== Write counter to page
%>

08.1389-6 CH05 100 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Interrupting the Program Flow 101

N O T E Remember, when evaluating condition statements in VBScript, if a condition is Null, the

condition returns a False. ■

The For Each...Next Statement

The For Each...Next loop is a new feature in VBScript 2.0. The For Each...Next loop repeats
a group of statements for each element in a collection or array. What this means is that you can
cycle through a group of objects, such as text boxes or database fields, and manipulate the
properties of those objects. A collection acts as a storage unit that contains a set of similar
objects. The collection is stored as an array that has a sequential index that identifies the indi-
vidual storage units, called elements. Because the objects are similar, they have the same prop-
erties and can be accessed by code.

This idea of a collection can best be illustrated by comparing collections to an apartment build-
ing. The apartment building serves as a large storage unit that contains the individual apart-
ments. This group of apartments shares common characteristics, such as having an apartment
number and number of bathrooms. If you want to determine the total number of kitchens that
exist in the entire apartment complex, you would use the following statement to examine the
AppartmentComplex collection:

For Each Apartment in AppartmentComplex

Total = Total + Appartment.kitchens
Next

In a code-specific situation, most of your Active Server Page objects use collections to house
information about the object. For example, you can display all the values stored in a cookie file
by using the Request object’s Cookie collection, as demonstrated below.
Part
<%
For Each key in Request.Cookies(“mycookie”)
II
%> Ch
<%= cookie %> (<%= key %>) = <%= Request.Cookies(cookie)(key)%>
<% 5
Next
%>

This example would write all the values stored in the cookie named “mycookie”. For more
information about cookies and the Request object, see Chapter 15, “Retrieving User Informa-
tion: The Request Object.”

Interrupting the Program Flow

You have seen the implementation of decision-based and iterative logic. VBScript also gives you
the ability to explicitly change the navigational flow of your application by using the Exit state-
ment. The Exit statement provides the functionality to jump out of the current script block
back into the host script code. The On Error statement helps control program execution when
a runtime error occurs.

08.1389-6 CH05 101 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 102 Chapter 5 Controlling Program Flow with VBScript

The Exit Statement

In some situations, you need to be able to stop the current executing logic found in a code
procedure. This often happens when a condition is met and additional code processing is un-
necessary. The Exit statement permits this change in programmatic flow from a Do...Loop,
For...Next, function, or subroutine to the calling script block. Exit uses the following syntax:

Exit ProcedureCall

where ProcedureCall can be the Do, For, Sub, or Function keywords.

Here’s how to control the program flow inside a For...Next loop: In order to exit a For...Next
loop, the Exit statement can be placed anywhere inside the calling For...Next loop. Recall the
For...Next syntax:

For counter = start To end [Step stepsize]

statements
Exit For
statements
Next

When the Exit For statement is encountered in the code, VBScript transfers control to the
statement after the calling Next statement. Therefore, if you have nested For...Next loops, as
shown in the example below, the program control will jump out of the embedded For K loop
back into the For J loop.
For J = 1 To 10
Private myLastRecord
For K = 1 To 10
If IsNull(myArray(K)) then ‘=== Tests for Null Values
Exit For ‘=== Exits out of For K loop
Else
myLastRecord = myarray(k)
End if
Next ‘=== Next I loop
Response.Write myLastRecord ‘=== On the Exit event, program
➥returns here
Next ‘=== Next j

The Exit For statement can also be applied to exiting the For Each...Next loop.

The same exiting process that exists for the For...Next loop can be applied to the Do...While
loop. Recall the Do...While loop syntax:
Do {While | Until} condition
statements Exit Do
statements
Loop

The following example expands on the counter loop illustrated in the previous example by
allowing a flag to be set by the inner Do loop to stop the execution of the outer loop. Also, notice
that the inner loop uses the Do...While loop and the outer loop uses a Do...Until loop.

08.1389-6 CH05 102 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 The On Error Statement 103

<%
Dim myCheck, myCounter
myCheck = True: myCounter = 0 ‘=== Initialize variables.
Do ‘=== Create outer do loop.
Do While myCounter < 100 ‘=== Inner loop.
myCounter = myCounter + 1 ‘=== Increment Counter.
If myCounter = 50 Then ‘=== If condition is True.
myCheck = False ‘=== Set value of flag to False.
Exit Do ‘=== Exit inner loop.
End If
Loop
Loop Until myCheck = False ‘=== Exit outer loop immediately.
%>

The Exit statement can also be used to return application flow to the host script that called a
function or subroutine. The Exit function and Exit Sub statements return control to the state-
ment following the calling statement.
Function mySearch(mySearchVar)
‘ VBScript statement ‘=== Sample code
If StrFound = False Then ‘=== Checks a variable
mySearch = False ‘=== Sets the return value of function
Exit Function ‘=== Exit the function
End If
‘ More statements ‘=== Sample code
End Function

The On Error Statement Part

The On Error statement is not actually used to interrupt program flow but to react to program
interruptions as a result of runtime errors. The On Error statement gives you a limited amount II
of error-handling control and uses the following syntax: Ch
On Error Resume Next
5
There are no options available with the On Error statement. The On Error statement simply
enables VBScript runtime error handling. Without the On Error statement, if a runtime error
occurs, the application stops and the error message is displayed to the user. In most situations,
a runtime error is caused by trying to complete an invalid operation. However, the On Error
Resume Next statement continues to process the application by returning control to the state-
ment immediately after the statement where the error was generated.

The following example illustrates using an On Error Resume Next statement. The improper
use of the Month function will produce a Type Mismatch error and would consequently close
the script, if the On Error was not included.
<%
On Error Resume Next ‘=== Turns on Error Handling at this point
Dim myMonth, myMessageStr
myMonth = Month(“Scot”) ‘=== Type Mismatch error is generated
Response.Write myMonth ‘=== Application Focus is returned here
%>

08.1389-6 CH05 103 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 104 Chapter 5 Controlling Program Flow with VBScript

In this example, the program control is returned to the line following the misuse of the Month
function. Because the variable myMonth was not set, nothing is written to the page.

Remember that VBScript, unlike its parent Visual Basic, does not support the GoTo option. The
GoTo functionality gave you the ability to write error-handling packages that could be refer-
enced when an error occurred. The GoTo feature has been omitted from VBScript. With only
inline error handling available in VBScript, trapping your errors becomes more cumbersome
than in Visual Basic. For more information on the Error object, please see the next section.

Error Trapping with the Err Object

The basis of VBScript error handling is made possible by using the Err object. The Err object
is created when a runtime error occurs. The Err object is used to store information about the
error and expose its different methods. The syntax for the Err object is:
Err[.{property | method}]

where property is the Err object properties listed in Table 5.1 and the method is the Err object
methods listed in Table 5.2. In VBScript code, the On Error statement is used to enable error
handling. However, the only error response that exists is to use the Resume Next statement.
The Resume Next statement is used to respond to an error condition by enabling the program
to continue execution with the next statement after the one that produced the error. Therefore,
to enable error handling in VBScript, use the following line of code:
<% On Error Resume Next %>

The On Error statement used by VBScript enables limited functionality when compared to its
native Visual Basic development counterpart. If no error handling is activated, any runtime
error that is generated at the server will be fatal, resulting in the immediate termination of the
script. However, if error handling is activated and you, for example, misspell a function, the
application continues by executing the next line in the code.

Remember that because ASP code is compiled from the top of the page to the bottom of the
page, the placement of the tags is important. Error trapping on the page can only begin after
the On Error Resume Next statement. Any runtime error that occurs before the On Error
Resume Next statement will cause a fatal error and shutdown of the script. This is demon-
strated in the following example where the On Error Resume Next statement is located in the
middle of the script. The improper use of the Month function produces a Type Mismatch error
and consequently closes the script.
<%
Dim myMonth, myMessageStr
myMonth = Month(“Scot”) ‘=== Type Mismatch error is generated

On Error Resume Next ‘=== Turns on Error Handling at this point

Response.Write myMonth ‘=== Application Focus is returned here
%>

Therefore, in most situations when error handling is going to be implemented, it is best to

place the On Error Resume Next statement at the very beginning of the page.

08.1389-6 CH05 104 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Error Trapping with the Err Object 105

N O T E Error handling begins after the On Error Resume Next statement in the script. Any runtime
errors generated before the error handling declaration will produce a fatal runtime error. ■

Remember: Error handling returns the application focus back to the line immediately following
the statement that generated the error. This return functionality is important to consider when
using subroutines and functions. For example, Listing 5.1 calls subroutines to perform a task.
However, code within the subroutine improperly declares the VBScript Round function and
produces a runtime error. Because error handling is initiated at the script level, the program
execution is returned to the following statement in the script-level code, not to the next line
within the GenerateError subroutine procedure.

Listing 5.1 Example of error trapping code.

<% On Error Resume Next %>
<%
‘=============================== Begin Sub Module ==============================
Sub GenerateError()
myVar = Round(“myNumber”,5) ‘=== Incorrect use of Round Function
Response.Write Err.Description ‘=== This line is not executed
End Sub
‘=============================== Close Sub Module ==============================
%>

<Center>Using the Error Statements</center><BR>

<%
Response.Write “Executed First Line <BR>” ‘=== Writes ‘Executed First Line’
➥to page
call GenerateError() ‘=== Error is generated here
Part
Response.Write “Executed Third Line <BR>” ‘=== Program control is returned
here II
%>
Ch

The error handling also has the capability to provide localized error trapping at the procedure 5
level. Listing 5.2 limits error handling only to code within the GenerateError function. This
example checks the return result of the function to determine whether an error was generated
within that function. If an error was generated (the result of GenerateError = True) the error
message is written to the page. If the GenerateError function returns False, a return message
stating that the transaction was successful is written to the page.

Listing 5.2 More error trapping code.

<%
‘================================== Begin Function Module ======================
Function GenerateError()
On Error Resume Next ‘=== Enable error trapping in function

continues

08.1389-6 CH05 105 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 106 Chapter 5 Controlling Program Flow with VBScript

Listing 5.2 Continued

only
myVar = Round(“myNumber”,5) ‘=== Generate error
If Err then
GenerateError = True
Else
GenerateError = False
End if
End Function
‘================================== Close Function Module ======================
%>

<% ‘=================================== Begin Script ===========================

%>
<center>Using the Error Statements </center><BR>
<%
if GenerateError() then ‘=== Check results of
➥Function: Returns True
Response.Write “Warning: The Following Error Has Occurred: <BR>”
Response.Write Err.Description & “<BR>” ‘=== Write the description
of the error
Else ‘=== No errors
Response.Write “Data Transfer Successful<BR>”
End if
%>
<% ‘=================================== End Script =============================
%>

Notice that even though the Err object was created in the GenerateError function, the Err
object has page-level scope and can be referenced anywhere on the page. However, if faulty
code generates a runtime error outside the GenerateError function, the server will process the
script up to the point where the error was generated and create a fatal error, as shown in the
following example.
<%
‘================================== Begin Function Module =======================
Function GenerateError()
On Error Resume Next
myVar = Round(“myNumber”,5) ‘=== Generate an error trying to round a
➥string
If Err then
GenerateError = True
Else
GenerateError = False
End if
End Function
‘================================== Close Function Module =======================
%>

<% ‘=================================== Begin Script ============================

%>
<center>Using the Error Statements </center><BR>
<%

08.1389-6 CH05 106 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 Error Trapping with the Err Object 107

if GenerateError() then ‘=== Check results of

➥Function: Returns True
Response.Write “Warning: The Following Error Has Occurred: <BR>”
Response.Write Err.Description & “<BR>” ‘=== Write the description of
➥the error
Else ‘=== No errors, proceed
Response.Write “Data Transfer Successful<BR>”
myVar = Round(“myNumber”,5) ‘=== Generate fatal error and
➥terminate script
End if
%>
<% ‘=================================== End Script ==============================
%>

In the previous example, the GenerateError function will generate an error trying to round a
string. However, because error trapping is enabled for the function, the script continue process
when program control is returned to the calling script. However, program execution will termi-
nate when a runtime error is generated at script level.

Also, keep in mind that if a property of the Err object is referenced before the Err object is
actually created, no runtime errors are produced when the properties are referenced. The
following example demonstrates referencing properties of the Err object without the explicit
creation of an Err object.
<%
If Not(Err) then ‘=== If no
➥Error Objects exist
Response.Write “Error Number: “ &Err.Number &”<BR>” ‘=== Returns 0
Response.Write “Error Description: “ &Err.Description ‘=== Return
➥zero length string
myErrorType = Vartype(Err.Number) ‘=== Returns Part
➥Long
myErrorType = Vartype(Err.Description) ‘=== Returns
II
➥String Ch
End If

%>
5
The Err Object’s Properties
Active Server Pages uses the error handling found in VBScript to process and manage runtime
errors. VBScript handles error trapping with the Err object. The Err object has several proper-
ties and two methods. The Err object’s properties are listed in Table 5.1.

Table 5.1 Err Object Properties

Property Description
Description Returns a descriptive string associated with an error.
HelpContext Returns a context ID for a topic in a help file.

continues

08.1389-6 CH05 107 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 108 Chapter 5 Controlling Program Flow with VBScript

Table 5.1 Continued

Property Description
HelpFile Returns a fully qualified path to a help file.
Number Returns a numeric value specifying an error (default property).
Source Returns the name of the object or application that originally
generated the error.

The Err object’s Description, Number, and Source properties are used to provide more infor-
mation about the runtime error. The Description property explains the VBScript error that
has occurred. The Number property returns the VBScript-specific error number. This property
is useful because you can retrieve the specific error number and tailor your application accord-
ingly. The Source property returns the application or instantiated object that generated the
runtime error.

The Source property is also important because now you can provide a customized error and
help response system. Although an application error lowers productivity and irritates users, a
customized error and help response system enables a smooth transition from an instantaneous
runtime crash to a customized help-based solution. This help-based solution system can not
only change the user’s perception of your application, but can also help the user by providing
additional debugging information.
You can use the properties of the Err object to access more information from the VBScript help
file and also reference your own help files from your own custom created application. The Err
object’s HelpFile and HelpContext properties provide the pointers to the help files. The
HelpFile property sets the path to the help file and the HelpContext identifies the context ID
used to display the help topic content.

Using the Err Object Methods

The Err object also provides methods to control the Err object. The Err object’s methods, as
shown in Table 5.2, can be used to generate a runtime error or reset the Err object’s properties
using the Clear method.

Table 5.2 The Err Object’s Methods

Method Description
Raise Generates a runtime error.
Clear Clears all property settings of the Err object.

The Err object has two methods that enable you to control the Err object. The Raise method
enables you to generate a runtime error, which is particularly useful in debugging your applica-
tions. The Raise method requires the following syntax:
object.Raise(number, source, description, helpfile, helpContext)

08.1389-6 CH05 108 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 From Here… 109

where object is the Err object, number is a long integer subtype used to uniquely identify the
number, source is a string used to describe the application or object that generated the error,
description is used to explain the raised error, helpfile is the fully qualified path to the help
file, and HelpContext is the pointer to the specific content topic.
<%
Err.Raise 6, “You just simulated an ‘overflow’ error! “ ‘=== Generate an
➥Err Object
Err.Clear ‘=== Clear the Err
%>

You have to watch when you are controlling the programmatic workflow in your application
that you don’t inadvertently reset your Err object. The Err.Clear method is automatically
executed whenever an On Error Resume Next, an Exit Sub, or Exit Function statement is
called.

From Here…
This chapter covered how VBScript controls the programmatic application flow. VBScript con-
trols the process flow by using decision-based logic and looping logic. Decision logic is based
on the If...Then, If...Then...Else, and Select Case statements. These statements are used
to evaluate conditions and make a decision based on the expression returned from the condi-
tional statement. VBScript uses looping logic to repeat a block of code until a condition evalu-
ates to True. VBScript uses the For...Next, Do...Loop, While...Wend, and the For
Each...Next statements to complete the cyclical processing. In addition, VBScript also pro-
vides a mechanism to stop the current processing flow by using the Exit statement. Finally, the
Err object was illustrated to examine how to handle error trapping with VBScript. Part

For related information, please see the following chapters: II

■ Chapter 6, “Working with Functions, Subroutines, and Dialog Boxes,” illustrates event- Ch
driven programming through the use of reusable code modules. 5
■ Chapter 7, “Datatype Conversion Features,” features the data type conversion functions
in VBScript to ensure proper application execution.
■ Chapter 13, “Managing the User Session: The Session Object,” demonstrates how to
track user-specific information through your Web application.
■ Chapter 14, “Speaking to the Web Client: The Response Collection,” describes how to
transfer the server-side information to the Web client using the Response collection.
■ Chapter 19, “Putting It All Together: Creating an Online Registration System,” illustrates
how to use the different ActiveX components that ship with the Internet Information
Server to create a Web site for the Amazing DataWidget WareHouse online catalog.

08.1389-6 CH05 109 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 110 Chapter 5 Controlling Program Flow with VBScript

08.1389-6 CH05 110 10/26/98, 9:09 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH05 lp#4
 6
111

C H A P T E R

Introduction to Event-Driven
Working with Programming
This section provides an overview of

Functions, Subroutines, event-driven programming and how

it applies to VBScript.

and Dialog Boxes Using Procedure Modules

Learn how to utilize functions and
subroutines to build re-useable code
and provide logical application flow
using procedures.

Client-Side Dialog Boxes

Discover how to utilize client-side
VBScript was initially developed by Microsoft to provide dialog boxes to display and retrieve
event-driven programming functionality to Web pages for information from the user.
users of the Internet Explorer browser. Event-driven pro-
gramming is the process of executing code that depends
on the actions that occur in an application. This custom-
ized programming approach is different from the tradi-
tional model of procedural programming, in which an
application was designed to complete a specific objective.
The traditional procedural approach relies on the applica-
tion to execute code, whereas event-driven code relies on Part
user input to execute code. This chapter will cover how II
VBScript uses procedures to create smaller modules of
code that are executed by a user’s actions. In addition, the Ch
Message and Inbox dialog boxes will demonstrate how to 6
provide user interaction on the Web client. ■

09.1389-6 CH06 111 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 112 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

Introducing Event-Driven Programming

In most event-driven application development environments, such as Visual Basic or Visual
C++, objects have specific events or actions that can be controlled or monitored. These actions,
such as clicking a button, can then be used to determine the code that is to execute next. Fig-
ure 6.1 demonstrates an example of event-driven programming. This example shows how
VBScript displays different pop-up menu boxes depending on the button that is pushed. If the
button labeled “Say Hello” is pushed, the OnClick event of the pressed button is executed. If
script code exists to be executed on the OnClick event, then that code would be executed. In
this situation, a message box with “Hello” is displayed. However, if the second button labeled
“Say Goodbye” is pushed, the OnClick event of the button labeled “Say Goodbye” is processed.

FIG. 6.1
Event-driven program-
ming reacts to user
events.

As mentioned in Chapter 3, VBScript code is actually text that is embedded as a scripting lan-
guage within the HTML page. The embedded code exists within the HTML page, and the
SCRIPT tag tells the browser what language is being used. The HTML syntax for the SCRIPT tag
is the following:
<SCRIPT LANGUAGE=’scripting language’>
Script Application logic here
</SCRIPT>

where ‘scripting language’ is any script-compliant language.

The language tag is used to determine the interpreter that processes the script. The HTML-
embedded script is actually interpreted as the page is initially loaded into the browser. The pre-
loading process enables objects to be verified as secure or unsecure before they are loaded
into the browser, as mentioned in Chapter 3.

Once the text has been loaded into the browser, HTML intrinsic objects are instantiated by the
browser. The term instantiated simply means that the browser has created the object and re-
served a named memory space where the object can be referenced. These instantiated objects
have a scope equal to the length of time that the page is loaded in memory of the browser. For
example, the following code creates an HTML button named btnHello.
<INPUT TYPE=BUTTON VALUE=”Say Hello” NAME=”btnHello”>

09.1389-6 CH06 112 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Introducing Event-Driven Programming 113

Once an object is instantiated, embedded script languages can access the object’s properties,
methods, and events. Most event-driven programming languages access an object’s events,
properties, or methods by using the following syntax:
Object.[Events, Properties, Methods]

Utilizing, managing, and coordinating an object’s events, properties, and methods form the
basis of event-driven programming. By separating the characteristics of an object into these
three categories, the script language can trap or react to an object’s properties, methods, and
especially its events.

An event is a predefined action of an object. VBScript can then react to the object’s action by
executing a code module that is associated with that event. For example, the intrinsic HTML
button has an OnClick event. This OnClick event is triggered when the button is pressed.
These predefined events are usually named by concatenating the name of the object with an
underscore and the name of the event: name_event. Therefore, when the HTML button is
clicked, the browser interprets an event named btnHello_OnClick. In this situation, you place
code in this event procedure to display a message to the user via a message box. The following
code illustrates utilizing the OnClick event of a button.
<SCRIPT LANGUAGE=”VBScript”>
Sub BtnHello_OnClick
MsgBox “Hello!”, 0, “VBScript Demo”
End Sub
</SCRIPT>

Even though trapping an object’s events and executing code are the foundations of event-
driven programming, being able to access an object’s properties and methods is also impor-
tant. An object’s property is used to describe a characteristic of that object. For example, the
HTML button described above has a Name property set to ‘btnHello’ and Value property set
to ‘Say Hello’. You can display the name of a control by accessing its Name property. Most
objects have a minimum of two properties: a Name property and a Value property. The following
example demonstrates how to display information that has been entered into a text field when
the OnClick event of a button is executed:
<SCRIPT LANGUAGE=”VBScript”>
Part
<!--
Sub btnDisplayText()
II
msg = frmtest.txtInformation.value Ch
call window.alert(msg)
End Sub 6
-->
</SCRIPT>

You can further manipulate the object via its methods. A method call is used to cause an action
to occur on a particular object. For instance, in the previous example, the window.alert(msg)
statement executes the pop-up dialog box from the window object.

09.1389-6 CH06 113 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 114 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

Using Procedures: Subroutines and Functions

Now that the basics of event-driven code have been discussed, it is time to investigate how
VBScript stores the code that is executed on an object’s events. One of the greatest benefits of
event-driven programming is the capability to store code that is bundled together in script
procedures. These procedures can be used on both client and server-side scripts and they
provide the most flexible way to re-use code and provide a clear and logical workflow.

Procedures in VBScript are either subroutines or functions. Functions are procedures that
return a value to the calling code module. Subroutines are procedures that do not return a
specific value; they simply execute code. Subroutines are used to act as a self-contained section
of code used to process information. Functions provide interaction with the calling script code
by not only processing logic, but also providing results back to the script.

General Procedure Rules

Although subroutines and functions differ in terms of returning values back to the calling
script, they share many common rules. Most of the similar rules for procedures are found in
the declaration, variable use, and naming conventions.

Procedure Declaration Because functions and subroutines are specific categories of script
procedures, the two share many of the same rules and logic. VBScript calls for both functions
and subroutines to use the following syntax:
Public | Private procedure name arglist
statements
Exit Procedure
statements
End Sub

where procedure identifies a function or subroutine, name identifies the procedure name, and
statements contains the code to be executed when the procedure is referenced. The name of
the procedure must follow the same naming rules that are applicable to variables:

■ Procedure names must begin with an alphabetic character.

■ They can’t contain an embedded period or type-declaration character.
■ Procedure names must be unique within the same scope.
■ They cannot be longer than 255 characters.

The subprocedure has two optional arguments—the Public/Private declaration and the argu-
ment list.
Public Versus Private Declarations Procedures, similar to variables, can be made public or
private within a page (either HTML or ASP). By declaring a procedure as private, you limit the
accessibility of that procedure to only those scripts where the procedure is declared. When a
procedure is declared as public, any script within that page can reference the procedure. If no
explicit declaration of public or private is made, the procedure is assumed to be a public proce-
dure.

09.1389-6 CH06 114 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Procedures: Subroutines and Functions 115

N O T E Variables declared in procedures have scope only within that procedure and are not
preserved when the application control is passed between procedures. ■

Argument Lists The second optional argument is the argument list. The argument list de-
clares the variables defined by the procedure that are assigned by the calling code. Placing a
variable in the argument list is the equivalent of implicitly declaring a variable within a proce-
dure. If you try to re-declare the variable using a Dim or equivalent statement, you will generate
a ‘Named Redefined’ error.

When the calling code calls the procedure, the variables in the argument list are first implicitly
instantiated. Then the calling values are transferred to the new localized variable located in the
procedure’s argument list. Creating your own localized variables enables you to create a de-
scriptive naming convention that facilitates easy code re-use and logical operation. The follow-
ing example demonstrates how procedures use their own localized variable names.
<%
Sub CheckUserId(UserAction, UserIssue) ‘=== Begin Subroutine
Dim IdValid ‘=== Create local variable
IdValid=(UserAction &” “ &UserIssue) ‘=== Uses Local Sub values
Response.Write “<BR>” &IdValid ‘=== Writes “Scan P34—Table” to page
Response.Write “<BR>” &MyClientAction ‘=== Writes “Scan” from calling code
End Sub ‘=== End Subroutine
%>

<%
...
Dim myClientAction, myClientIssue
myClientAction=”Scan”: myClientIssue=”P34—Table” ‘=== Initialize Variables
Call CheckUserId(myClientAction, myClientIssue) ‘=== Calling VBScript code
Response.Write “<BR>” &Idvalid ‘=== Writes a blank line
Response.Write “<BR>” &UserAction ‘=== Writes a blank line

...
%>

The client code calls a subroutine called CheckUserId and supplies two variables named
myClientAction and myClientIssue. The values of these variables are then transferred to the Part
local subroutine variables named UserAction and UserIssue. The subroutine creates these II
procedural scope variables and writes the strings ‘Scan’ and ‘P34-Table’ to the HTML page.
Ch
After the last statement of the subroutine is executed, program control is returned to the line
immediately after the calling of the subroutine. At this point, the application tries to write the 6
subroutine’s local variable, ‘IdValid’, and the subroutine’s declaration variable, UserAction,
to the page. However, because those variables have a scope limited to that subroutine, no val-
ues are actually written to the page.

09.1389-6 CH06 115 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 116 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

TIP Placing a variable in the argument list is the equivalent of implicitly declaring a variable within a
procedure.

Subroutines and functions do not require an argument list. An argument list simply acts as a
mechanism to transfer data from the calling client code into the host-based procedure. How-
ever, if multiple variables exist, they must be separated by a comma. Furthermore, unless
public variables are created outside procedures, variables processed by the procedure must
enter through the argument list. The values of a local argument list are not preserved from
procedure to procedure.

The arglist argument has the following syntax and parts:

ByVal | ByRef varname( )

where ByVal declares that the variable is passed by value, ByRef declares that the variable is
passed by reference, and varname is the name of the argument variable.

Why is this important? Passing a variable by reference allows the variable’s value to be perma-
nently changed within the procedure. Variables passed by value allow a temporary copy of the
variable to be made within the procedure. This temporary variable can be manipulated at will
by the procedure level script. However, when the program execution is returned to the calling
code, the temporary copy of the variable is destroyed and the original value is restored.

In most situations, you don’t expect a procedure to permanently modify its arguments. Careful
attention must be given when passing arguments ByRef or ByVal to prevent the corruption of
your data. For example, assume that you are writing code that is used to manipulate account
balance information.
<%
sub CalcNewBalanceVal(ByRef Balance)
Balance=(FormatCurrency(Balance+Balance*22.5/100/12))
End sub

sub StoreBalance(AcctNumber, AcctBalance)

Response.Write “<BR> Your Balance is “ &AcctBalance
End sub

dim myAcctBalance
myAcctBalance = 3250
Call CalcNewBalanceVal(myAcctBalance)
...
[some time later in code]
...
Call StoreBalance(myAcctNumber,myAcctBalance) ‘=== Stores $3250.00
%>

In this example, you use two subroutines to assist your calculations. One procedure calculates
a balance and the other stores the information. The first procedure demonstrates using the
ByVal function to pass the Balance variable into the subroutine for processing. Afterward, this
value is called by the Store Balance procedure to store the value of Balance into a database
with the value of $3250.

09.1389-6 CH06 116 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Procedures: Subroutines and Functions 117

However, passing the value by reference can cause a variable to retain its changed value else-
where in code:
<%Sub CalcNewBalanceRef(ByRef Balance)
Balance=(FormatCurrency(Balance+Balance*22.5/100/12))
End sub

sub StoreBalance(AcctNumber, AcctBalance)

Response.Write “<BR> Your Balance is “ &AcctBalance
End sub

dim myAcctBalance
myAcctBalance = 3250
Call CalcNewBalanceRef(myAcctBalance)
Call StoreBalance(myAcctNumber,myAcctBalance) ‘=== Stores $3310.94
%>

Notice what happens when the Balance variable is passed by reference in the
CalcNewBalanceRef subroutine. The variable Balance maintains its new value even after the
programming focus has returned to the calling script. When that variable is later called to be
stored, the incorrect value is used because the variable value is permanently changed.

Nesting Procedures If you want to permanently change the value of variable, the proper way
to accomplish this is to use a function that returns a value to a variable, which ensures a clear
way to track and manage variables in your code.

Further, similar to the logical operators discussed in Chapter 5, procedures (both subroutines
and functions) can be nested within each other. In the example above, if storing the new bal-
ance information always occurs after the calculation of the new balance, you might want to
consider calling the StoreBalance routine from inside the CalcNewBalanceRoutine. The follow-
ing code list demonstrates the capability to nest procedures:
<%
Sub CalcNewBalanceRef(ByRef Balance)
Balance=(FormatCurrency(Balance+Balance*22.5/100/12))
Call StoreBalance(Balance)
End Sub
Part
Sub StoreBalance(AcctBalance)
Response.Write “<BR> Your Balance is “ &AcctBalance
II
End Sub Ch
%>
6
<%
Dim myAcctBalance
myAcctBalance = 3250
Call CalcNewBalanceRef(myAcctBalance)
%>

As you can see, the nesting of procedures makes the calling client code much easier to read.
However, this recursion technique can lead to problems if it is not properly managed. Problems
arise because procedures can endlessly call themselves until stack overflow occurs.

09.1389-6 CH06 117 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 118 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

Exit Statements If you need to exit a procedure before the End Sub or End Function state-
ments, use the Exit statement. The Exit statement is used to immediately exit from a subrou-
tine or function procedure, and it is typically used when a condition is met and does not require
additional code processing within the procedure. The program execution continues with the
statement following the statement that called the subroutine. Furthermore, you are not limited
to just one Exit statement in your procedure. Multiple exit conditions can exist anywhere in
the procedure. However, in most situations, you might want to create smaller subroutines and
functions to further compartmentalize programming flexibility. Using smaller procedure mod-
ules in lieu of multiple Exit statements helps prevent unstructured logic errors and erratic
application flow.

Subroutines
In the previous section, the general topics governing the rules of using procedures (subrou-
tines and functions) were discussed. In general, most rules apply to both subroutines and
functions, but some individual differences exist. Remember that subroutines are used to store
a collection of code that is used to perform a commonly used task. Subroutines require the
following syntax:
Public | Private Sub name (arglist)
statements
Exit Sub
statements
End Sub

Subroutines are called or executed in one of two ways. The first method explicitly calls the
subroutines using the Call statement. Call adheres to the following syntax:
Call subname argumentlist

where subname is the subroutine name.

The argumentlist is an optional value that is used to reference the variables, if any, that are
needed by the subroutine. If you use the Call statement to declare your subroutine and the
subroutine uses variables, the variables must be enclosed in parentheses.

The second manner of calling the subroutine is to implicitly call the subroutine. This type of
activation takes place by simply referencing the name of the subroutine and its arguments. If
arguments are required, parentheses are not required in the calling code. For instance, notice
how the example below activates the same subroutine using the explicit Call statement and
the implicit execution by using the WriteInfoMsg subroutine name:
<%
Sub WriteInfoMsg(StrTitle, MsgType)
Response.Write “<BR>” &MsgType &StrTitle &”<BR>”
End Sub
%>

09.1389-6 CH06 118 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Procedures: Subroutines and Functions 119

<%
Dim StrCompany
StrCompany = “BrainWare Inc”
WriteInfoMsg StrCompany, “Welcome to “ ‘=== Implicit subroutine
➥call
Call WriteInfoMsg(StrCompany, “Thanks for visiting “) ‘=== Explicit Subroutine
➥Call
%>

TIP Use parentheses when using the Call statement to explicitly execute a subroutine with arguments. No
parentheses are required when implicitly executing a subroutine.

Functions
Functions, on the other hand, are used to accept arguments, process these arguments, and
return the results to the calling code. The Function statement has the following syntax:
Public | Private Function name (arglist)
statements
name = expression
Exit Function
statements
name = expression
End Function

The foundation of the Function procedure is built on the grouped processing functionality
found in the subroutine procedure. The Function procedure is a separate procedure that ac-
cepts arguments to perform a series of code statements and returns the results of the pro-
cessed code statements directly to the client code. The only way to accomplish this return of
processed information using Subroutines is to manipulate a variable that has global scope
inside the procedure. The Function procedure enables you to immediately have access to the
manipulated data. This instantaneous return of processed data is performed the same way as
other native VBScript function, such as Sqr, Cos, or Chr.

For instance, the following example uses a Function to return the calculated balance:
Part
<%
Function CalcNewBalance(Balance) II
Balance=(FormatCurrency(Balance+Balance*22.5/100/12))
Ch
CalcNewBalance = Balance
End Function
6
Dim myOldAcctBalance, myNewAcctBalance
myOldAcctBalance = 3250
myNewAcctBalance = CalcNewBalance(myOldAcctBalance)
Response.Write “<BR> Your Balance is “ & myNewAcctBalance
%>

In this example, the CalcNewBalance function returns the new calculated balance, which is
assigned to the variable myNewAcctBalance.

09.1389-6 CH06 119 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 120 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

Using Dialog Boxes to Interact with Users

Dialog boxes are used to present information to and request information from the user within
the client’s browser. These pre-built, modal dialog boxes are only instantiated on the Web client
to perform client-side interaction. Active Server Pages, because they are executed on the
server, have no need for pop-up dialog boxes to occur on the server. If you try to use a server-
side dialog box, you will generate a Permission Denied error. However, you can use return
values from the dialog boxes to solicit information from the user and transfer that data to your
Active Server Page. The message dialog box is used to present a message box to the user and
trap the results. The input box is used to gather textual information from the user.

Message Box
The message box is used to present warnings, errors, alerts, or information to the user. In
addition, the message box gives you the flexibility to determine what button the user has se-
lected. These available options enable you to control programming workflow depending on
which buttons were pressed. To use the message box, use the following syntax:
MsgBox(prompt, buttons, title, helpfile, context)

where

■ prompt is the required message string

■ buttons are the available message box buttons
■ title is the text that appears in the title bar of the message box
■ helpfile is the name of the helpfile used to provide contextual help
■ context is the numerical identifier used to display the topic information from the
Helpfile

The message box has several optional parameters that enable you to create appropriate
prompting mechanisms needed for your applications. The capability to customize the function-
ality of the message box dialog occurs through the button option of the message box. The
button option enables you to control three visual components of the message box:

■ The type of buttons

■ The icon style of the button
■ The default button

VBScript’s message box provides six different prompting mechanisms to display information,
warnings, or confirmations to the user. The default prompt is the standard OK dialog box. To
change which buttons are displayed on the message box, use the constants listed in Table 6.1.

09.1389-6 CH06 120 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Dialog Boxes to Interact with Users 121

Table 6.1 Message Box Buttons

Constant Value Description
vbOKOnly 0 Display OK button only. (Default.)
vbOKCancel 1 Display OK and Cancel buttons.
vbAbortRetryIgnore 2 Display Abort, Retry, and Ignore buttons.
vbYesNoCancel 3 Display Yes, No, and Cancel buttons.
vbYesNo 4 Display Yes and No buttons.
vbRetryCancel 5 Display Retry and Cancel buttons.

The default message box is simply used to present a message to the user and present only the
OK button on the message box, as seen in Figure 6.2.

FIG. 6.2
Use the default
VBScript message box
when you want the user
to acknowledge a
request.

The vbOKOnly message box is generated by the following code: Part

Sub loadOK()
dim strMsg, iButton, strTitle, index
II
strMsg = “Please Enter a Login ID!” ‘=== set message string Ch
iButton = 0 ‘=== default: vbOkOnly
strTitle = “VBScript Default Message Box” ‘=== set title bat text 6
index = MsgBox(strMsg, iButton, strTitle) ‘=== trap the results
End sub

In most decision-making applications, you have to prompt the user with one of two options,
either Continue or Stop the process or Continue, Stop, or Cancel the process. For example,
Figure 6.3 illustrates the use of the button constant vbYesNoCancel to present the user the Yes,
No, and Cancel buttons on the message box.

09.1389-6 CH06 121 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 122 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

FIG. 6.3
Use VBScript’s dialog
constants to display the
Yes, No, and Cancel
dialog box.

The button options also allow you to control the displayed icon in the message box. This ability
to choose the icon gives you the opportunity to express the tone of the message. For example,
Figure 6.4 illustrates displaying the Critical Message icon.

FIG. 6.4
Dialog boxes can
display different icons to
express the tone of
message.

Notice the use of the carriage return and line feed characters in the following code, which
drives Figure 6.4. The carriage return, chr(13), and line feed, chr(10), characters are used to
help control how the information is displayed in the message box.
Sub loadOKCancel()
dim strMsg, iButton, strTitle, index
strMsg = “Permission Denied!” &(Chr(13) & Chr(10)) & “Continue?” ‘=== Set
➥the message string
iButton = vbOkCancel + vbCritical ‘=== OK and Cancel, critical icon
strTitle = “VBScript OK Cancel Message Box” ‘=== Set the title bar text
index = MsgBox(strMsg, iButton, strTitle) ‘=== Trap the results
End Sub

VBScript supports four different icons that can be displayed in dialog boxes. The available
constants and values are listed in Table 6.2.

09.1389-6 CH06 122 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Dialog Boxes to Interact with Users 123

Table 6.2 The Dialog Boxes Icons

Constant Value Description Icon

vbCritical 16 Displays Critical Message icon.

vbQuestion 32 Displays Warning Query icon.

vbExclamation 48 Displays Warning Message icon.

vbInformation 64 Displays Information Message icon.

The third option the button tag controls is the selection of the default button on the messaging
window. The default button is used to automatically set the focus on a specific button. Because
the application focus is set onto a button, code behind the default button is executed when the
user hits the Enter key. For example, Figure 6.5 depicts setting the default button to the Retry
button of an Abort, Retry, or Ignore dialog box.

FIG. 6.5
Use the button
property to set the
default button option of
the dialog box.

Part
II
Ch

6
The capability to set the default focus to a particular button is based on the keywords listed in
Table 6.3.

Table 6.3 The Default Button Options in VBScript’s Dialog Boxes

Constant Value Description
vbDefaultButton1 0 First button is default.
vbDefaultButton2 256 Second button is default.

continues

09.1389-6 CH06 123 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 124 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

Table 6.3 Continued

Constant Value Description
vbDefaultButton3 512 Third button is default.
vbDefaultButton4 768 Fourth button is default.

The code behind Figure 6.5 is a combination of all the available functions of the button option.
This code, shown below, is using both the VBScript dialog constants as well as the constants’
values. The three-option dialog that generates the Abort, Ignore, and Retry option buttons are
created using the constant vbAbortRetryIgnor or its value 2. The value 32, or the constant
vbQuestion, displays the question mark icon. The constant vbDefaultButton2 or its value of
256 sets focus to the second button on the dialog window.

Sub loadAbortRetryIgnor()
dim strMsg, iButton, strTitle, index
strMsg = “Could Not Access Database” ‘=== Set the message string
‘iButton = vbAbortRetryIgnore +vbQuestion +vbDefaultButton2 ‘=== Using
➥Constants
iButton = 2 + 32 + 256 ‘=== Using values
strTitle = “VBScript Abort, Ignore and Retry Message Box” ‘=== Set the
➥title bar text
index = MsgBox(strMsg, iButton, strTitle) ‘=== Trap the results
End sub

After the user has selected from an available action, the core functionality of the message box
is to retrieve the options that the user has selected. To obtain the results the user has selected
from the dialog, keep in mind that the dialog boxes act as a prepackaged function within
VBScript. To determine the option that the user has selected, simply assign the results of the
dialog box to a variable, as shown below. Notice that this same assignment methodology is
used to retrieve results from a function.
index = MsgBox(strMsg, iButton, strTitle)

Once this value is determined, use the Message Box result constant, listed in Table 6.4, to
determine the results.

Table 6.4 Message Box Result Constants

Constant Value Description
VbOK 1 OK
VbCancel 2 Cancel
VbAbort 3 Abort
VbRetry 4 Retry
VbIgnore 5 Ignore
VbYes 6 Yes
VbNo 7 No

09.1389-6 CH06 124 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 Using Dialog Boxes to Interact with Users 125

Input Box
Unlike the limited button set returned by the message box, the input box is designed to accept
textual input from the user. The input box dialog displays a prompt for the user to enter data
and click a button, and returns the text entered into the text box when the user selects the OK
button. The InputBox uses the following syntax:
InputBox(prompt[, title], default, xpos, ypos, helpfile, context)

where

■ prompt is the required string expression used to display a question to the user
■ title is the text to appear in the title bar of the input window
■ default is the default text that will appear in the text box
■ xpos and ypos represent the position on the screen where the box is displayed
■ helpfile refers to the name of the help file for the application
■ context refers to the numeric ID needed to retrieve context information

N O T E When both the helpfile and context options are used, a Help button is automatically
added to the dialog box. ■

The InputBox dialog returns values in two situations: when the user presses OK or when the
user selects the Cancel button. When the user clicks OK or presses Enter, the InputBox func-
tion returns whatever is in the text box. If the user clicks Cancel, the function returns a zero-
length string (“”). The following code sample illustrates how to display the input dialog box
shown in Figure 6.6.
Sub LoadInputBox()
dim Index, strMsg, strTitle, strDefualt
strMsg = “Welcome to BrainWare International!”
strMsg = strMsg &(Chr(13) & Chr(10))
strMsg = strMsg &(Chr(13) & Chr(10))
strMsg = strMsg &”Where did you hear about this site?”
strTitle = “Request Information” Part
strDefault = “AltaVista Search Engine”
Index = InputBox(strMsg, strTitle, strDefault) II
End Sub
Ch

FIG. 6.6 6
Use the InputBox to
collect textual data
from the user.

09.1389-6 CH06 125 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 126 Chapter 6 Working with Functions, Subroutines, and Dialog Boxes

This code uses the variable strMsg and the concatenation operator to sequentially build the
prompt string. The text that appears at the top of the InputBox is set by the strTitle variable
and the strDefault variable is used to initially populate the text box.

From Here…
In this chapter, you have learned how VBScript uses procedures to help create logical blocks of
code. These logical units of code can be used to modulize your application into smaller pieces
of code, which enables you to reuse the same code procedures in multiple places in script, as
well as make your code easier to follow and debug.

Functions are used to process information and return the results back to the calling script.
Subroutines are used to process a block of code without returning information to the script.
The Function and Subroutine procedures can accept values to be processed within the code
module. These arguments can be passed into the procedures by either ByRef or ByVal. The
ByVal declaration passes the variable by its value, whereas ByRef declares that the variable is
passed by reference. Keep in mind that passing variables by reference will cause the variable’s
value to permanently change beyond the scope of the procedure. Whereas, using the ByVal
declaration limits the scope of the changed variable value to the procedure.

This chapter also identified VBScript’s dialog box functions. The message box is used to dis-
play information to the user and require the user to respond. The message box can use a vari-
ety of constants to customize the display mechanism presented to the user. The Input box is
also a VBScript function that is used to solicit information from the user.

At this point, you should be well poised to start using the multitude of built-in VBScript func-
tions that are used to manipulate the Variant subtypes. After that, the Microsoft Script
Debugger will be demonstrated to illustrate how to use inline debugging capabilities for your
client- and server-side scripts.

■ Chapter 7, “Datatype Conversion Features,” illustrates the subtype conversion functions.

■ Chapter 8, “Putting It All Together with VBScript,” solidifies Part I by demonstrating
VBScript running on the Web client and IIS server.
■ Chapter 10, “The Composition of an Active Server Application,” lays the foundation for
building the Active Server Pages application.
■ Chapter 15, “Retrieving User Information: The Request Object,” demonstrates how to
implement Active Server Pages and VBScript to retrieve information from the user.

09.1389-6 CH06 126 10/26/98, 9:10 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH06 lp#1
 7
127

C H A P T E R

Identifying the Variant Sub

Datatype Conversion Datatype
Learn how to prevent unexpected

Features logical and programmatic errors by

identifying the sub datatypes.

Converting the Datatype

Utilize subtype conversion features
to force the generic Variant datatype
to specific sub datatypes.

In the previous three chapters, the various effects from the Using Formatting Functions
Variant datatypes were discussed. With operators that Discover how to utilize the VBScript
change behavior as the sub datatype changes for a vari- 2.0 formatting features to take
able, VBScript has various sub datatype identification advantage of built-in date/time,
currency, percent, and number
tools to maximize and convert the specific sub datatypes.
formatting displays.
Furthermore, VBScript 2.0 ships with formatting func-
tions to help control how date/time, currency, percent,
and numbers are displayed. Together, these features
combine to identify, manage, and display variable
information. ■

Identifying the Datatype

VBScript uses identification functions to determine a
specific sub datatype. The identifier functions return a
Boolean value indicating whether the given datatype is
determined. This Boolean value can be used to provide
the logical decision-making points in your application.
Part
II
Ch

10.1389-6 CH07 127 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 128 Chapter 7 Datatype Conversion Features

The IsArray Function

The IsArray function is used to determine whether a Variant is an array. An array is a series
of variables that are stored as one variable, with each variable uniquely identified by an index
number. The value of the variable can be determined by referencing the index number. The
IsArray function requires the following syntax:

IsArray(myVarName)

where myVarName is any valid variable name.

The IsArray function returns True if the expression is an array and False if the expression is
not an array. The following example illustrates the successful results of the IsArray function.
<%
Dim myRray(), myArrayResult ‘=== Declares dynamic array
Redim myRray(1) ‘=== Re declares the array (needed to
➥populate w/data)
myRray(0)=”myData” ‘=== Loads first element
MyArrayResult = IsArray(myRray) ‘=== Returns True
%>

Arrays are particularly useful because they can be treated as a variable. This variable can be
passed from page to page to act as an individual storage unit to house data. The following ex-
ample illustrates testing for arrays using the IsArray function.
<%
Dim myArray, myResult
myArray = Array(ID, Password, Permissions) ‘=== Loads Array with the new
➥Array function
myResult = IsArray(myArray) ‘=== Returns True
myResult = IsArray(myResult) ‘=== Returns False
%>

In this example, the new VBScript 2.0 IsArray function is used to create an array and store the
array as a Variant variable. The first use of the IsArray function returns a True value because
the variable myArray is actually storing an array. However, the second result returns False
because the variable myResults is a Boolean value.

The IsDate Function

The IsDate function examines an expression and determines whether the expression can be
converted into a date. The expression can already be a date expression or can be a string ex-
pression that can be interpreted by the Variant datatype as a date expression. The IsDate uses
the following syntax:
IsDate(myExpression)

The IsDate function returns True if the expression can be converted to a date and False if the
expression is unable to convert the expression. For example, the following demonstrate the
results of IsDate function and various date formats.

10.1389-6 CH07 128 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 Identifying the Datatype 129

<%
dim myDateLong, myDateLiteral, myDateShort, myResults

myDateLong = “APRIL 30, 1971”

myDateLiteral = #4/30/71#
myDateShort = “APR 30, 71”
myDateDash = “04-30-71”
myBadDate = “Not a Date”

myResults = IsDate(myDateLong) ‘=== Returns True.

myResults = IsDate(myDateLiteral) ‘=== Returns True.
myResults = IsDate(myDateShort) ‘=== Returns True.
myResults = IsDate(myDateDash) ‘=== Returns True.
myResults = IsDate(myBadDate) ‘=== Returns False.
%>

The IsEmpty Function

The IsEmpty function is used to determine whether a value has been assigned since the vari-
able was created. When a variable is first created, it is assigned a special value called the Empty
value. This Empty value is different from a zero length string (“”) or a NULL value. The Empty
value is determined by using the IsEmpty function with the following syntax:
IsEmpty(myExpression)

where myExpression is a valid expression.

The IsEmpty expression returns True when the expression is an Empty value. Not only can an
expression exist as an Empty value before it is assigned, but a variable can also be explicitly set
to Empty. For example, in the following code sample, the value of myUserId is declared and
tested for the Empty value.
<%
Dim myUserId

If IsEmpty(myUserId) Then '=== Returns True, value still unassigned

MyUserID = “Administrator” ‘=== Sets the value
Response.Write MyUserId ‘=== Displays the value to page
End if

If Not IsEmpty(myUserID) Then

myUserID = Empty ‘=== Sets to Empty Data Type
Response.Write (IsEmpty(MyUserId)) ‘=== Returns True
End If
%>
Part

The IsNull Function II

One of the biggest sources of confusion over datatypes comes from the lack of understanding Ch
of the NULL value. The NULL value, like the Empty value, is treated as a special type of value in
the datatypes. The NULL datatype represents the anti-datatype, meaning it represents an expres-
7
sion that does not contain an existing datatype. As mentioned previously, the NULL is not the

10.1389-6 CH07 129 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 130 Chapter 7 Datatype Conversion Features

same as the Empty or zero length strings. The IsNull function returns a Boolean value indicat-
ing whether the evaluated expression contains a NULL value. The IsNull function requires the
following syntax:
IsNull(myExpression)

where myExpression is a valid expression.

The IsNull function returns True if the expression contains a NULL. If the expression contains
any other datatype, the IsNull function returns False.

The IsNull function is important for evaluating an expression because of the unique behaviors
of the NULL datatype. The NULL datatype, when used with any expression or any operator, repli-
cates itself to produce a NULL value. The following example illustrates three results of the
IsNull function.

<%
Dim myNullVar, MyNullResults
MyNullResults = IsNull(myNullVar) ‘=== Returns False, variable is empty

myNullVar = “”
MyNullResults= IsNull(myNullVar) ‘=== Returns False, variable is zero length

myNullVar = Null
MyNullResults = IsNull(myNullVar) ‘=== Returns True, variable is Null
%>

If you are testing for NULL, you must use the IsNull function. The IsNull function is necessary
because of the residual effects of NULL. Any existence of NULL in an expression or operator will
transform the expression to NULL itself. Therefore, if any comparison operator is used to test for
NULL, the entire expression itself becomes NULL, resulting in the perpetual failure of the expres-
sion. This perpetual failure is seen in the following example in which an expression is tested to
see if the variable is equal and then not equal to NULL.
<%
Dim MyNullVar
myNullVar = Null ‘=== Set variable to Null

If myNullVar = Null Then ‘=== Test for Null using an

➥operator
response.write “Never Will Occur” & “<BR>” ‘=== This will never execute
Else
response.write “Equal to Null Comparison Failed” & “<BR>”
End If

If myNullVar <> Null Then ‘=== Test for Not

➥Null condition
Response.Write “Never Will Happen Either” & “<BR>” ‘=== This will never
➥execute
Else
Response.Write “Not Equal to Null Comparison Failed” & “<BR>”
End if
%>

10.1389-6 CH07 130 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 Identifying the Datatype 131

The IsNumeric Function

The IsNumeric function determines whether an expression can be evaluated as a number. You
will notice that the IsNumeric function has similar functionality to the IsDate function by evalu-
ating the possibility of converting a specific datatype. The IsNumeric function uses the follow-
ing syntax:
IsNumeric(myExpression)

where myExpression is a valid expression.

The IsNumeric function returns True if the entire expression can be converted to a numeric
expression. If the expression contains any other datatypes, the IsNumeric expression returns
False, as shown in the following example:

<%
Dim myNumericVar, myNumResults

myNumericVar = “53”
myNumResults = IsNumeric(myNumericVar) ‘=== Returns True.

myNumericVar = “459.95”
myNumResults = IsNumeric(myNumericVar) ‘=== Returns True.

myNumericVar = “45 Help”

myNumResults = IsNumeric(myNumericVar) ‘=== Returns False.

myNumericVar = “$45.65”
myNumResults = IsNumeric(myNumericVar) ‘=== Returns True.

myNumericVar = “April 30, 1971”

myNumResults = IsNumeric(myNumericVar) ‘=== Returns False.
%>

The IsObject Function

The IsObject function indicates whether or not a variable is an ActiveX object, and requires
the following syntax:
IsObject(myexpression)

where myExpression is a valid ActiveX automation object. An ActiveX automation object is an

object that exposes its own properties and methods through COM interfaces.

The IsObject returns True if the expression is evaluated as an object. If the expression repre-
sents any other datatype, the IsObject function returns False. For example, the following code Part
example returns True: II
<% Ch
Set bc = Server.CreateObject(“MSWC.BrowserType”)
Response.Write IsObject(bc) & “<BR>”
%>
7

10.1389-6 CH07 131 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 132 Chapter 7 Datatype Conversion Features

The VarType Function

The core functionality of the IS datatype identifiers is powered by the VarType function. The
VarType function presents the subtype category of a variable by using the following syntax:

VarType(varname)

where varname is any variable name. Unlike the other identifier functions, the VarType function
returns the VBScript constants that are used to identify the datatype. For instance, in the fol-
lowing example the result of the VarType function indicates that the tested variable is an
ActiveX object type.
<%
dim myVarType ‘=== Declare variables
Set bc = Server.CreateObject(“MSWC.BrowserType”) ‘=== Create Browser sniffer
myVarType = VarType(bc) ‘=== Returns a 9, vbObject
%>

The complete listing of variable types is presented in Table 7.1.

Table 7.1 Identify the Sub Datatype with the VarType Function
Constant Value Description
vbEmpty 0 Uninitialized (default)
vbNull 1 Contains no valid data
vbInteger 2 Integer
vbLong 3 Long
vbSingle 4 Single
vbDouble 5 Double
vbCurrency 6 Currency
vbDate 7 Date
vbString 8 String
vbObject 9 Object
vbError 10 Error
vbBoolean 11 Boolean
vbVariant 12 Variant (used only for arrays of variants)
vbDataObject 13 Data access
vbDecimal 14 Decimal
vbByte 17 Byte
vbArray 8192 Array

10.1389-6 CH07 132 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 The Formatting Functions 133

The Formatting Functions

One of the major stumbling blocks that developers encounter when learning VBScript is the
lack of formatting functions that are inherent to programming languages like Visual Basic.
However, VBScript 2.0 has introduced new formatting strings to aid in the conversion and
manipulation of currency, data and time values, numbers, and percent values. These formatting
functions are based on the number, currency, time, and date settings found in the Regional
Settings of the Control Panel. Figure 7.1 illustrates the options available in the Regional Set-
tings component of the Control Panel. Keep in mind that Regional Settings are only specific to
the machine the VBScript code is being executed on. For example, if you are using a format-
ting function on an Active Server Page, the Regional Settings on the Web server will be used to
dictate formatting. Whereas, if you are using client-side VBScript to display the current time
and date, the local Web browser’s Regional Settings would be applied.

FIG. 7.1
Configure the Regional
Settings within the
Control Panel.

The FormatDateTime Function

The FormatDateTime function enables you to present a date expression in five convenient com-
mon date/time formats. The formats, as listed in Table 7.2, present the dates in a general for-
mat, a long and short date format, and a long and short time format.

Table 7.2 Date Format Constants Part

Constant Value Description II
VbGeneralDate 0 Displays a date or time. Ch
VbLongDate 1 Displays the long date format. 7
VbShortDate 2 Displays the short date format.
VbLongTime 3 Displays a long time format.
VbShortTime 4 Displays a short time format.

10.1389-6 CH07 133 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 134 Chapter 7 Datatype Conversion Features

Using these settings, you can dictate how date and time information will be manipulated. This
formatting feature is often used when cleaning data for database storage or is used to present
date and time information to the user without a large amount of hand coding.

The FormatDateTime function returns the formatted date or time value by using the following
syntax:
FormatDateTime(Date[,NamedFormat])

where Date is a date expression and NamedFormat is the VBScript constant that determines the
presentation display format. For example, the variable myDate is set to the current date and
time of the system using the Now function. However, you can extract and display the date and
time components of the time in long or short formats, as shown in the following code:
<%
dim myDate, myFormatDate ‘=== Variable declaration
myDate = cdate(Now) ‘=== Initiates Variables
myFormatDate = FormatDateTime(myDate,vbGeneralDate) ‘=== Produces 6/20/97
➥7:51:21 PM
myFormatDate = FormatDateTime (myDate,vbLongDate) ‘=== Produces Friday,
June 20, 1997
myFormatDate = FormatDateTime (myDate,vbShortDate) ‘=== Produces 6/20/97
myFormatDate = FormatDateTime (myDate,vbLongTime) ‘=== Produces 7:51:21 PM
myFormatDate = FormatDateTime (myDate,vbShortTime) ‘=== Produces 19:51
%>

The FormatCurrency Function

The FormatCurrency function focuses on returning a formatted currency value. The currency
symbols are based on the selected number, currency, time, and date settings used in the Re-
gional Settings within the Control Panel. The FormatCurrency function returns the currency
format using the following syntax:
FormatCurrency(Expression[,NumDigitsAfterDecimal [,IncludeLeadingDigit
[,UseParensForNegativeNumbers [,GroupDigits]]]])

where expression is the required numeric expression to be formatted. As you can see, the
FormatCurrency has several options. The NumDigitsAfterDecimal option indicates the optional
number of digits to display to the right of the decimal point. The IncludeLeadingDigit is the
optional constant used to determine whether or not to include the leading zero. The
UseParensForNegativeNumbers enables parentheses to be displayed in place of the negative
sign to represent numbers less that zero. The GroupDigits gives the capability to follow the
group delimiter options specified in the Currency tab of the Control Panel.

The last three options, IncludeLeadingDigit, UseParensForNegativeNumbers, and

GroupDigits are Tristate constants. The Tristate constants provide the flexibility to use a con-
stant that has three positions or values. These display values are usually dictated by the Re-
gional Settings of the Control Panel. The Tristate constants are defined in Table 7.3.

10.1389-6 CH07 134 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 The Formatting Functions 135

Table 7.3 The FormatCurrency Tristate Constants

Constant Value Description
TristateTrue -1 Option is True.
TristateFalse 0 Option is False.
TristateUseDefault -2 Use the settings from the computer’s Regional Settings.

For the FormatCurrency function, the specific Regional Settings are located in the Currency
and Number tabs.

For example, the following code demonstrates the FormatCurrency features:

<%
dim myCurrencyValue, myInterestValue, myFormatCurrency, myTotal

myCurrencyValue = 16299.98: myInterestValue = 0.21 ‘=== Initialize Variable

myTotal = myCurrencyValue + (myCurrencyValue*myInterestValue)
myInterestLost = myCurrencyValue - myTotal

myFormatCurrency =FormatCurrency(myCurrencyValue) ‘=== Displays $16,299.98

myFormatCurrency =FormatCurrency(myTotal, 4) ‘=== Displays $19,722.9758
myFormatCurrency =FormatCurrency(myInterestLost, TristateFalse, 0, TristateTrue)
‘=== Displays -$3,423
myFormatCurrency =FormatCurrency(myInterestLost, TristateTrue, 0, -1)
‘=== Displays ($3,423)
%>

The FormatPercent Function

The FormatPercent function returns the formatted percent value of an expression. The
FormatPercent function accomplishes this by multiplying the value by 100 and concatenating
the percent (%) character onto the results. The FormatPercent function uses the following
syntax:
FormatPercent(Expression[,NumDigitsAfterDecimal [,IncludeLeadingDigit
[,UseParensForNegativeNumbers [,GroupDigits]]]])

where Expression is the required expression to be formatted. The remaining options,

NumDigitsAfterDecimal, IncludeLeadingDigit, UseParensForNegativeNumbers, and
GroupDigits, are the same options available in the FormatCurrency function. However, in this
situation, the default system settings are located in the Number tab of the Regional Settings in
the Control Panel. Part

The following code demonstrates examples that utilize the different FormatPercent options.
II
Ch
<%
Dim myInterestRate, myFormatPercent
myInterestRate = 0.2159 ‘=== Initialize Interest Rate
7

10.1389-6 CH07 135 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 136 Chapter 7 Datatype Conversion Features

myFormatPercent = FormatPercent(myInterestRate) ‘=== Returns 21.59%

myFormatPercent = FormatPercent(myInterestRate, 1) ‘=== Returns 21.6%
myFormatPercent = FormatPercent(-myInterestRate,TristateTrue, 0, -1)
➥‘=== Returns (22%)
myFormatPercent = FormatPercent(myInterestRate/100, , -1, TristateTrue)
➥‘=== Returns 0.22%
%>

The FormatNumber Function

The FormatNumber function enables the customized display of a numeric expression. Following
in the tradition of the other formatting functions, the FormatNumber uses the following syntax:
FormatNumber(Expression[,NumDigitsAfterDecimal [,IncludeLeadingDigit
[,UseParensForNegativeNumbers [,GroupDigits]]]])

where Expression is the required numeric expression to be formatted. The available options
and Tristate constants are also the same as the FormatCurrency and FormatDateTime functions.
<%
Dim myNumber, myFormatNumber
myNumber = 22/7 ‘===
➥Initialize variables
myFormatNumber = FormatNumber(myNumber) ‘=== Results
➥in 3.14
myFormatNumber = FormatNumber(myNumber, 8) ‘=== Results
➥in 3.14285714
myFormatNumber = FormatNumber(-myNumber, 1, 0, -1) ‘=== Results
➥in (3.1)
myFormatNumber = FormatNumber(3-myNumber, 3, -1, TristateFalse ‘=== Results
➥in -0.143
%>

Datatype Conversions
Throughout the investigation of the various operators and functions in VBScript, the potential
hazards of the automatic conversion of the Variant datatypes abound. To prevent un-intentional
errors from occurring, VBScript provides conversion utilities to tighten the control on the sub
datatypes. Furthermore, as Active Server Pages begin to distribute applications worldwide, is-
sues such as regional formatting become a concern. The conversion functions were designed
with this in mind. Most of the conversion functions recognize regional currency and formatting
issues and can seamlessly interact with different display settings without changing the underly-
ing datatypes.

VBScript provides nine conversion functions to help reduce the chance of runtime or logic-
based errors. The conversion functions and descriptions are listed in Table 7.4.

10.1389-6 CH07 136 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 Datatype Conversions 137

Table 7.4 Sub Datatype Conversion Functions

Function Return Type Data Range
CBool Boolean True or False

CByte Byte 0 to 255

CCur Currency -922,337,203,685,477.5808 to 922,337,203,685,477.5807
CDate Date Any valid date expression
CDbl Double -1.79769313486232E308 to -4.94065645841247E-324 for
negative values; 4.94065645841247E-324 to
1.79769313486232E308 for positive values
CInt Integer -32,768 to 32,767; fractions are rounded
CLng Long -2,147,483,648 to 2,147,483,647; fractions are rounded
CSng Single -3.402823E38 to -1.401298E-45 for negative values;
1.401298E-45 to 3.402823E38 for positive values
CStr String Returns string values, except Null

When using the conversion functions, the data ranges are important to keep in mind. If you try
to use a function to evaluate an expression that is not in the targeted conversion group, a
runtime error will occur.

Cbool Function
The CBool function converts an expression into a subtype of Boolean. The CBool function ac-
complishes this by using the following syntax:
CBool(expression)

The Boolean conversion process will convert any numeric expression to either True or False.
Any other expressions will produce a Type Mismatch error. For example, the following code
uses the CBool function to convert an expression into Boolean subtype.
<%
Dim myBoolValue1, myBoolValue2, myBoolResult, myDataType
myBoolValue1 = 100: myBoolValue2 = 99.5 ‘=== Initialize variables.
myBoolResult = CBool(myBoolValue1 = myBoolValue2) ‘=== Returns a Boolean of
➥False
myDataType = VarType(myBoolResult) ‘=== Identifies Boolean Part
subtype
II
myBoolValue1 = 100: myBoolValue2 = 100 ‘=== Reset Values
Ch
myBoolResult = CBool(myBoolValue1 = myBoolValue2) ‘=== Returns a Boolean of
➥True 7
myBoolValue1 = 100 ‘=== Resets Value
myBoolResult = CBool(myBoolValue1) ‘=== Returns a Boolean of
➥False
%>

10.1389-6 CH07 137 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 138 Chapter 7 Datatype Conversion Features

Notice in the first two examples, the Boolean results of the expression myBoolValue1 =
myBoolValue2 were successful in returning Boolean values. In the first situation, the expres-
sion returned False because the numeric values were not the same. However, in the second
situation, the CBool function returned True because the variables were equal. In the third situa-
tion, the CBool function returns False because even though the variable exists, the expression
of the single variable by itself does not return a value. Keep in mind that the Boolean conver-
sion process can only evaluate numeric expressions. In the following code example, the Bool-
ean conversion process produces runtime errors when trying to evaluate string values.
<%
myBoolValue1 = “Foo” ‘=== Set Value to string Foo
myBoolResult= CBool(myBoolValue1) ‘=== Results in Type Mismatch Error

myBoolValue1 = “” ‘=== Reset Value

myBoolResult= CBool(myBoolValue1) ‘=== Results in Type Mismatch Error
%>

CByte Function
The CByte function converts an expression into a subtype of type Byte. The CByte function
requires the following syntax:
CByte(expression)

The following example illustrates using the CByte conversion function:

<%
Dim myByteValue, myDataType
myByteValue = 99.987654 ‘=== myByteValue is a Double.
myDataType = CByte(myByteValue) ‘=== Returns 100
%>

N O T E Before using the conversion functions, remember to check the range of the input expres-
sion to prevent conversion errors. ■

CCur Function
The CCur function is used to convert ordinary variant expressions into the currency subtype.
The CCur function uses the following syntax to return conversion results.
CCur(expression)

The currency conversion function is important when considering your potential global user
base. The currency conversion functions recognize the different currency formatting issues,
such as various separator symbols, different thousand separators, and other currency display
options.
<%
Dim myCurrValue, myDataType ‘=== Initialize Values
myCurrValue = 16299.214588 ‘=== Sets myCurrValue is a Double.
myDataType = CCur(MyCurrValue * 2)‘=== Returns
32598.4292
%>

10.1389-6 CH07 138 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 Datatype Conversions 139

The currency conversion function converts the value 32598.429176 to 32598.4292.

CDate Function
The CDate function is used to merge a variant variable into a date subtype. The CDate function
requires the following syntax:
CDate(date)

where date is a valid date expression. The IsDate function mentioned in the beginning of this
chapter is used to determine whether an expression can be converted into a date subtype. The
CDate function is actually the function that does the conversion into the date subtype.

One advantage that the CDate function has at its disposal is the capability to recognize date and
time literals. Date literals are comprised of a defined set of characters surround by pound (#)
signs. The defined character set is defined in the Regional Settings of the Control Panel. You
can use the CDate function to display localized regional formats when building multi-region
applications.

<%
Dim myDateValue, myTimeValue, myDataType
myDateValue = “February 12, 1969” ‘=== Initialize date.
myDataType = CDate(myDateValue) ‘=== Convert to Date 2/12/69

myTimeValue = “4:35:47 PM” ‘=== Initial time.

myDataType = CDate(myTimeValue) ‘=== Convert to Date 4:35:47 PM.
%>

N O T E Use date literals, such as #4/30/1997# and #12:30:00 PM#, to ensure that your date
and time strings are portable across multiple regions. ■

CDbl, CInt, CLng, CSng

The CDbl, CInt, CLng, and CSng conversion functions are all used to provide conversion be-
tween the number datatypes. The CDbl function is used to convert variables to double preci-
sion subtypes. The CInt function is used to transform variables into single precision subtypes.
The CLng function is used to convert variables into Long subtypes. The CSng function is used to
convert an expression into a single-precision sub datatype. These numeric conversion func-
tions use the following syntax:
nConversionFunctions(expression)
Part
where nConversionFunctions is the CDbl, CInt, CLng, or CSng conversion names and expres-
sion is the evaluated expression in any valid expression.
II
Ch
The following example illustrates the conversion of a double-precision sub datatype to a single-
precision and Integer datatype. 7

10.1389-6 CH07 139 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 140 Chapter 7 Datatype Conversion Features

<%
Dim myFirstValue, mySecondValue, myDataType, myValue
myFirstValue = 157.3585705: mySecondValue = 194.6465794 ‘=== Initializes
➥variables
myDataType = VarType(myFirstValue) ‘=== Returns a Double
myDataType = VarType(mySecondValue) ‘=== Returns a Double
myValue = CSng(myFirstValue) ‘=== Converts to
➥Single 157.3586
myValue = CSng(mySecondValue) ‘=== Converts to
➥Single 194.6466
myValue = CInt(mySecondValue) ‘=== Converts to
➥Integer 195
%>

N O T E When using the CInt and CLng functions, remember that these functions always round to
the nearest even number, therefore 2.5 rounds to 2, and 3.5 rounds to 4. ■

CStr Function
The CStr function is used to convert an expression into a string subtype. The CStr function
uses the following syntax:
CStr(expression)

where expression is a valid expression. The result of the CStr function always returns a string
subtype, with the exception of when the expression contains a NULL value. Table 7.5 illustrates
the results of the CStr function when a variety of subtypes compose the expression.

Table 7.5 The CStr Return Values

If CStr(expression) CStr Returns
Boolean A string containing True or False
Date A string containing a date in the short-date format
Empty A zero-length string (“”)
Error A string containing the word Error followed by the error number
Numeric A string containing the number
Null A runtime error

The following example uses the VarType function to determine the subtypes.
<%
Dim myValue, myDataType
myValue = 16299.99 ‘=== Initialize variable
myDataType = VarType(myValue) ‘=== Initially a Double
myValue = CStr(myValue) ‘=== Converts to String
myDataType =VarType(myValue) ‘=== Results in a String

10.1389-6 CH07 140 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 From Here… 141

myValue = CStr(Now) ‘=== Convert a Date to String, produces

➥6/21/97 12:39:05 AM
myDataType =VarType(myValue) ‘=== Results in a String
%>

From Here…
From this chapter, you learned that the VBScript datatype identification and conversion func-
tions are used to explicitly convert Variant sub datatypes. In addition, VBScript 2.0 has intro-
duced four new formatting functions that are used to easily format date/time, currency,
percent, and numeric values. These new functions eliminate the need to build VBScript format-
ting code from scratch into your scripts.

For related information, please see the following chapters:

■ Chapter 10, “The Composition of an Active Server Application,” will illustrate the
different pieces of an ASP application.
■ Chapter 12, “Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object,” demonstrates how the Server Object
controls and manages the ASP applications.
■ Chapters 17 and 18, “Implementing Existing Server-Side Components” and “Creating
Your Own Active Server Components,” will teach how to use server-side VBScript to
provide connections to pre-built and customizable servers.
■ Chapter 20, “Database Access with ADO,” provides a firsthand approach on using
VBScript to manipulate your databases.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.
■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,”
demonstrates how to build ActiveX components to create an n-tier distributed application
for the Amazing DataWidget WareHouse Catalog.

Part
II
Ch

10.1389-6 CH07 141 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 142 Chapter 7 Datatype Conversion Features

10.1389-6 CH07 142 10/26/98, 9:11 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH07 lp#1
 8
143

C H A P T E R
Part
II
Ch

Create Active Server Scripts

Putting It All Together with VBScript
Learn how to implement VBScript

with VBScript within your Active Server Pages to

create application and processing
logic that runs on the Web server.

Create Client-Side VBScript

Utilize client-side scripts to process
application logic that will be ex-
ecuted within the Internet Explorer
This chapter demonstrates how to use VBScript to build a browser to perform client-side
validation.
miniature application that uses client- and server-side
processing scripting to calculate the aerodynamic drag for
Create Browser-Specific Code
a vehicle over a range of speeds. This application uses
Discover how to use Active Server
server-side ASP scripts to perform to the core processing
Pages to identify the requesting
of the engineering calculations, which can be processed
browser and tailor the client-side
by any browser. Once the functionality of the VBScript scripting to the specific browser
programming logic has been demonstrated, VBScript will type.
also be used on the client to ensure that the required
HTML fields are properly filled in by the user. ■

11.1389-6 CH08 143 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 144 Chapter 8 Putting It All Together with VBScript

Application Overview
One of the largest benefits of using ASP on the server is that any browser can have access to
the application logic embedded on the Web server. To illustrate that fact, this application will
demonstrate a sample application that performs aerodynamic drag calculations. These calcula-
tions are performed by executing ASP scripts on the Internet Information Server from any
HTML-compliant browser.

Figure 8.1 demonstrates the interface to enter information on the vehicle specification page in
Internet Explorer.

FIG. 8.1
Using Internet Explorer
to gather vehicle
specifications.

The advantage of using an HTML-based system is that you are not constrained to one specific
browser type; you can gather the same information from other browsers. For example, Figure
8.2 illustrates using Netscape Navigator to enter vehicle information.

After the user enters the vehicle information, the ActiveX scripting engine on the Internet
Information Server processes the calculations. In this situation, the ActiveX scripting engine
will process VBScript to perform the calculations, but the calculations can be written in JScript,
PERL, REXX, or any other script-compliant engine. After the drag values have been calculated,
the results are delivered from the Internet Information Server to the browser as HTML (see
Figure 8.3).

The key point to remember here is that only HTML is delivered to the browser. The process-
ing of VBScript occurs only on the server. Processing VBScript on the server does not limit the

11.1389-6 CH08 144 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Application Overview 145

access to ActiveX-compatible browsers or operating systems. Figure 8.4 illustrates the com- Part
plete results being delivered to Netscape Navigator (with no ActiveX plug-in installed—there
are no cards up my sleeves!).
II
Ch
FIG. 8.2
Any HTML-compliant
8
browser can be used to
access the application.

FIG. 8.3
After processing, the
results of the server-
side calculations are
delivered to the
browser.

11.1389-6 CH08 145 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 146 Chapter 8 Putting It All Together with VBScript

FIG. 8.4
Processing VBScript on
the server does not limit
application access to a
browser-specific delivery
mechanism.

You will build this mini-application in three phases. The first phase demonstrates how to use
server-side VBScript to create the basic functionality of the Active Server application. Second,
you will add client-side VBScript and JavaScript to perform client-side validations. Finally, you
will return to the server-side VBScript to maximize the flexibility and power of the server-side
scripts.

Creating Server-Side Active Server Scripts

with VBScript
The first phase of building this application demonstrates some of the features of server-side
VBScript processing. The mini-application will use Active Server Pages to run a fuel economy
simulation model for vehicles. Specifically, this application demonstrates the relation of drag or
air resistance to velocity. This chapter will not cover the specifics of the engineering formulas
used to calculate this information, but will focus on how to use VBScript 2.0 within Active
Server Pages. Keep in mind that this application serves as a template that can contain any type
of application logic.

As you saw in Figures 8.1 and 8.3, two pages are used in the application. The first page is a
simple forms-based page that collects information from the user. This page passes information
to the second page via the Internet Information Server. The second page is the page that actu-
ally contains the application logic for the calculations, and it is also responsible for displaying
the calculation results to the requesting browser.

11.1389-6 CH08 146 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Server-Side Active Server Scripts with VBScript 147

To start building this application, you can use your favorite method of creating and managing Part
Active Server applications. You can use the Internet Service Manager in IIS 3.0 to create a new
virtual directory to house your pages. Alternatively, if you are using IIS 4.0, you can use the
II
Microsoft Management Console to create a new Web site or virtual directory. (If you need Ch
more information on configuring your Web site with the MMC, please see Chapter 31, “Using
Internet Information Server 4.0 to Manage Your Web Applications.”)
8
As far as Active Server Page development tools, choose your favorite ASP development envi-
ronment. If you like, you can use your favorite ASCII editor to create and edit your Active
Server Pages. Alternatively, FrontPage enables you to create Active Server Pages and server-
side scripting, and Visual InterDev provides the most functionality tailored toward application
developers. Regardless of the development environment you choose, the functionality of Active
Server Pages remains the same.

Creating the Visual InterDev Project

Visual InterDev will be used in this chapter for this VBScript simulation project. After creating
a new Web project on the Web server, create two Active Server Pages that will act as the input
page and the results page. To create new Active Server Page files within Visual InterDev, select
New from the File menu. Next, choose Active Server Page from the File tab, and name the first
ASP file Vehicle Info and the second ASP file CalcDrag. Visual InterDev will create Active
Server Pages based on the default ASP template. For more information on using Visual
InterDev, please see Chapter 27, “Visual InterDev: The Active Server Pages Integrated Devel-
opment Environment.”

When you create the new files in Visual InterDev, the files are automatically opened in the
Visual InterDev editor. This editor serves as your primary development environment.

Creating a Static Page The first page, Vehicleinfo.htm, does not require any server-side
processing when the page is delivered to the browser. It is simply used to collect information
from the user. With the Active Server Page revolution, you may be tempted to convert all of
your HTML files to ASP files. However, this is not recommended. When Internet Information
Server processes files with an .ASP extension, it uses the Active Server scripting engine to
interpret the page—even if no Active Server script is embedded on the page. This relatively
small amount of processing power can be significantly amplified depending on your Web site’s
size and load.

TIP Resist the temptation to convert all HTML files to ASP files. To prevent the unnecessary use of
processing power, only convert the HTML files that require script processing.

To create this page, you can code the HTML by hand or use any GUI HTML creation tool, such
as FrontPage or HotMeTal. You can activate any external tool from within Visual InterDev by
right-clicking the file you want to open. When you choose the Open With option, a pop-up
dialog displays the configured tools for this file type (see Figure 8.5).

11.1389-6 CH08 147 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 148 Chapter 8 Putting It All Together with VBScript

FIG. 8.5
With Visual InterDev, you
can use external tools to
create and edit files.

This example will use Visual InterDev FrontPage to create the template page. The FrontPage
version that ships with Visual InterDev is a scaled-down version of the full Microsoft FrontPage
product that enables basic Web creation and script wizard interface, but does not include items
such as the various wizards and clipart libraries that are available in the full product. If you do
not have FrontPage for Visual InterDev configured, select the Add button to add the new edi-
tor. The FrontPage editor that ships with Visual InterDev is usually located at /devestudio/
VintDev/ISFP/Bin/isfp.exe.

After the Visual InterDev FrontPage editor is open, create the Web page that will collect the
user information. This page consists of several nested tables, text boxes, and Submit and Reset
buttons. The nested tables are used to control positioning of objects on a Web page. Figure 8.6
illustrates using tables with the border property set to zero.

FIG. 8.6
Use invisible nested
tables to control page
layout.

11.1389-6 CH08 148 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Server-Side Active Server Scripts with VBScript 149

Text boxes are used to enable the user to enter information into the page. The HTML Submit Part
and Cancel buttons are used to control the processing of the form objects on a page. The Sub-
mit button posts or sends the information to the designated page on the Web server. The spe-
II
cific page is set within the form Action event. For example, the following code will send the Ch
information entered on a form to the calcdrag.asp page.
8
<FORM ACTION=”Calcdrag.asp” METHOD=”POST” NAME=”frmInfo”>

The Cancel button is used to reset all HTML objects to their initial values.

Take time to name all your HTML objects on the page using a standard naming convention.
There are several naming conventions available, just be sure to pick a method and stick to it.
Standardizing your code will benefit you in the long run by creating code that is easier to read
and reducing errors due to misspellings and bad references.

For example, name the HTML forms with the frm prefix to help you visually identify an object’s
type. If the form is used to collect information about the user’s vehicle, name the form
frmVehicleleInfo. You name the objects in FrontPage by double-clicking the object or right-
clicking and selecting the appropriate Properties tag. Standardizing the names of your objects
is important because it will also help to prevent errors. For example, if you name a text box
txtVehicleWeight, take care to remember the capitalization scheme you have chosen.
VBScript, which is case insensitive, can reference an object using any combination of upper-
and lowercase letters. However, JavaScript is case sensitive. Therefore, if you try to reference
the text box txtVehicleWeight as txtvehicleweight within JavaScript, an error will occur.

Creating the Dynamic Template After you create the information collection page, save the
application. Now it is time to start creating the core VBScript programming functionality that
performs the vehicle calculations in Active Server Pages. The role of the second page is to
accept the vehicle information passed to it, process it, and display the results. The second page
will act as a template to host whatever information is processed by the ASP script. For example,
the physical HTML page should have the same format regardless of the range of valid vehicle
weights or speed ranges to be calculated.

Using FrontPage to Create a Template In most situations, developers can perform data
collection, manipulation, and retrieval with relative ease, but displaying this information is often
the tedious part. This is particularly true with Web deployment and the joys of learning HTML.
To help take away some of the pains of developing HTML, use the same HTML GUI tools that
you used to create the static page to create the template to display the dynamic information.

In this example, you want to return the information entered about the test vehicle specifica-
tions at the top of the page. After creating the header, you need to create a template that will be
repeated for every record displayed. To do that, create an HTML table with two rows under the
header information. The first row is used as the table header describing the columns. The
second row represents the row that will be repeated within the VBScript looping code to dis-
play the results, as shown in Figure 8.7.

After saving and closing the page, you are now ready to start creating the core server-side
VBScript.

11.1389-6 CH08 149 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 150 Chapter 8 Putting It All Together with VBScript

FIG. 8.7
Use GUI HTML editors to
quickly create HTML
templates to write your
ASP data into.

Building Data Retrieval and Display

To help further define how to code this mini-application, you should review the workflow pro-
cess. The first phase obtains the information that is passed to the page and displays the values
in the header. The second phase takes the information and processes it according to the given
engineering equations and writes the output to the page.

VBScript Declarations The first line of the Active Server Page should contain process direc-
tives that are used to help the Active Server scripting engine process that Active Server Page.
For instance, to set the default scripting engine required to process the script contained in the
<% and %> tags, use the LANGUAGE directive:

<%@ LANGUAGE = VBScript %>

This tag should be placed within the ASP file when the file is initially created via Visual
InterDev. If the tag has been deleted, add it to the first line of text.

Remember that the process directive must be on the first line of the ASP, and you must include
a space between the @ and the LANGUAGE tag. You can also set other characteristics of the page
within the process directive. The process directive enables you to use the CODEPAGE keyword
to work with multilingual sites, the LCID to set the locale identifier, the TRANSACTION keyword to
enable transaction processing, and ENABLESESSIONSTATE to track user sessions. These direc-
tives are covered in greater detail in Chapter 10, “The Composition of an Active Server Applica-
tion.”

After setting the default scripting language for the page, insert the VBScript Option Explicit
statement. The Option Explicit statement forces all variables to be explicitly declared by
using the Dim, ReDim, Public, or Private keywords.

11.1389-6 CH08 150 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Server-Side Active Server Scripts with VBScript 151

The Option Explicit tag must be declared before any header information is sent to the Part
browser. Failure to initiate this will generate an Expected statement error, as shown in
Figure 8.8.
II
Ch
FIG. 8.8
Failing to define the
8
Option Explicit
statement before
sending information to
the browser will
generate an error.

Placing the Option Explicit statement after any HTML tag, as shown below, generates this
error:
<%@ LANGUAGE=”VBSCRIPT” %>
<!DOCTYPE HTML PUBLIC “-//IETF//DTD HTML//EN”> <!--Output is sent to computer
➥à
<% Option Explicit %> <!--Generate error à
<HTML>

To correct this, simply place the Option Explicit tag after the LANGUAGE directive:
<%@ LANGUAGE=”VBSCRIPT” %>
<% Option Explicit %> <!--Sets Option Explicit à
<!DOCTYPE HTML PUBLIC “-//IETF//DTD HTML//EN”> <!--Output is sent to computer
➥à
<HTML>

Next, create variables to store the input from the user so that you can manipulate the values in
the engineering formulas. You can accomplish this by using the Dim keyword and the variable
name. For example, the following code declares variables that represent the input items from
the user:
<% ‘=== User Variables
Dim Weight ‘=== Vehicle Weight
Dim RRConst ‘=== Rolling Resistance Constant
Dim RRXcoef ‘=== Rolling existence Coefficient
Dim Grade ‘=== Grade
Dim Cd ‘=== Coefficient of Drag
Dim Sarea ‘=== Frontal Surface Area
Dim Headwind ‘=== Headwind
Dim Density ‘=== Density of medium
Dim StepSize ‘=== Display Increment
Dim VelBegin ‘=== Begin Display at this velocity
Dim VelEnd ‘=== Close Display at this velocity
%>

11.1389-6 CH08 151 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 152 Chapter 8 Putting It All Together with VBScript

Comments are used after each declaration statement to help identify the purpose of each vari-
able. However, to improve performance, the multiple variable declarations can all occur on one
line:
<%
Dim Weight, RRConst, RRXcoef, Grade, Cd, Sarea, Headwind, Density, StepSize,
VelBegin, VelEnd
➥%>

The improvements come not from the reduced number of comments on a page, but from the
number of times VBScript has to process each line. By collecting the multiple declaration state-
ments within one statement, VBScript only has to parse the line one time. If declare statements
are made on individual lines, VBScript must parse each line separately.

Retrieving User information Now that you have created variables on the server to hold the
user information, you are going to use an Active Server object to help retrieve information
from the requesting Web page. The Request object is used to extract values from items con-
tained on an HTML form. Because you have to use this information to process logic on the
page, assign these values to the variables declared in the previous section.

To obtain the values the users entered in a text box on a form, use the text box name within the
Request object, as shown in Listing 8.1.

Listing 8.1 Gathering information from the HTML form.

<%
Weight = Request.Form(“txtweight”)
RRConst = Request.Form(“txtrrconstant”)
RRXcoef = Request.Form(“txtxcoef”)
Grade = Request.Form(“txtgrade”)
Cd = Request.Form(“txtcd”)
Sarea = Request.Form(“txtarea”)
Headwind = Request.Form(“txtheadwind”)
Density = Request.Form(“txtdensity”)
StepSize = Request.Form(“txtstepsize”)
VelBegin = Request.Form(“txtbegin”)
VelEnd = Request.Form(“txtend”)
%>

For more information on the Request object, see Chapter 15, “Retrieving User Information:
The Request Object.”

Writing the Information to the Page Now that you have the information in memory on the
page, you can create the page header, which returns the information the user entered for the
calculations. To accomplish this, you have to write the variable information to the page. Active
Server script uses two different ways to write information to a page: the Response.Write
method or its shorthand notation of <% =%>.

To write these values in the appropriate location on the template page that was created in
FrontPage, you should mix HTML tags with Active Server tags. Find the location where you

11.1389-6 CH08 152 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Server-Side Active Server Scripts with VBScript 153

need to write information on the page with the Visual InterDev editor. In this case, look for the Part
location of the string (weight). This string was added to the page to serve as a placeholder
when you created the page in FrontPage. When you have found this string in the Visual
II
InterDev editor, replace it with the Active Server tags to write the variable weight to the page. Ch
Specifically, you are looking for the following HTML:
8
<td valign=”top”> (Weight) <td> <! -- Original HTML from FrontPage à

The string (Weight) must be replaced with the Response tag, as shown in the following code:
<td valign=”top”> <% Response.Write weight %> </td>

Continue replacing the placeholders that were created in FrontPage with the appropriate
Response.Write tags. If you save and reopen the calcdrag.asp page in FrontPage, FrontPage
interprets the Active Server script as script, as shown in Figure 8.9.

FIG. 8.9
FrontPage uses a script
icon to represent the
Active Server script.

Processing the Logic

Now that you know how to integrate Active Server script within HTML, you will use VBScript
to perform calculations and use looping logic to complete a task a certain number of times.

VBScript has a variety of built-in mathematical functions to help perform the various calcula-
tions. For example, the drag force due to air resistance is expressed by the following equation:
Fdrag = 0.5 * Density * ((Velocity + Headwind)^2)*(88/60)^2 * Cd * Sarea

11.1389-6 CH08 153 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 154 Chapter 8 Putting It All Together with VBScript

As you can see from this equation, VBScript handles addition, multiplication, division, and
exponential arithmetic statements. Furthermore, you can tap into the trigonometric functions.
The following example illustrates using the arctangent function to calculate the slope:
slope = atn(Grade/100)*(180/(22/7))

Because you want to perform the drag calculations over a range of velocities, you need to
implement looping logic. This looping logic will repeat a section of code either until a condition
is met or for a specific number of times. In this situation, you will implement a For...Next loop
to repeat the drag calculation for a specific number of times. Usually you implement a
For...Next loop from 1 to an upper limit with a step size of 1. However, in this situation, you
have asked the user to specify the start and end positions and at what increments to perform
the calculations. You can implement this conditional logic to calculate the drag force for mul-
tiple velocities starting at velBegin and ending at velEnd with an increment size of stepsize:
<%
Dim iVelocity
For iVelocity = velbegin to velend step stepsize
Velocity = iVelocity
Fdrag = 0.5 * Density * ((Velocity + Headwind)^2)*(88/60)^2 * Cd * Sarea
Next
%>

This iterative looping procedure enables the application to calculate and show the drag values
from 5 to 95 mph at every 10 mph.

Now that the calculations are done, you need to write these values to the results template
table, which is similar to the approach you used to generate the header information. To write
the calculations as new rows in a table, you have to create new table rows (<TR>) and table
cells (<TD>) within the For...Next loop by using the Response.Write method, as shown in
Listing 8.2.

Listing 8.2 Using a For...Next loop to calculate drag at various speeds.

<%
Dim Frolling, Slope, Velocity, Fdrag, iVelocity, Ftotal, Fgrade
Slope = atn(Grade/100)*(180/(22/7))
For iVelocity = velbegin to velend step stepsize
Velocity = iVelocity
Response.Write “<TR>”
Response.Write “<TD width=’18'>&nbsp;</TD>”
Response.Write “<TD valign=’top’><font face=’arial’ size=’2'
color=#000080>” &Velocity “</font></TD>”
Response.Write “<TD width=’2'>&nbsp;</TD>”
Fdrag = 0.5 * Density * ((Velocity + Headwind)^2)*(88/60)^2 * Cd * Sarea
Response.Write “<TD valign=’top’><font face=’arial’ size=’2'
color=#000080>” &Round(Fdrag,6) &”</font></TD>”
Fgrade = Weight*Sin(slope * (22/7)/180)
Response.Write “<TD valign=’top’><font face=’arial’ size=’2'
color=#000080>” &Round(Fgrade,6) &”</font></TD>”
Response.Write “<TD width=’2'>&nbsp;</TD>”
Ftotal = Fdrag+Frolling+Fgrade
Response.Write “<TD valign=’top’><font face=’arial’ size=’2'

11.1389-6 CH08 154 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Client-Side VBScript 155

color=#000080>” &Round(Ftotal,6) &”</font></TD>”

Part
Response.Write “<TD width=’6'>&nbsp;</TD>”

Next
Response.Write “</TR>” II
%> Ch

8
Notice the use of batch processing of server scripting instead of mixing HTML tags and Active
Server script delimiters. This helps increase the performance by reducing the need to switch
back and forth between processing server script and HTML tags.

This looping logic also incorporates the use of a VBScript formatting feature. The Round func-
tion is used to round a number to a specific number of decimal places. In this situation, round
all calculated values to six decimal places.

Minor Error Trapping

Finally, you want to include some server-side validation before you try to process the calcula-
tions. Even though the server is not the optimal location to validate input, this preliminary
example is used to demonstrate that it is possible. The best place to perform the validation is
within the client browser’s environment. The advantage of processing information on the client
is that it reduces the server load and network traffic. However, for the purposes of the VBScript
server examples, the application will display an error page if the user does not properly enter
certain values.

To accomplish this simple validation and display technique, server-side script will determine
whether the values of the weight and surface area variables are greater than zero. To check the
value of both the variables, use the conditional If...Then statement with the OR comparison
operator. The If...Then statement is used to test a specific condition and take action depend-
ing on the result. The conditional operator OR is used to test whether either condition evaluates
to True. Together, these can be implemented to ensure that the variables Weight and Sarea are
both greater than or equal to zero.
<%
If Weight <= 0 Or Sarea <= 0 Then
Response.Redirect “Error.asp”
End If
%>

In this situation, if either the Weight is negative (less than or equal to zero) or the Sarea is
negative, the user is automatically routed to an error page. It is true that there are various
error navigation methods you can implement. However, for this example, I want to simply
demonstrate how to implement VBScript on the server without going into too much com-
plexity.

Creating Client-Side VBScript

As I mentioned earlier, the server is not the ideal place to validate user input. To reduce server
load and network traffic, client-side scripting should be implemented. This section will demon-
strate how to build both VBScript and JavaScript client-side validation scripts. The JavaScript

11.1389-6 CH08 155 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 156 Chapter 8 Putting It All Together with VBScript

functions are included because other browsers do not interpret VBScript, although it is sup-
ported on the IIS Web server and on the Internet Explorer browser. Loading VBScript-based
client-side scripting in other browsers will generate scripting runtime errors, as shown in
Figure 8.10.

FIG. 8.10
VBScript client-side
script will generate
errors on non VBScript-
compliant browsers.

The simplest validation to deploy on the client is to test the values of the text boxes. Because
the purpose of this application is to perform calculations, you must trap for blanks, zero values,
and non-numeric values to ensure proper arithmetic operations.

To build this validation method, you are going to create a subroutine that is responsible for
validating field values. This subroutine will first check to see if the values of the fields are valid.
If all the fields pass the inspection, the form is submitted to the server for processing. If a field
contains an invalid number, an error message will prompt the user about the incorrect entry
and set the focus to the troubled field.

Restructuring the HTML

The first step needed to build this functionality is to add the validation subroutine. The valida-
tion subroutine will check certain HTML fields for proper values before the user tries to per-
form the calculations. The validation routine is actually triggered with the Click event of the
Calculate button. If the validation passes, the form data will be passed onto the calculation
page. If the validation fails, the user is prompted and control is returned back to the browser.
If the validation passes, the information is sent to the Web server for processing. However, to
build this functionality, you have to change the Submit button into a regular HTML button
because the Submit button will always execute the Submit event of the form. By converting the
Submit button into a normal button, the Submit event of the form can be controlled via script-
ing. To change the Submit button to a normal button, either double-click the Submit button
within FrontPage to change its property or find the following line of code in the HTML file:
<INPUT LANGUAGE=”VBScript” TYPE=submit VALUE=”Calculate” NAME=”btncalcclient”>

Change the TYPE from Submit to Button, and add the Onclick event to reference a subroutine
called validate, as shown below:
<INPUT LANGUAGE = “VBScript” TYPE = button VALUE = “Calculate” ONCLICK = “call
validate()”
NAME = “btncalcclient”>

11.1389-6 CH08 156 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 Creating Client-Side VBScript 157

Notice that if the button type is left as a Submit button with the Onclick event, as shown below, Part
the form will be sent to the Web server for processing even if the validation subroutine fails.
II
<INPUT LANGUAGE=”VBScript” TYPE=SUBMIT VALUE=”Calculate” ONCLICK=”Call
➥Validate()” Ch
NAME=”btncalcclient”>
8
Building the Validation Subroutine
Now you are ready to build the validation subroutine. Because you are checking multiple text
box values on the form, you might be inclined to develop a series of nested If...Then...Else
loops. This brute-force method requires every if condition to perform a comparison operation
to determine whether the condition is met. A more efficient way to complete this task is to
implement the Select Case statement. The Select Case statement enables the condition
statement to be evaluated only one time and picks the execution option that matches the re-
sults. Listing 8.3 demonstrates using the Select Case statement to find all conditions that
return True. If an expression returns True, an error has been found and the return value of the
Select Case statement is set to False. After exiting the Select Case statement, the ret value
is examined to determine if all the values tested in the Select Case statement are valid. If so,
the form is submitted to the Web server by calling the Submit event of the form.

Listing 8.3 Using the Select Case statement to perform client-side validation.
<SCRIPT LANGUAGE=”VBScript”>
<!--
Sub Validate()
Select Case True ‘=== Test all conditions that return True
Case frminfo.txtweight.value <= 0
Ret = False
Msgbox “Weight must be greater than zero”
Call frminfo.txtweight.focus()
Case frminfo.txtarea.value <= 0
Ret = False
Msgbox “Frontal surface area must be greater than zero”, 16
Call frminfo.txtarea.focus
Case frminfo.txtbegin.value < 0
Ret = False
Msgbox “Beginning time value must be greater than zero”, 32
Call frminfo.txtbegin.focus()
Case frminfo.txtdensity.value <= 0
Msgbox “Density must be greater than zero”, 48
Ret = False
Call frminfo.txtdensity.focus()
Case Else
Ret = True
End Select
If Ret Then ‘=== Did validation pass?
frminfo.submit ‘=== If so, submit the form
End if
End sub
-->
</SCRIPT>

11.1389-6 CH08 157 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 158 Chapter 8 Putting It All Together with VBScript

If the subroutine finds an invalid value, the user is prompted with the appropriate warning, as
shown in Figure 8.11.

FIG. 8.11
Client-side scripting
reduces server load and
network traffic by
executing within the
browser environment.

You can expand the functionality of the Validate subroutine to be as complex as needed. For
example, you can tap in to the built-in VBScript variable functions to test whether a field is
empty or whether the value is a numeric value, as shown in Listing 8.4.

Listing 8.4 Using VBScript functions to perform client-side validation.

<%
Select Case True
Case Not IsEmpty(frminfo.txtdensity.value)
Ret = False
Msgbox “Please enter a density value”
Call frminfo.txtdensity.focus()
Case Not IsNumeric(frminfo.txtweight.value), frminfo.txtweight.value<=0
Ret = False
Msgbox “Incorrect data for weight”
Call frminfo.txtweight.focus()
Case Else
Ret = True
End Select
If Ret Then ‘=== Did validation pass?
frminfo.submit ‘=== If so, submit the form
End if

End Select
%>

Testing for a VBScript-Compliant Browser

To prevent VBScript client-side coding from being delivered to a non-VBScript–compliant
browser, Web server variables can be used to determine information about the requesting
client. The Web server can extract information about the requesting client from variables
passed to the server from the client’s HTTP header request (see Listing 8.5).

Listing 8.5 Using HTTP header information to detect the browser type.
<%
If Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”) Then
‘=== This is an IE browser
‘=== Put client-side VBScript functions here

11.1389-6 CH08 158 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 From Here… 159

Else
Part
‘=== This is not an IE browser. Is it a Netscape Browser?
If Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”Mozilla”) Then
‘=== Netscape Browser
II
‘=== Put JavaScript Functions here Ch
End If
End If 8
%>

This server-side VBScript uses the InStr function to search for a string within another string.
The line of VBScript looks for the string MS IE within the HTTP_USER_AGENT server variable. If
the string is found, the requesting browser is Internet Explorer. If the browser is not Internet
Explorer, the same test can be applied to determine if the user has a Netscape browser.

Using this capability to determine the browser type, you can customize your script to deliver
VBScript only to Internet Explorer browser. However, this might not always be the best ap-
proach to solving your client-side scripting needs and depends on the target user base of your
application. If you are building applications specifically for Internet Explorer, a VBScript envi-
ronment can help you leverage your in-house Visual Basic skills. If you are building applica-
tions for an HTML 3.0 environment, there is no need to build two different client-scripting
versions because both Internet Explorer and Netscape Navigator support JavaScript. If you are
building applications to run at a lower compliance level, such as in text-based browsers, valida-
tion on the server is your only option.

You can also use the Browser Capability Component to detect the specific browser type and the
available features of the specific browser. The Browser Capability Component is one of the
various installable components that ship with the Internet Information Server. The Browser
Capability Component uses the HTTP_USER_AGENT server variable to look up the capabilities of
the different browser types in an INI file. For a more detailed discussion of the Browser Capa-
bility Component, see Chapter 17, “Implementing Existing Server-Side Components.”

From Here…
In this chapter, you discovered how to use VBScript to implement execution logic to run on the
ActiveX Server and within the Internet Explorer browser. VBScript provides an excellent devel-
opment tool that enables you to test variable datatypes and control program flow by using
operators and conditional statements for both server-side and client-side processing. In addi-
tion, the need to offload processing power onto the browser client helps to reduce the process-
ing requirement of the server and reduce network bandwidth.
For related information, please see the following chapters:

■ Chapter 10, “The Composition of an Active Server Application,” will illustrate how six
built-in ASP objects collectively form an ASP application.
■ Chapter 12, “Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object,” demonstrates how the Server object
controls and manages ASP applications.

11.1389-6 CH08 159 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 160 Chapter 8 Putting It All Together with VBScript

■ Chapter 18, “Creating Your Own Active Server Components,” will teach how to create
your own ActiveX components to meet the needs of your applications.
■ Chapter 16, “Putting It All Together: Building an Online Catalog,” demonstrates how to
use the existing components, including the Browser Capability Component, that ship
with IIS 4.0.
■ Chapter 20, “Database Access with ADO,” provides a firsthand approach on using
VBScript to manipulate your databases.

11.1389-6 CH08 160 10/26/98, 9:13 AM

B6 SEU Active Server Pages #1389-6 mary ellen 9/8/98 CH08 lp#1
 9
161

C H A P T E R

Part
Using the Script Debugger II
Using the Script Learn how the Script Debugger can
be activated to process server–side
Ch

9
Debugger ASP scripts and client-side VBScript
and JScript.

Trapping and Debugging

Scripts
Discover how to set breakpoints to
systematically step through the
code to check and verify the script’s
The Microsoft Script Debugger 1.0 finally provides an in- variable values and properties.
line debugging tool that has been missing from the family
Identifying Errors
of scripting programming tools. This logical extension is
necessary to start transforming scripting languages into The Script Debugger helps mini-
mize the runtime errors, syntax
more mature development environments. The Script
errors, and logical errors that might
Debugger provides a debugging capability for VBScript
plague your scripting applications.
and JScript, and can also be used to debug Java programs.
This flexibility enables the Scripting Debugger to control,
trap, and change variables in both server-side and client-
side scripts. Server-side scripts embedded in Active
Server Pages are trapped and processed on the Internet
Information Server before the results are sent to the re-
questing browser. Client-side scripts are embedded within
the HTML on the browser client. ■

12.1389-6 CH09 161 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 162 Chapter 9 Using the Script Debugger

Using the Script Debugger

The Script Debugger is installed on the client’s machine to perform server- and client-side
debugging capabilities with the Internet Explorer browser. The Script Debugger built into
the Internet Explorer 4.0 can be started as a stand-alone application or can be activated in
programmatic ways. Figure 9.1 displays a client machine debugging a server-side ASP script.

FIG. 9.1
The Script Debugger
debugs client and
server VBScript, JScript,
and Java programs.

The focus of this book has been primarily on Active Server Pages running on the Internet
Information Server, but the Script Debugger can also debug client-side scripts. To debug
server-side scripts, the debugging property must be activated for the ASP application. The
Internet Information Server 4.0 enables you to control the ASP application settings using the
Microsoft Management Console, as shown in Figure 9.2.

To debug Java applications, the Microsoft Script Debugger can only be used with the Microsoft
Java Virtual Machine (VM) version 2174 or later. The ActiveX-enabled Java VM is shipped with
the Internet Explorer 4.0 and the Internet Information Server 2.0. To enable Java debugging in
the Microsoft Script Debugger, select the option to enable Java debugging from the Options,
View menu. Then select the root path for the Java source files in the Source Path location.

12.1389-6 CH09 162 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Using the Script Debugger 163

FIG. 9.2
Activate server-side
ASP script debugging
from the Microsoft
Management Console.

Part
II
Ch

To activate script debugging capabilities for server-side scripts, right-click the ASP application
within the Microsoft Management Console and select Property. Then click the Configuration
button in the Application Settings section of the Virtual Directory tab. If only the Create button
is available from the Application Settings section, create an ASP application by clicking the
Create button. Clicking the Configuration button exposes the various properties of the ASP
application, such as ASP application caching and programmatic settings. For more information
on using the MMC to configure your ASP applications, please see Chapter 31, “Using Internet
Information Server 4.0 to Manage Your Web Applications.”

After Script Debugger permissions have been set on the Server, the Script Debugger can be
activated manually or automatically when server-side errors occur. The Script Debugger pro-
cesses the error via a dialog box, as shown in Figure 9.3.

The Script Debugger can also be activated programmatically from within HTML. To activate
the Script Debugger from VBScript, use the Stop keyword; if you are using JScript, use the
keyword Debugger. When the Internet Information Server processes the Stop or Debugger
request, the Script Debugger is started and program control is returned to the current line
(see Figure 9.4).

The Script Debugger can also be automatically activated when viewing the client source code
on the Internet Explorer browser. After the Script Debugger has been installed, select Source
from the View menu within the Internet Explorer browser.

12.1389-6 CH09 163 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 164 Chapter 9 Using the Script Debugger

FIG. 9.3
The Script Debugger
automatically detects
ASP scripting errors and
directs you to the
scripting error.

FIG. 9.4
The Script Debugger can
be programmatically
activated from within
the script.

12.1389-6 CH09 164 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Using the Script Debugger 165

Managing the Script Components

The Script Debugger consists of four windows, as shown in Figure 9.5. This group of windows
enables you to manage and view the script source code, control the script workflow, and deter-
mine and set variable and property values for your client and server scripting needs. The four
windows that make up the scripting wizard are the Script window, the Command window, the
Running Documents window, and the Call Stack window (see Table 9.1).
Part
FIG. 9.5 II
The Script Debugger
mimics the functionality Ch
and look and feel of
Visual Basic to provide
9
script debugging.

Script window

Running Documents
window

Call Stack window

Command window

Table 9.1 The Script Debugger Windows

Window Name Purpose
Script window Displays the script source code.
Command window Sets and retrieves variable and property values.
Running Documents window Manages the files that can be debugged.
Call Stack window Traces the procedures and functions within the script.

The Script window is primarily responsible for displaying the source code. The Script window
uses syntax coloration, images, and highlighting to indicate scripting syntax, breakpoints, and
the current line the script is processing. The Command window is used to set and retrieve
object values within the executing script page. The Running Documents window manages
multiple script windows in a hierarchical tree-view format. The Call Stack window is used to
manage and track procedure calls between subroutines and procedures within a page.

12.1389-6 CH09 165 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 166 Chapter 9 Using the Script Debugger

Keep in mind that the scripts displayed in the Script Debugger are activated on a read-only
basis. In order to manipulate the script code, a new script document must be created from
within the Script Debugger and saved to the appropriate location. Once the new document is
created, you can copy the script code into the new document and have complete control to
manipulate that document as needed within the Script Debugger.

N O T E The source code in the Script window is read only. In order to have full read-write
permissions to a file, create a new script code from within the scripting wizard and copy the
script code into the new file for testing, editing, and debugging purposes. ■

The Script Debugger has a toolbar that provides shortcuts to a variety of debugging features.
Table 9.2 illustrates the toolbar icons with a brief description.

Table 9.2 The Scripting Debugger Debug Toolbar

Icon Description
Starts or continues execution of the current script.

Stops debugging the current script.

Breaks into Debugger at the next statement.

Step into next statement.

Step over next statement.

Step out of current execution function.

Insert or remove breakpoint at the current line.

Clear all breakpoints.

Display the Running Documents window.

Display the Command Input/Output window.

Display the current Calling Stack window.

12.1389-6 CH09 166 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Using the Script Debugger 167

Of all the different ways to start the Script Debugger, the easiest method is to use the Running
Documents window. The Running Documents window contains a list of pages that have been
loaded into the browser for ASP applications that have server-side scripting capabilities turned
on. You can use this listing to open documents in the Script window to start debugging by
double-clicking the page. After the page is loaded, you can set breakpoints within your script to
help control where the Debugger will stop processing the script. To start debugging, simply
reload or refresh the page in the browser. Alternately, you can also directly start debugging a Part
page by right-clicking the document in the Running Document window and selecting the Break
At Next Statement option, as shown in Figure 9.6. II
Ch
FIG. 9.6 9
Use the Running
Documents window to
start the debugging
process.

Although selecting this action does not immediately bring up the script code in the Script win-
dow, when the page is refreshed in the browser, the code debugger “breaks” in to the normal
execution and starts stepping through the first line of script in the Script Debugger environ-
ment. This debugging technique is what is displayed in Figure 9.6.

Controlling Script Flow

The majority of script debugging effort involves setting breakpoints in the code to temporarily
pause the script processing. This temporary suspension of processing enables you to check
and verify the values of properties or variables at that point in the script. Controlling the appli-
cation flow and monitoring application objects is accomplished through the use of breakpoints,
stepping mechanisms, and the Call Stack window.

12.1389-6 CH09 167 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 168 Chapter 9 Using the Script Debugger

Breakpoints and code stepping mechanisms provide the foundation of the Script Debugger.
Breakpoints are used to set locations within scripts where the execution of the script is tempo-
rarily stopped and program control is transferred to the Script Debugger. After the Script
Debugger has control of the script, different step mechanisms and program windows are used
to incrementally process the remaining script and to set and retrieve values. To set a
breakpoint, place your cursor on the line of code where you want the breakpoint to occur and
click the SetBreakpoint icon (refer to Table 9.2). The breakpoint is symbolized by the red
highlighting of the script line of code and the stop-sign shaped icon to the left of the Script
window, as shown in Figure 9.7.

FIG. 9.7
Use breakpoints to
transfer program
execution from the
processing client to the
Script Debugger.

The yellow arrow, shown at the left side of the screen in Figure 9.7, indicates the line the Script
Debugger is waiting to execute.

Notice that breakpoints can only be set on lines of code where code execution occurs. For
example, a breakpoint cannot be set on the first line of code in Figure 9.7, the <% Option
Explicit %>, or on lines where variables are declared using the Dim keyword. Furthermore,
breakpoints only apply to the first statement in a line of code. For example, in VBScript, mul-
tiple statements can be embedded on one line of code using the colon (:) as a separator (see
Figure 9.8).

In Figure 9.8, a breakpoint is placed on a code line that contains multiple statements separated
by colons (:). If the single step method is chosen, only the first statement on that line is ex-
ecuted, as you can see by viewing the values of the variables in the Command window in the
bottom of Figure 9.8. The value of variable vWarningStr is assigned User Profile, but the
remaining variable assignments for the variables vServerLoc and vProfile are unassigned.

12.1389-6 CH09 168 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Using the Script Debugger 169

FIG. 9.8
Breakpoints only apply
to the first statement in
a line of code.

Part
II
Ch

After program control is transferred to the Script Debugger, the Command window can be
used to set and retrieve information from within the script. (The Command window is similar
to the Debug window found in Visual Basic.) For example, Figure 9.1 is using the Command
window to determine the value of the Request(UpdateUser) object using VBScript syntax.
Notice that you are not limited to extracting only ASP components, you can access any object
or variable on that page. The Command window will accept any script command supported by
the scripting language that is interpreting the current page. To determine any property on a
page using VBScript, use a question mark (?) followed by the object, or use the Debug.Write
method followed by the object name. For example, to determine the document title, use the
following syntax:
? Document.Title

Alternatively, you can use the Write method to display a procedure level variable named
iCounter:

Debug.Write iCounter

If you want to determine a variable or property value using JScript, simply type in the variable
name in the Command window.

You can also assign variables from within the Command window using the same naming rules
that apply for your scripting language. For example, to assign the variable iCounter the value
of 3 in VBScript, the following line of code would be entered into the Command window:
Icounter = 3

12.1389-6 CH09 169 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 170 Chapter 9 Using the Script Debugger

When the Script Debugger reaches a breakpoint, all data transfer to the requesting client is
stopped and is explicitly controlled by the Script Debugger. This means that if a breakpoint is
located in the middle of a Web page, HTML is delivered to the browser up until the breakpoint
is processed. The Script Debugger now regulates the flow of the remaining data.

The Script Debugger has three stepping mechanisms to control how to execute selected por-
tions of the script: Step Into Next Statement, Step Over Next Statement, and Step Out of Cur-
rent Execution Function. The Step Into Next Statement is used to process the next line of
script. The Step Over Next Statement is used to skip over calls to functions and subroutines.
The current procedure is still executed, but the current line pointer does not enter the called
procedure statement. This enables you to determine the result of a function without having to
step through the function’s code details. Finally, the Step Out statement enables the remaining
script within a procedure to be executed and debugging control returned to the line preceding
the calling procedure.

These different navigation techniques can be applied by using the Script Debugger toolbar
buttons shown in Table 9.2.

To fully understand the different debugging step-through methods, consider the ASP debug-
ging session shown in Figure 9.9.

FIG. 9.9
The Script Debugger
offers different methods
to step through the
scripting code.

If the Step Into Next Statement method is used, the Script Debugger would start to evaluate
the IsWholeNum function. As the Script Debugger starts to evaluate the function, the program
control is evaluating the first line of the IsWholeNum function, as shown in Figure 9.10. Notice
that the Call Stack window identifies that the IsWholeNum procedure is being called.

12.1389-6 CH09 170 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Using the Script Debugger 171

FIG. 9.10
The Step Into Next
statement method
sequentially processes
script one line at a
time.

Part
II
Ch

Consequently, if the Step Out method is selected from the toolbar, the Script Debugger would
process the remaining script within the function and return control to the next line following
the calling parent function, as shown in Figure 9.11.

FIG. 9.11
The Step Out method
finishes processing the
function and returns
control to the
statement after the
parent calling function.

12.1389-6 CH09 171 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 172 Chapter 9 Using the Script Debugger

If the Step Over Next Statement stepping method is initially selected from the starting point
displayed in Figure 9.10, the function is calculated and the Script Debugger returns program-
matic control to the appropriate If...Then condition, as shown in Figure 9.12.

FIG. 9.12
The Step Over Next
statement enables you
to see the results of a
function or subroutine
without having to step
through the procedure
itself.

The Call Stack window is used to track the script procedures and events that have been loaded.
The Call Stack contents are arranged historically with the most recent scripts located at the top
of the list (refer to Figure 9.11). This window is particularly useful when you need to trace a
series of nested procedures within an application. It enables you to quickly navigate from pro-
cedure to procedure without having to incrementally follow every line of executing code. Keep
in mind that the Call Stack window does not display the values of variables within the proce-
dure, just the name of the currently executing procedure. If you want to retrieve and set vari-
able and property values for the current procedure, use the Command window.

Types of Errors
Unfortunately, the Script Debugger is still incapable of helping application developers create
bug free code. Regardless of the result of scripting errors, the Script Debugger helps minimize
the debugging runtime errors, syntax errors, and logical errors.

12.1389-6 CH09 172 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 Types of Errors 173

Runtime Errors
Runtime errors are a result of trying to process a line of code that is impossible to complete or
are the result of using incorrect script syntax. In most situations, runtime errors are inadvert-
ently overlooked as a result of not considering all the possible situations for a conditional state-
ment. For example, consider the following statement:
Fuel_Economy = (Miles_Traveled) / (Gallons_Used)
Part
In most situations, this calculation would successfully return the calculated fuel economy of a II
vehicle. However, what happens if the variable Gallons_Used is zero? The resulting division by
zero is an invalid operation and would generate a runtime error (see Figure 9.13). Ch

9
FIG. 9.13
Runtime errors are
the result of trying to
execute code that is
impossible to process.

To prevent most runtime errors from occurring, familiarize yourself with the situations where
these errors might occur. Then you can trap or construct code to prevent these runtime errors
from occurring. For instance, in the previous example to determine fuel economy, first ensure
that the variable Gallons_used is a non-zero, real number before you try the division operation.

Syntax Errors
Syntax errors are a direct result of the improper use of the language’s object model. Each lan-
guage has a specific set of explicit instructions that are needed for the compiler to process the
embedded application logic. If the script syntax is incorrect, the compiler does not know how
to process the request and will generate a runtime error. For example, consider the following
faulty statement:
Response.Write FormatDateTime(“Now”)

12.1389-6 CH09 173 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 174 Chapter 9 Using the Script Debugger

When the script engine processes this request, the FormatDateTime function is expecting a
date/time variable subtype. However, in this situation, the application developer inadvertently
put quotes around the Now keyword. As a result of trying to process a string datatype, the
FormatDateTime function will generate an error, as shown in Figure 9.14.

FIG. 9.14
Syntax errors are
generated when the
compiler attempts to
compile improper code
statements.

The Script Debugger currently does not scan your scripts for incorrect script syntax to warn
you of potential runtime errors. However, the Script Debugger does identify the specific line of
script code where the runtime error occurs.

Logical Errors
Logical errors are often the most frustrating errors to trap. Logical errors are the direct result
of code not executing as the application developer intended. The dilemma of developing code is
that the compiler does exactly what the code tells it to do, nothing more, nothing less. If there
is a gap between how the application is intended to behave and how it actually behaves, logical
errors will result. There are no bullet-proof methods to prevent logical errors from occurring;
only testing, analysis, and experience can prevent faulty application logic.

12.1389-6 CH09 174 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 From Here… 175

From Here…
This chapter showed that the Script Debugger is a valuable resource in testing and debugging
your server- and client-side scripts. Thin-client development tools are slowly maturing into the
ranks the other traditional client-server integrated development environments have enjoyed for
years. The Script Debugger provides inline debugging capabilities for VBScript, JScript and
Java. The Script Debugger mimics the debugging functionality found in Visual Basic, which
Part
enables:
II
■ Color syntax highlighting to identify code modules via the Script windows.
Ch
■ Setting and removing multiple breakpoints within a page.
■ Step Into, Step Over, and Step Out of navigation methods. 9
■ Direct retrieval and setting of variable and property values via the Command window.
■ Easy page and procedural navigation with the Running Document and Call Stack
windows.

■ Chapter 10, “The Composition of an Active Server Application,” will illustrate the various
components of an ASP application.
■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of
tracking a user through an ASP application.
■ Chapter 15, “Retrieving User Information: The Request Object,” demonstrates how to
collect information from the user using forms, hidden fields, and cookies.
■ Chapter 16, “Putting It All Together: Building an Online Catalog,” demonstrates how to
use the existing components, including the Browser Capabilities Component, that ship
with IIS 4.0.
■ Chapters 17 and 18, “Implementing Existing Server-Side Components,” and “Creating
Your Own Active Server Components,” will teach you how to use server-side VBScript to
provide connection to pre-built and customizable servers.
■ Chapter 20, “Database Access with ADO,” provides a firsthand approach to using
VBScript to manipulate your databases.

12.1389-6 CH09 175 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 176 Chapter 9 Using the Script Debugger

12.1389-6 CH09 176 10/26/98, 9:14 AM

Brands3/stone3 SEU Active server Pages #1389-6 SRing 11/11/97 CH09 lp#3
 P A R T

III
Active Server Objects: The Essential
Building Blocks
10 The Composition of an Active Server Application 179

11 Controlling the Application: The Application Object 211

12 Controlling the Server with the Server Object and Using

Transaction Processing with the ObjectContext Object 229

13 Managing the User Session: The Session Object 247

14 Speaking to the Web Client: The Response Collection 267

15 Retrieving User Information: The Request Object 297

16 Putting It All Together: Building an Online Catalog 315

13.1389-6 Part III 177 10/26/98, 9:15 AM

Brands SEU Active Server Pages #1389-6 Part III SRing 10/28/97 lp#1
13.1389-6 Part III 178 10/26/98, 9:15 AM

Brands SEU Active Server Pages #1389-6 Part III SRing 10/28/97 lp#1
 10
179

C H A P T E R

The Active Server Page Appli-

The Composition of an cation Blueprint
Discover the text-based files, direc-

Active Server tories, and virtual directories and

scripting languages that interact to
Part
form Active Server Page applica-

Application tions. III

Ch
The Application Manager: The
GLOBAL.ASA file 10
Learn about the importance of
the GLOBAL.ASA file and how it
manages the Application and
Session objects and in what situa-
This chapter provides an overview of how the various sys- tions is it activated.
tems that form the present-day Web enterprise are identi-
fied and encapsulated into separate, manageable Active Understanding ASP Scoping
Server objects. These various objects each have their Issues
specific functionality but collectively interact to form the Discover how the ASP objects use
foundation of Active Server Pages. ■ four different scoping and lifetime
issues and in what situations should
they be applied.

The Fundamental ASP Server

Objects
Introduce yourself to the six ASP
objects that drive the functionality of
delivering ASP applications.

Debugging Active Server

Pages
Although no built-in debugger exists
for ASP, learn how to use the Err
object, the Registry settings, and
the FileSystem object to assist in the
debugging process.

14.1389-6 CH10 179 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 180 Chapter 10 The Composition of an Active Server Application

The Components of an Active Server Page

Application
An Active Server Page application is comprised of various items that together form the ASP
application. The collection consists of various text-based files, Server objects and components,
and ActiveX Server scripting, as shown in Figure 10.1.

FIG. 10.1 Server-Side Scripting

• VBScript, JavaScript
Client-Side Scripting

The eclectic collection • Perl, CGI, REXX

•VBScript, JavaScript
•ActiveX Controls
• Java, ISAPI
that forms Active Server • Java

applications.

Server Side Objects

Active Server Pages
• Built-in object that controls
the Active Server scripting engine

Server-Side Components
File System • Com-based objects
• Scalable, manageable, and
• Virtual, physical, and
distributable via MTS
root directories
• Controls permissions Text
•ASCII text represents
• Defines application content and forms HTML tags

Text Files
Active Server Page applications are really a collection of text files that are interpreted by the
Internet Information Server. These ASP scripts are composed of HTML tags, Active Scripting
code (using languages such as JScript and VBScript), and standard text, as shown in Listing 10.1.

Listing 10.1 An ASP file uses HTML tags, ASP scripts, and standard text.
<HTML> ‘=== Traditional HTML Tags
<HEAD> ‘=== Standard HTML
<TITLE>
Welcome ‘=== Standard Text
</TITLE>
</HEAD>
<BODY>
<% ‘=== Active Server Script
Sub WriteWelcomeMessage ()
Dim myGreeting
If Time >=#12:00:00 AM# And Time < #12:00:00 PM# Then
myGreeting = “Good Morning!”
Else
myGreeting = “Hello!”
End If
Response.Write mygreeting
End Sub
%>

14.1389-6 CH10 180 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 The Components of an Active Server Page Application 181

<% Call WriteWelcomeMessage %> ‘=== Executing the Active Server

subroutine
</BODY>
</HTML>

Traditional HTML files consist of only HTML tags for page layout and standard text to display
text. The HTML tags are used to create and display the content in the Web page. The differ-
ence between the HTML and ASP files is that ASP files simply expand the role of HTML tags
by adding scripting processing tags to control the HTML tags.

Keep in mind that pages with just HTML are a subset of ASP pages. This means that any tradi-
tional HTML-compliant Web page is already an Active Server Page, with one exception: the file
extension. To convert your existing HTML pages to Active Server Pages, simply rename the
HTM or HTML file with an ASP file extension. The Internet Information Server will see the file Part
extension and automatically interpret the ASP file.
III
N O T E The ActiveX scripting engine will process all pages with an .ASP file extension. Therefore, to Ch
maximize server performance, only convert HTM or HTML files to ASP files when scripting
functionality is needed for the page. Otherwise, unnecessary resources will be consumed when
10
processing the ASP file for namesake only. ■

Furthermore, the ASP and the IIS Web server do not have support issues with HTML versions
you want to process and deliver to your sites and applications. Because the server doesn’t view
the actual HTML embedded in the ASP files, the delivered HTML version is limited only by the
HTML/browser level of your targeted user base. You can create your Active Server scripts to
sense what type of browser is visiting your site and send the appropriate HTML version accord-
ing to requesting browser’s capabilities. Chapters 12, “Controlling the Server with the Server
Object and Using Transaction Processing with the ObjectContext Object,” and 17, “Imple-
menting Existing Server-Side Components,” illustrate how to use the HTTP server variable and
the Browser Capabilities Component to determine the type of browser visiting your site.

Physical, Root, and Virtual Directories

Your ASP application is simply a collection of text files located in a directory or its
subdirectories residing on the Web server. The base directory for your ASP application is
referred to as a virtual root. The virtual root is what is referenced by the Web server as the
container that holds all the files needed for the ASP application. Virtual directories are also
used by the IIS Web server as file containers. Virtual directories help prevent you from having
to know the exact file paths of files and to prevent users from obtaining more information about
the Web server’s underlying sub-systems.

An example of a virtual root is VBScript in the URL www.microsoft.com/VBScript. The

VBScript virtual directory is use by the Internet Information Server to map this virtual path to
a physical directory location. The virtual directory settings are managed by the Internet Ser-
vice Manager. The Internet Service Manager manages the physical drive mappings with virtual
roots, permissions, and logging. Figure 10.2 illustrates the role of physical and virtual directory
structures.

14.1389-6 CH10 181 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 182 Chapter 10 The Composition of an Active Server Application

FIG. 10.2 www root

Internet Service
Manager regulates the
relationship between Directory 2
virtual and physical Directory 1
Directory 3
directory settings.
Virtual Directory / Application

Sub Directory 1
Sub Directory 2

Virtual Directory / Application

Sub Directory 1 Sub Directory 2

ASP Scripting Languages and HTML

The heart of Active Server Pages programming is the processing of server-side scripts. Server-
side scripting provides the capability to manipulate variables and application flow through
procedure-based coding, connect to a variety of external sources to retrieve information, and
control output to the Web browser. Part II of this book covered the foundations of VBScript, the
default scripting language of the Internet Information Server. However, you can also use
JScript, REXX, PERL, or any other script-compliant language that can be processed by the
server. The scripting language, now processed by both the Web server and Web client, acts as
an interpreter to execute the text-based Internet protocol. To set the scripting language for an
Active Server Page, use the LANGUAGE directive:
<%@ LANGUAGE = ScriptingLanguage RUNAT = Location %>

where ScriptingLanguage is the primary scripting language for that page and Location is the
location of script interpretation.

An ASP directive tells the IIS Web server how to interpret the individual ASP file. The LANGUAGE
directive is used to set the proper interpreter for your primary scripting language for the page.
It is important to note that ASP pages can use a variety of scripting languages throughout your
ASP application that are different from the primary scripting language. Active Server applica-
tions can be comprised of pages written for several scripting languages. To use VBScript as the
primary scripting language for a given ASP page, use the LANGUAGE directive in the first line of
the Active Server Page.
<%@ LANGUAGE = “VBScript” RUNAT = “Server”%>

CAUTION
If no space is placed between the directive symbol (@)and the LANGUAGE attribute, the server will generate
a ‘Missing Language attribute’ error.

The LANGUAGE directive is just one of the directives available when using ASP. The various ASP
directives will be covered in greater depth later in this chapter.

14.1389-6 CH10 182 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 The Components of an Active Server Page Application 183

Remember, the primary scripting language determines the language that will be used to pro-
cess the Active Server script contained within the <% %> delimiters on your pages. If you want
to change the default scripting language for your Web server, you have to change the
DefualtScriptLanguage key within the Registry in IIS 3.0 or use the Microsoft Management
Console in II 4.0, as shown in Figure 10.3. For more information on using the MMC to config-
ure the IIS 4.0 and Active Server settings, please see Chapter 31, “Using Internet Information
Server 4.0 to Manage Your Web Applications.”

FIG. 10.3
The Microsoft
Management Console
regulates the HTTP and
ASP settings for IIS 4.0.
Part
III
Ch

10

By default, VBScript is the default scripting language when you install IIS. Therefore, explicitly
setting the primary scripting language on each page in your application prevents errors that
might otherwise occur if the host Web server has a different default scripting language de-
fined. In other words, don’t assume that every Internet Information Server uses VBScript as its
default scripting language.

The <% %> tags enable the Internet Information Server to differentiate between HTML and
Active Script functions located between the Active Server delimiters. The Active Server delimit-
ers can be placed either inline or within delimited code blocks. The following example demon-
strates combining HTML and inline Active Server script to write the name of the user to the
page and the estimated shipping date.
<Center>Thank you, <% Response.Write Session(“UserName”) %>, for your
order.</Center><Br>
<Center>Your order should arrive by <% Response.Write FormatDateTime(Now + 7,
vbShortDate) %></Center>

The first line of the previous code example tells the Internet Information Server to write a
value to a page by using the <% %> tags. In this example, the username has been stored in a
session variable named “UserName”. Therefore, the expression <% Response.Write
Session(“UserName”) %> will write the username onto the Web page. The second line of code
uses VBScript functions, covered in Part II, “VBScript: The Foundation of Active Server
Pages,” to add a week to the current date. After the estimated shipping date is determined, the
Response object is used to write the shipping date onto the Web page.

14.1389-6 CH10 183 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 184 Chapter 10 The Composition of an Active Server Application

Listing 10.2 demonstrates a code block that is used to contain a subroutine called GenerateMsg.

Listing 10.2 Using delimiters to execute blocks of code.

<%@ LANGUAGE=”VBSCRIPT”%>

<%
‘============== Begin Subroutine Modules ==============
Sub GenerateMsg()
Response.Write “You have mail!” ‘=== Writes message to page
End Sub
‘================ End Subroutine Modules ==============
%>

<HTML>
<BODY>
<%
Call GenerateMsg() ‘=== Call the subroutine GenerateMsg
%>
</BODY>
</HTML>

Remember, the <% %> tags are used to indicate Active Server scripts. The Active Server script
can use any script-compliant language (such as VBScript, JScript, PERL, or REXX) within these
tags to manipulate variables, process logic, and make database connections.

The Active Server Page Project File: GLOBAL.ASA

As you start building ASP applications, the GLOBAL.ASA file will play a central role in managing
them. The .ASA file extension is an acronym for Active Server Applications. The GLOBAL.ASA
file enables the management of two critical objects used to track state information in an ASP
application. The advent of Active Server Pages gives developers the ability to track user infor-
mation from page to page within a Web-based application. Remember, the Hypertext Transfer
Protocol (HTTP) is a connectionless or stateless protocol. Users connect to a Web server,
request a page, receive a page, and then disconnect. In the past, there was no way to track a
user’s activity as he jumped from page to page. ASP helps to bring state and session tracking to
the Web for cookie-compliant browsers and networks. For more information about tracking
state, see Chapter 13, “Managing the User Session: The Session Object.”

This tracking capability is essential to implement any Web-based workflow application. When
an application is first started, the GLOBAL.ASA file is loaded into memory. This file is used to
track application events and individual user events, variables, and objects. The GLOBAL.ASA file
is an optional file located in the root directory of any ASP application. An example of a
GLOBAL.ASA file is shown in the following code.

14.1389-6 CH10 184 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 The Active Server Page Project File: GLOBAL.ASA 185

<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

‘====================== Session and Application Procedures ======================
Sub Session_OnStart ‘=== Session Level events and object placed here
End Sub

Sub Session_OnEnd ‘=== Close session event code is placed here

End Sub

Sub Application_OnStart ‘=== Application level variables used here

End Sub

Sub Application_OnEnd ‘=== Close any application objects here

End Sub
‘=================================================================================

‘====================== Session and Application Events ==========================

Session_OnStart ‘=== Runs the first time a user runs any page in Part
your application
Session_OnEnd ‘=== Runs when a user’s session times out or
III
quits your application Ch
Application_OnStart ‘=== Runs once when the first page of your
application is run 10
‘=== for the first time by any user
Application_OnEnd ‘=== Runs once when the Web server shuts down
‘================================================================================
</SCRIPT>

TIP In previous versions of Active Server Pages, the GLOBAL.ASA file was named GLOBAL.ASP.

The GLOBAL.ASA file tracks the OnStart and OnEnd events of the Application and Session
objects, as demonstrated in Figure 10.4. The Application_OnStart and Application_OnEnd
are used to monitor all Application objects that are available to all users within the ASP appli-
cation. For more information on the Application object, please see Chapter 11, “Controlling
the Application: The Application Object.”

The Session_OnStart and Session_OnEnd events are used to manage the Session objects. The
Session objects are used to track individual user information. However, unlike the Application
object, the Session object cannot share information between other user sessions; it can only be
used to pass information from page to page. It is also important to point out that you cannot de-
clare a variable outside a <SCRIPT> tag within the GLOBAL.ASA file. Trying to use any text outside
the <SCRIPT> tag, with the exception of the <OBJECT> tag, will generate a runtime error. The
specific details of the Session object will be fully discussed in Chapter 13.

14.1389-6 CH10 185 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 186 Chapter 10 The Composition of an Active Server Application

FIG. 10.4
Global ASA File
The GLOBAL.ASA
file manages
Application and
Application Events
Session objects.
• on start
• on end

Session Events
•on start
• on end

Tracking Application Events

In most traditional application environments, a user must take some action to launch an appli-
cation. Usually a command is issued from the keyboard or a command button is pressed in a
graphical environment. As application development swings back to the server-centric deploy-
ment model, the start of an application is not necessarily connected to a specific user action.
For example, assume a proposal letter generation system is installed on a user’s desktop. When
the user needs to generate a new proposal letter, the application is started, the user uses the
application, and when finished, the user shuts down the application and exits the system. In a
server-centric world, such as database servers or Web server environments, these application
services may already be running when the user decides to access them. The user can use the
application and, when he is finished, leave the application, yet the application process is not
terminated on the server. This separation between application services and user services is the
underlying function of the GLOBAL.ASA file.

The GLOBAL.ASA file is an optional file that is used to manage the Application object. The
Application object has two events: the Application_OnStart event and the
Application_OnEnd event. The Application_OnStart event is triggered the first time any page
in application is called. The Application_OnEnd event occurs when an application is closed
down. The Application shutdown event occurs when the Internet Information Server needs to
recompile and reload the GLOBAL.ASA file back into memory. The Application object also lets
you lock and unlock Application object properties to ensure that values don’t inadvertently
get changed. The Application object will be discussed in much greater detail in Chapter 11.

TIP The GLOBAL.ASA file is an optional file that does not have to be used when creating ASP applications.
If the file exists, the Application and Session events can be processed accordingly. If the
GLOBAL.ASA file does not exist, the corresponding events are not processed and the Application
and Session events cannot be trapped.

14.1389-6 CH10 186 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 The Active Server Page Project File: GLOBAL.ASA 187

Tracking Session Events

The Session object is used to track and manage individual user sessions in an application.
The Session object has two events: the Session_OnStart event and the Session_OnEnd
event. Session_OnStart is triggered when a new user enters the application. When the
Session_OnStart event is triggered, a unique ID is assigned to the user and is used to monitor
the user within the ASP application. However, the Session_OnEnd event is not so well defined.
The Session_OnEnd event usually occurs when a timeout value has expired on the Web server
or the session is explicitly ended in code.

Tracking and monitoring an individual user session is one of the biggest challenges in building
ASP applications because the Web uses the Hypertext Transfer Protocol (HTTP). The HTTP
protocol is described as stateless because it is used to initiate a request from an HTTP client
and deliver the response from the server. After the request has been fulfilled, the connection Part
between the client and server is eliminated. Although this model may be ideally suited for
simple request processing systems, the lack of client/server concurrency plagues application III
development via HTTP. Ch

Unlike the application events, objects and variables that are activated in the session event are 10
localized only to that user session and cannot be referenced by another user session. But the
user session information can be passed from page to page within the application. If variables
need to be shared across individual sessions, the application-level object would have to be
used.

Activating the GLOBAL.ASA File

The GLOBAL.ASA file is primarily accessed based on session-level events, and is called in three
situations:

■ When the Application OnStart or OnEnd event is triggered.

■ When a Session_OnStart or Session_OnEnd event is triggered.
■ When a reference is made to an object that has to be instantiated in the GLOBAL.ASA file.
If the GLOBAL.ASA file is used, it is implicitly called in one of three ways. First, assuming that the
ASP application is not currently running, a request is made from a Web browser to the Internet
Information Server for an ASP page. The Internet Information Server in turn looks for the
requested ASP page in a virtual directory. The Web server associates the virtual directory
representing an ASP application, and searches for an existing Application object created from
the requested ASP application. If no Application objects exist, the Web server tries to locate
the GLOBAL.ASA file in the application’s root directory. If the GLOBAL.ASA file is found, the
Application_OnStart and Session_OnStart events are sequentially triggered to create an
Application and Session object. After the Session object is created, the user is assigned
unique session identification.

The second situation in which the GLOBAL.ASA file is used is when a new user accesses
an already running application and the user does not have an application session identifier.

14.1389-6 CH10 187 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 188 Chapter 10 The Composition of an Active Server Application

As mentioned above, when a user makes a call to an ASP application, a unique session ID is
generated to identify and track the user. If no session identification is defined for the current
application, the GLOBAL.ASA file is used to generate the unique ID on the Session_OnStart
event.

The third situation in which the GLOBAL.ASA file is activated is when a calling Active Server
Page references an object declared in the GLOBAL.ASA file. You can create an application- or
session-level object by using the <OBJECT> tag in the GLOBAL.ASA file, as shown below. It is
important to point out that the <OBJECT> tags must be placed outside the <SCRIPT> tags to
prevent errors from occurring.
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

‘===================== Session and Application Procedures =======================

Sub Session_OnStart ‘=== Session Level events and object placed here
End Sub

Sub Application_OnStart ‘=== Application Level events and object placed here
End Sub
‘=================================================================================

Session_OnStart
‘=== Session_OnEnd
‘=== Application_OnStart
‘=== Application_OnEnd
‘================================================================================

</SCRIPT>

<OBJECT RUNAT=Server SCOPE=Application ID=MyDBConnection

PROGID=”ADODB.Connection”>
</OBJECT>

In the preceding example, even though the global or application-level database connection is
declared in the GLOBAL.ASA file, the object itself is not created until it is referenced. This late
binding feature helps to conserve resources by creating the objects only when they are
needed. For example, the ActiveX Data Object (ADO) declared in the GLOBAL.ASA file in the
previous example can be referenced by calling any of its methods from any ASP within the
application.
<% Response.Write MyDBConnection.Open(“myDSNConnection”) %>

The details of using the ADO and other objects that ship with the Internet Information Server
will be covered in greater depth in Part IV, “Active Server Components.”

N O T E To prevent runtime errors, the <OBJECT> tag used to declare application-level objects
must be outside the <SCRIPT> tags in the GLOBAL.ASA file. ■

14.1389-6 CH10 188 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Understanding Scoping Issues in Active Server Applications 189

Upon finding the GLOBAL.ASA file and determining that the Application object does not exist
for that project, the Application_OnStart event is executed. After the Application_OnStart
event is complete, the Session_OnStart event is triggered because this is a new user to the
application.

Therefore, the order of execution in which an ASP application performs is as follows:

1. GLOBAL.ASA
2. Application_OnStart
3. Session_OnStart
4. Session_OnEnd
5. Application_OnEnd

One of the most common misconceptions about the GLOBAL.ASA file is that it can be used as a Part
library for commonly used functions and subroutine procedures. The GLOBAL.ASA file can only III
be used to create references to objects and trap the start and ending events of the Application
Ch
and Session objects. The best way to share code modules across Active Server Pages is to use
Server Side Includes (SSI). SSI provides a convenient means to extract data (functions, subrou- 10
tines, and so on) from a source file and place it into an Active Server Page.

One of the drawbacks of the stateless request process using HTTP is that once the GLOBAL.ASA
file is loaded into memory for a user session, any consequential change to the GLOBAL.ASA file
is not implemented until the user starts a new user session. If you decide to make changes to
the GLOBAL.ASA file, IIS will start the process to reset the application. First, the HTTP service
stops accepting any new requests for pages and continues to process any requests that are
waiting in the queue. After the HTTP service is finished with the requests, the Internet Infor-
mation Server closes all sessions by executing the Session_OnEnd event. After the Session_
OnEnd event is finished, IIS closes the application by executing the Application_OnEnd event.
At this point, the GLOBAL.ASA file is recompiled and reloaded into memory. However, during
that time, any additional HTTP requests made to the Web server will generate an error mes-
sage stating “The request cannot be processed while the application is being
restarted.” or “Create object failed”. After the recompiled GLOBAL.ASA file is reloaded,
new users will be able to access the changed file.

Understanding Scoping Issues in Active Server

Applications
As you discovered in Part II of this book, variables in VBScript have a defined lifetime or scope.
Often these scoping issues are related to how you declare the variable or at what level you
declare the variable. A similar situation exists when using variables in ASP applications.

14.1389-6 CH10 189 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 190 Chapter 10 The Composition of an Active Server Application

GLOBAL.ASA Scoping Level

If it is used, the GLOBAL.ASA file represents the highest level available to set application- and
session-level variables. A variable’s scope depends on where the variable is declared. However,
just because you don’t declare a variable within the application and session OnStart events
does not mean you can’t create application- and session-level variables later in your ASP appli-
cation.

N O T E Any text not included within the <SCRIPT> tags in the GLOBAL.ASA file will produce a
syntax error. ■

The GLOBAL.ASA file can use multiple <SCRIPT> tags and multiple supported scripting lan-
guages. However, all text in the GLOBAL.ASA file must be contained within the <SCRIPT> tags.
The following two code examples demonstrate using VBScript and JScript in the GLOBAL.ASA
file.
<SCRIPT LANGUAGE = “VBScript” RUNAT = “Server”>
Sub Session_OnStart
‘== Data Connection Variable
Session(“ConnectionString”) = “DSN=CWI”
Session(“ConnectionTimeout”) = 15
Session(“CommandTimeout”) = 30
Session(“RuntimeUserName”) = “”
Session(“RuntimePassword”) = “”
End Sub
</SCRIPT>

This can also be illustrated using JScript within the GLOBAL.ASA file:
<SCRIPT LANGUAGE = “JScript” RUNAT = “Server”>
Function Session_OnStart() {
//== Data Connection Variables
Session(“ConnectionString”) = “ DSN = CWI”
Session(“ConnectionTimeout”) = 15
Session(“CommandTimeout”) = 30
Session(“RuntimeUserName”) = “”
Session(“RuntimePassword”) = “”
}
</SCRIPT>

These examples are both triggered when a new Session object has been created. When the
Session_OnStart event is processed, as shown in the preceding ASP scripts, session variables
are set to help open database connections as the user walks through the application.

Application-Level Scope
In Active Server Pages, a global variable is now referred to as an application-level variable.
Application-level variables are global in the sense that they are accessible across multiple user
sessions. To declare an application-level variable that can be shared across the entire applica-
tion with multiple users, you use the Application_OnStart event.

14.1389-6 CH10 190 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Understanding Scoping Issues in Active Server Applications 191

Application-level variables often include commonly referenced items that are needed through-
out your application. For example, suppose your ASP application includes a drop-down box
whose values are drawn from a database. It is more efficient to make the database query one
time and store the values in an application-level variable than it is to query the database for
every user. The application-level variable can then be referenced by individual user sessions to
minimize database input/output. However, if the values in the drop-down box change fre-
quently, the application-level variable might not contain the correct data because the applica-
tion variable is only refreshed when the ASP application is restarted.

Chapter 11 illustrates several examples that demonstrate how to use application variables. The
following code example illustrates loading database connection variables as application vari-
ables.
<SCRIPT LANGUAGE = “VBScript” RUNAT = “Server”>
Sub Application_OnStart Part
‘== Data Connection Variable
Application (“ConnectionString”) = “DSN = CWI”
III
Application (“ConnectionTimeout”) = 15 Ch
Application (“CommandTimeout”) = 30
Application (“RuntimeUserName”) = “” 10
Application (“RuntimePassword”) = “”
End Sub
</SCRIPT>

Once this script is loaded in the GLOBAL.ASA file, all users of the ASP application can reference
these variables whenever a database connection is needed.

Session-Level Scope
If you need to create user-level variables, you can declare the variables in the Session_OnStart
event in the GLOBAL.ASA file. This is commonly done when a database connection is needed in
the application. The actual database connection, in most situations, should not be made in the
Session_OnStart event because you want to utilize the database connection, thread, and re-
source pooling available with ODBC 3.0 and the Microsoft Transaction Server.

TIP To increase your ASP application performance, establish and release database connections as needed
to utilize ODBC 3.0’s database connection, threading, and pooling features.

Chapter 16, “Putting It All Together: Building an Online Catalog,” demonstrates using applica-
tion- and session-level variables to create the Amazing DataWidgets WareHouse.

Procedure-Level Scope
Procedure-level variables are those variables that are declared and used on individual ASP
pages. These variables are subject to the limitations defined by the scripting language respon-
sible for processing the ASP. Variables in VBScript can exist at two levels: either at the script or
procedure levels. (Refer to Chapter 3 for a detailed discussion of variable scoping.)

14.1389-6 CH10 191 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 192 Chapter 10 The Composition of an Active Server Application

Unlike the objects declared in the GLOBAL.ASA file, the Application_OnStart,

Application_OnEnd, Session_OnStart, and Session_OnEnd events cannot be called from any
ASP within the application. If you try to call those events, the ASP page thinks that you are
trying to reference an undefined procedure call on the current page and generates an error.

The Fundamental Active Server Page Objects

Active Server Pages derive their rich functionality from the use of six inherent objects that do
have to be instantiated. These objects have their own predefined functionality and their own
respective properties, methods, and events. These objects form the foundation of Active Server
Pages development and consequently have their own chapters dedicated to fully describe
them.

The Application Object

The Application object is used to manage all information in the ASP application. The informa-
tion can be accessed and passed between different users in the application. Because multiple
people can try to simultaneously change application variables, the Application object enables
locking and unlocking of its variables. When a variable is locked, other users cannot modify the
properties of an Application object. The Application object is covered in more detail in Chap-
ter 11, “Controlling the Application: The Application Object.”

The Server Object

The Server object is responsible for the administrative functionality of the server. One of the
most exciting aspects of the Server object is the capability to create an instance of a server via
its CreateObject method. This capability to tap into existing ActiveX servers transforms the
Web server from a simple file manager into a data translator. The MapPath method of the
Server object maps a physical or virtual path to a directory on the server. The ScriptTimeOut
property determines the amount of time a script can run before the script process is termi-
nated. The HTMLEncode methods enables HTML encoding of a string. A complete listing of all
the properties, methods, and events for the Server object can found in Chapter 12.

The ObjectContext Object

The ObjectContext object is used by ASP to control transaction processing using the
Microsoft Transaction Server (MTS). This direct processing within the MTS environment
enables Active Server Pages to explicitly control, commit, and rollback features of objects
managed by MTS. For more details on the ObjectContext object, please see Chapter 12.

14.1389-6 CH10 192 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Debugging Active Server Pages 193

The Session Object

The Session object is responsible for managing information for a specific user session. These
variables are not accessible by other user sessions, but they can be passed from page to page
within the ASP application. The SessionID property uniquely identifies the particular user
session. The Timeout property sets the time in minutes that the SessionID is valid. Once the
SessionID is invalid, the application destroys the user’s variables and the user must enter the
application from the beginning. If the Timeout property is not reached, executing the Abandon
method will destroy the specific user session. Details of the Session object can be found in
Chapter 13.

The Response Object

The Response object is responsible for sending output from the server to the requesting Part
client. The three most common Response object methods are the Write, Redirect, and
Response methods. The Response object uses the Write method to send HTTP output to the
III
browser. The Redirect method provides the capability to redirect the client browser to a Ch
different URL. The Response object enables a Cookie collection to set cookie values on the
client’s browser. More details on the Response object can be found in Chapter 14, “Speaking to
10
the Web Client: The Response Collection.”

The Request Object

The Request object is primarily responsible for retrieving information from the client browser,
and it uses five collections to provide for communication between the client browser and Web
server. (A collection is commonly referred to as a grouping of common objects.) The
ClientCertificate collection retrieves certification fields from Web browsers. The Cookies
collection enables the server to collect information from client-side cookies. The Form collec-
tion retrieves values posted by a client form. Finally, the QueryString collection enables values
to be parsed from the HTTP URL string. The Request object can also retrieve server environ-
mental variables. To learn more about the Request object, see Chapter 15, “Retrieving User
Information: The Request Object.”

Debugging Active Server Pages

Although Active Server Page technology is rapidly changing the way applications are being
deployed, the debugging of Active Server Pages is still an area that needs to be improved. The
role of debugging is primarily the role of the ASP scripting language itself. If the host scripting
language supports error handling, processing, and debugging, errors generally fall into two
categories. The first role is the capability of Active Server Pages to handle error trapping when
scripts are processed on the Web server. VBScript provides a limited runtime error object that
provides some runtime error processing.

14.1389-6 CH10 193 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 194 Chapter 10 The Composition of an Active Server Application

The second debugging role is providing real-time debugging capabilities that enable the Active
Server Page code to be tested at design time. This type of debugging is the responsibility of the
integrated development environment (IDE) tool used to create the application. Currently the
most common tools used to create Active Server Pages are text editors, such as Notepad, or
development environments such as Microsoft Visual InterDev. Inline debugging is obviously
not available in simple editors or even Visual InterDev. However, the Microsoft Script
Debugger provides inline debugging capabilities for VBScript, JScript, and Java applications.
For more information on using the Script Debugger, please see Chapter 9, “Using the Script
Debugger.”

Trapping Errors with the Err Object

Active Server Pages use the error handling found in VBScript to process and manage runtime
errors. VBScript handles error trapping with the Err object. The Err object has several proper-
ties and two methods. The Err object’s properties and methods can be accessed using the
following syntax:
Err[.{property | method}]

where Err is the generated error object. The Err object’s properties provide information about
the error object. These properties are listed in Table 10.1.

Table 10.1 The Err Object’s Properties

Property Description
Description Sets or returns the referenced error description.
Number Sets or returns the number value of the error.
Source Sets or returns the Application object that generated the error.
HelpFile Sets or returns the fully qualified path to the help file.
HelpContext Sets or returns the topic ID with the help file.

The Err object also provides methods that control it. The Err object’s methods, as shown in
Table 10.2, can be used to generate a runtime error or reset the Err object’s properties.

Table 10.2 The Err Object’s Methods

Method Description
Raise Generates a runtime error.
Clear Resets all object properties.

14.1389-6 CH10 194 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Debugging Active Server Pages 195

In VBScript code, the On Error statement is used to enable error handling. The Resume Next
statement is used to respond to an error condition by enabling the program to continue execu-
tion with the next statement after the one that produced the error. Therefore, to enable error
handling in VBScript, use the following line of code:
<% On Error Resume Next %>

If no error handling is activated, any runtime error that is generated at the server will result in
the immediate termination of the script. However, if error handling is activated, the application
continues by executing the next line in the code.
<%
On Error Resume Next ‘=== Turns on Error Handling
Dim myMonth
myMonth = FormatDate(Now) ‘=== Spelling Error produces error
Response.Write myMonth ‘=== Application Focus is returned here Part
%>
III
In this example, a blank string is written to the page because the VBScript does not have a
Ch
built-in function named FormatDate. When the Err object is created because of the improper
spelling of the FormatDateTime function, the script is not immediately terminated and process- 10
ing continued throughout the ASP. In this example, a blank line is written to the page. For
detailed information about error trapping in VBScript, please refer to Chapter 5, “Controlling
Program Flow with VBScript.”

Trapping Errors with the Web Server Log File

You can also utilize two Registry Settings, the ScriptErrorsSentToBrowser and the
ScriptErrorMessage, to help control and manage primitive error trapping in Active Server
Pages. The ScriptErrorsSentToBrowser Registry Setting is used to determine whether the
specific error information will be written to the Web server log file and to the requesting
browser. The ScriptErrorsSentToBrowser enables you to trap and log useful debugging, such
as the filename, error line numbers, and description messages. If ScriptErrorsSentToBrowser
is disabled, the string in the ScriptErrorMessage is displayed to the browser. The
ScriptErrorMessage contains either the default error message or it can be customized to meet
your needs. The default message is An error occurred on the server when processing the
URL. Please contact the system administrator.

Trapping Errors by Writing Text Files

Another primitive method you can use to debug your ASP applications is to create your own
log file using the VBScript FileSystem object. The FileSystem object enables you to read and
write to text files on the Web server. Using this object, you can create a log file that can be
viewed for bugging purposes, as shown in the following code:
<%
DebugFileName = Server.MapPath (“/myApp”) + “\debuglog.txt”
Set FileObject = Server.CreateObject(“Scripting.FileSystemObject”)
Set FileOut= FileObject.OpenTextFile(DebugFileName, 8, True)
%>

14.1389-6 CH10 195 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 196 Chapter 10 The Composition of an Active Server Application

This example first uses the Server.MapPath method to create the physical path setting to the
debuglog.txt file. The FileSystemObject is used to return a TextStream object to read and
write to a customized debugging text file. The FileObject.OpenTextFile method is used to
read from or append to the file. In this example, any new information will be appended to the
text file and a new text file will be created if the existing file cannot be found.

This is particularly useful in the OnStart or OnEnd events of the Application and Session
objects in the GLOBAL.ASA file, where certain objects cannot be referenced in different events.
For example, you can write variable values to a text file on the Application_OnEnd event to
check the values of the variable and to verify that the Application_OnEnd event was actually
called.

You can also use the text file error-trapping approach to determine the variables or execution
points in an ASP, which is accomplished by writing values at different points in the ASP. For
example, the following example demonstrates writing a debug file to track the application
workflow.
<%
DebugFileName = Server.MapPath (“/myApp”) + “\debuglog.txt”
Set FileObject = Server.CreateObject(“Scripting.FileSystemObject”)
Set FileOut= FileObject.OpenTextFile(DebugFileName, 8, True)

FileOut.WriteLine(“Function PadStr Initiated”) ‘=== Writes Initiated string

to LogFile
newValue = PadStr(myString, 5) ‘=== Calls Function
FileOut.WriteLine(“Function PadStr Completed”) ‘=== Writes Complete string to
➥LogFile
%>

In this situation, the string Function PadStr Initiated is written to the debugging file using
the WriteLine method. Then the custom-created PadStr function is called. If the PadStr is
successful, the TextStream object writes the string Function PadStr Completed to the debug
file.

Integrating the debugging technique throughout your ASP would lead to a debug file from
which you could follow the process that leads to the error. The following debug file illustrates
the user and ASP events:
‘=== Debug.Txt: Generate 11/23/97 10:32 AM
Welcome Page Initiated
Function LoadCookieValues Initiated
Function LoadCookieValues Completed
Welcome Page Completed
Login Page Initiated
Function dbLogin initiated

Reviewing the debug log shown above, the Welcome page was successfully loaded, but the
problem appears to be somewhere after the function dbLogin is called.

14.1389-6 CH10 196 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Controlling Access to the Site 197

Although not as robust as traditional debuggers, this textual debugging feature can be very
useful especially when dealing with debugging from an offsite or remote location. For example,
if an application encounters an error, you can have the application variables trigger a miniature
debugging system. This system can track application variables and user events to a text file for
record keeping and debugging. This process of tracking the issues that lead up to the error
will help to systematically reproduce the error and reduce debugging time.

Controlling Access to the Site

Controlling who has access to your site and application information is critical when building
Web-based applications. Specifically, for many intranet sites are adding extranet capabilities,
restricting access to your site is critical to maintain control over sensitive information. This
control can be established in two ways for your ASP Web sites and applications: at the Web Part
server level and at the file level. Regardless of the different level of security you choose to III
implement, setting up Web security requires close communication between Web site adminis-
trators and network administrators. Ch

10
Web Server Packet Filtering
The first layer of control is through TCP/IP packet filtering at the Web server level. The
Web server can accept or reject requests based on where the request is being generated. To
grant or deny access to your Web server, open the Property settings for the site. To regulate
access using the Microsoft Management Console, use the Directory Security tab, as shown in
Figure 10.5.

FIG. 10.5
The Directory Security
tab of the MMC
enables TCP/IP filtering
at the Web server level.

If you are using Internet Information Server 3.0, the TCP/IP packet filter is located in the
Internet Information Service Manager on the Advanced tab, as shown in Figure 10.6.

14.1389-6 CH10 197 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 198 Chapter 10 The Composition of an Active Server Application

FIG. 10.6
IIS 3.0 also provides
Web site filtering
through the Internet
Service Manager.

Both the MMC and the Internet Service Manager grant and restrict access based on individual
computer IP addresses, a group of computers using an IP and submask, or entire domains of
computers.

File Level Access Control

The second control layer to regulate the permission of the site is to manage the permission at
the file level. Typically, when you set up the Internet Information Server, the username
IUSR_machinename is created to represent the Internet guest account. When an anonymous
user requests pages from your public site, NT uses this security setting to access the requested
files. Therefore, to restrict access to only authorized users or groups, use the NT file permis-
sion to administer user rights. To view the security rights of files on the Web server, right-click
the file or directory and select the Property item. The Security tab is used to manage the differ-
ent levels of access on the file or directory.
Clicking the Permission button displays who has rights to the file or directory and what level of
access they have. For example, Figure 10.7 illustrates that the Internet guest account has per-
missions to this file. If you double-click the user account, the detailed permissions of the file
can be viewed.

FIG. 10.7
Access permissions can
be assigned by using
the File Permissions
dialog box.

14.1389-6 CH10 198 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Configuring Active Server Pages 199

To prevent anonymous users from accessing this file, remove the anonymous guest account
and assign the desired user or group permissions. Now that the Internet guest account has
been removed, when the Internet Information server tries to access this file, the user will be
prompted for proper NT identification. This technique enables only user with valid NT user
accounts to access this site. Now, when any page on the Web site is requested, the user is
prompted with an authorization dialog box, as shown in Figure 10.8.

FIG. 10.8
Remove anonymous
access permissions to
prompt user for valid NT
user accounts.

Part
III
Keep in mind that the username and password are being passed in non-encrypted form from Ch
the Web browser to the Web server. A potential hacker could intercept the packet and retrieve
the NT username and password. To prevent this situation from occurring, only transmit 10
usernames and passwords over a Secure Socket Layer or other encrypted channels.

Configuring Active Server Pages

Active Server Pages require configuration at two levels. For IIS 3.0, the Registry provides a
storage location needed for the Web service configurations. In IIS 4.0, the Metabase is used to
store information about the IIS Web server. The Metabase is a hierarchical database that ex-
pands the functionality of the Registry by allowing properties to be set at the computer, Web
site, virtual directory, directory, and file levels.

The second location where Active Servers Pages are configured is at the file level. Usually, a
Web server interface tool, such as the Internet Service Manager, performs the file level con-
figuration automatically for you. These custom interfaces provide a way to configure the secu-
rity permissions that are needed to view, read, and execute your Web subsystems without
using traditional administration tools.

The Metabase
The Metabase is a hierarchical database used to store configuration information about the IIS
4.0 server. The easiest way to configure the metabase is through the Internet Service Manager.
The metabase ASP settings can be found by selecting the App Options property sheet on the
Home Directory property sheet, and then clicking the Configuration button.

The Metabase is used to store all configuration issues about the IIS 4.0 Web server. Table 10.3
illustrates the related metabase tags for Active Server Pages.

14.1389-6 CH10 199 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 200 Chapter 10 The Composition of an Active Server Application

Table 10.3 The Metabase Tags

Name Description
AspBufferingOn Controls whether Active Server Pages buffers output to a
browser. If set to True (that is, set to 1), HTTP headers
can be sent to the browser from anywhere within a script.
AspLogErrorRequests Sets whether the Web server writes unsuccessful client
requests to the Windows NT event log file. The value 1
turns error logging on, and 0 turns it off.
AspScriptErrorSentToBrowser Determines whether the Web server writes the filename,
error and line numbers, and description to the client
browser and to the Windows NT event log.
AspScriptErrorMessage Sets the error message that appears in a client browser if
detailed ASP error messages are disabled.
AspScriptFileCacheSize Sets the amount of memory in bytes to allocate for the
caching of precompiled script files. A setting of 0 indicates
that no script files will be cached. A value of –1 indicates
that all script files will be cached.
AspScriptEngineCacheMax Sets the value for the maximum number of ActiveX
language engines that Active Server Pages keep cached
in memory.
AspScriptTimeout Sets the length of time Active Server Pages will allow a
script to run before terminating the script and recording
the event to the event log.
AspSessionTimeout Sets the default amount of time a Session object should
be maintained after the last request associated with the
Session object is made.

AspEnableParentPaths Sets the flag to control whether Active Server Pages can
enable paths relative to the current directory using the ..
operator.
AspAllowSessionState Sets the flag controls to determine whether Active Server
Pages can maintain session state in the ASP applications.
If disabled, the ASPSessionID cookie will not be sent to
clients and any attempts to store anything in the session
or to use Session_OnStart or Session_OnEnd will
generate an error.
AspScriptLanguage Sets the primary scripting language for all Active Server
Page applications.
AspExceptionCatchEnable Used to set controls flag to monitor exception catching
from components. If enabled, Active Server Pages will
catch exceptions that are thrown by components.

14.1389-6 CH10 200 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Configuring Active Server Pages 201

Name Description
AppAllowDebugging Sets whether or not Active Server Page debugging is
enabled.
AppAllowClientDebug Determines whether or not Active Server Pages client-
side debugging is enabled.
ScriptMaps Used to set what applications will interpret the script
filename extensions.
CGITimeout Sets the maximum time in seconds the WWW service will
wait for a response from CGI scripts.
CacheISAPI Used to indicate whether ISAPI DLLs are loaded and
cached if the DLLs are unloaded after they are run.
Part
III
The Registry Settings Ch
In Windows-based operating systems, the Registry contains important information that can
help you optimize your Active Server Pages. These Active Server settings are used to control
10
the interaction between the Internet Information Server HTTP service and the Active Server
Page interpreter engine. From an organizational aspect, the Active Server Pages process is a
sub-category of the Web server service. Naturally, the Active Server process running as a
standalone process would be worthless if it were unable to accept and deliver HTTP requests
to the client.

The Registry itself represents a hierarchical database used to store system-level information.
This system database contains critical information about configuration issues within Windows
95 and Windows NT environments. The Registry acts as the central repository for your system
settings, and manages settings from one location across different applications. The Registry
also prevents user and application settings from being disbursed in various directories in vari-
ous files on the system. Specifically, the centralized storage location prevents configuration
files from being loaded in various .INI, .BAT, .SYS, and other files.

The database is typically managed through native Windows controls, such as the components
found in the Control Panel. Because the database is central to maintaining your operating
system, improper settings made in the Registry can cause catastrophic operating system
failure.

The Registry is best viewed and configured with a graphical Registry navigation tool, such as
the Registry Editor. The Registry Editor is executed by running REGEDIT.EXE from the com-
mand line.

The Registry Editor provides a convenient hierarchical viewing mechanism. The Active Server
Pages Registry settings are found in the following file path:

14.1389-6 CH10 201 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 202 Chapter 10 The Composition of an Active Server Application

HKEY_LOCAL_MACHINE\SYSTEM
\CurrentControlSet
\Services
\W3SVC
\ASP
\Parameters

The Registry settings, as shown in Table 10.4, contain many settings that are central to admin-
istrating the Active Server Pages. For example, you can use the Registry settings to change the
default scripting language of the Active Server, which is VBScript. As mentioned earlier in the
chapter, unless otherwise specified by the <% LANGUAGE %> on a page, the default scripting
language is responsible for processing the script contained within the <% %> tags. To change
the default scripting language of the machine, change the date value of the
DefaultScriptLanguage Registry setting. For example, if the default scripting is JScript, you
can convert the default scripting language to VBScript by changing the
DefaultScriptLanguage Registry name to VBScript.

Unless otherwise stated, most changes require the Web server to be restarted for changes to
take effect.

Table 10.4 Active Server Pages Registry Settings

Name Description
AllowSessionState Sets whether Active Server Pages can make session state
in applications. If the session state is not allowed, the
ASPSessionID cookie will not be sent to clients that access
the Web site. Consequently, if code existed in the
GLOBAL.ASA file, within the Session_OnStart or
Session_OnEnd events, an error will be generated.

BufferingOn Sets whether Active Server Pages buffers output to a

browser. If BufferingOn is activated, all output is com-
pletely collected in memory before being sent to the
browser.
DefaultScriptLanguage Sets the primary scripting language for all Active Server
Pages applications on the server. To override this setting
on a page basis, use the <%@ LANGUAGE =
ScriptingLanguage %> on the first line of an Active Server
Page.
EnableParentPaths Sets whether Active Server Pages can enable browser
navigation from the current directory. If so, the browser is
able to navigate up the directory tree using the parent
directory operator (..).
LogErrorRequests Sets whether the Web server can write unsuccessful client
requests to the Windows NT event log file.
MemFreeFactor Sets the maximum length of the free memory as a percent-
age of the used memory list.

14.1389-6 CH10 202 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Configuring Active Server Pages 203

Name Description
MinUsedBlocks Sets the minimum length of the used memory list before
elements can be freed.
NumInitialThreads Sets the number of worker threads that Active Server
Pages creates when it is started.
ProcessorThreadMax Sets the maximum number of worker threads to create per
processor. According to documentation, it is not recom-
mended to create more than 20 threads per processor.
RequestQueueMax Sets the maximum number of ASP requests to keep in the
queue for each thread. Once the maximum requests have
been made, the server returns a ServerTooBusy message
to the browser. Part
ScriptEngineCacheMax Sets maximum number of ActiveX language engines that
Active Server Pages will keep cached in memory.
III
Ch
ScriptErrorMessage Sets the error message that appears on a browser if
ScriptErrorsSentToBrowser. 10
ScriptErrorsSentToBrowser Sets whether the Web server writes debugging specifics,
such as the filename, error and line number, and error
descriptions to the browser in addition to the log.
ScriptFileCacheSize Sets the amount of memory in bytes to allocate for the
caching of precompiled script files, which improves the
performance of Active Server Pages.
ScriptFileCacheTTL Sets the amount of time that script files will remain in the
memory cache before being phased out if there have been
no references to those scripts.
ScriptTimeout Sets the length of time Active Server Pages will allow a
script to run to completion. This can be overridden in a
script by calling the Server.ScriptTimeout method.
SessionTimeout Sets the default amount of time a Session object should be
maintained after the last request associated with the object
is made. This can be overridden in a script by using the
Session.Timeout method call.

StartConnectionPool Sets whether ODBC connection pooling is turned on

(default) or off.
ThreadCreationThreshold Sets the number of requests that can be maintained in the
common queue to process Active Server Pages. If requests
exceed this number, a new thread is created in the thread
pool but is always less than the amount specified in the
ProcessorThreadMax amount field.

14.1389-6 CH10 203 10/26/98, 9:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 204 Chapter 10 The Composition of an Active Server Application

Web Server Interface Utilities

An ASP application consists of a group of ASP files collected in a virtual directory. The permis-
sions on this virtual or application directory regulate the rights and permissions of the ASP
application. Most Web servers provide an interface that enables you to configure and adminis-
ter the Web server. Usually this interface is provided in two forms, through a native client
utility and through an HTML interface. Regardless of the method used to administer the Web
service, the settings needed for ASP applications are the same. In order for the Web server to
process the ASP files, the permission on the files must have read and execute permissions set.
Assuming that IIS and Active Server Pages have been properly installed, setting these permis-
sions is most easily done via the Internet Service Manager. To configure the permissions on a
directory via the Internet Service Manager, select a directory from the Virtual Directory tab,
as shown in Figure 10.9.

FIG. 10.9
Configuring directory
permissions to control
the ASP execution
permissions.

After selecting the appropriate directory, set the read and execution check boxes to enable all
files in that directory to be read and executed by the Web server. If you were using the HTTP
service to simply deliver ordinary HTML files with no embedded scripting, the files in that
directory would only need to have read access. If just read permission is set when using ASP
files, the Active Server scripts would be displayed on the requesting client’s browser, as shown
in Figure 10.10. The script is shown because the Web server is simply transferring all the text
in the page to the browser without executing it.

However, if you have a directory that contains all your Perl scripts for a Web application, the
directory permissions would have to be set to execute in order for the files to process. If just
execute permission is set for the directory that contains static content, the Web server will only
display the text on the requesting page. All links that require the server to read any other type
of information will display as broken links.

Typically, if you use graphical user interface (GUI) tools such as Visual InterDev or FrontPage
to create ASP applications, the read and execute permissions for the virtual directory of the
Web would be set.

14.1389-6 CH10 204 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Configuring Active Server Pages 205

FIG. 10.10
Active Server script is
displayed as text if the
execute permission is
not set for the directory.

Part
III
Ch

10
For a more detailed discussion on administering your site, Chapter 31 provides an overview of
how to manage your sites using the Internet Services Manager.

Using ASP Directives

Active Server Pages have the capability to override page-level settings that have been stored in
the Metabase or Registry settings. Directives are used by ASPs to let the IIS know how to
process the individual ASP files. IIS 4.0 uses five directives to control page-level processing, as
shown in Table 10.5.

Table 10.5 IIS 4.0 ASP Directives

Directive Description
@CodePage Sets the code page of an ASP script.
@EnableSessionState Sets whether or not session information is stored.
@Language Sets the primary scripting language of the ASP file.
@LCID Sets the local identifier for the ASP file.
@Transaction Indicates whether ASP scripts should be processed as a
transaction.

14.1389-6 CH10 205 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 206 Chapter 10 The Composition of an Active Server Application

The ASP directives must be located on the first line of the ASP file. The syntax for the process-
ing directives is
<%@DirectiveName=Value%>

where DirectiveName is a valid processing directive and Value is a value assigned to the direc-
tive.

The @CodePage Directive The @CODEPAGE directive is used to set the code page for an ASP
file. The code page represents a unique character set used to define a specific language and
localized region. To set the code page for an ASP file, use the following syntax on the first line
of ASP script:
<%@CODEPAGE=codepage%>

where codepage is a valid code page. The @CodePage directive can be superceded by the
Session.CodePage property. For example, the following code is used to change the code page
used to display Japanese Kanji:
<%@CODEPAGE = 932%>

For more information on the code page, code page values, and the Session object, please see
Chapter 13.

The @EnableSessionState Directive The @EnableSessionState processing directive is used

to control whether session tracking is applied to the ASP file. The @EnableSessionState direc-
tive enables session tracking to be turned on or off. To use this page-level feature, use the
following syntax:
<%@EnableSessionSstate=True|False %>

For example, to disable session tracking information on a specific ASP file and consequently
decrease the processing time, the following code would be used:
<%@EnableSessionsState = False %>

This feature is beneficial when processing functionality is required in ASP scripts but tracking
the user information is not necessary. If monitoring user information via the Session object is
not necessary, the scripting engine processing load can be reduced.

The @Language Directive The @Language directive is used to set the primary scripting lan-
guage for the ASP file. The primary scripting language is responsible for processing all script
between the <% %> delimiters. To set the primary scripting language for a page, use the follow-
ing syntax on the first line of the ASP file:
<%@Language=scriptengine %>

where scriptengine is a valid script engine on the server. IIS 4.0 ships with the VBScript and
JScript scripting engines. Therefore, the following code example sets the primary scripting
engine to VBScript:
<%@Language = VBScript %>

14.1389-6 CH10 206 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 Configuring Active Server Pages 207

The @LCID Directive The @LCID processing directive is responsible for setting the local iden-
tifier for an ASP file. The local identifier uses standard international numeric abbreviations to
specify local date, currency, and time formats. To use the LCID directive, use the following
syntax on the first line of the ASP file:
<%@LCID=localeidentifier %>

where LCID is the valid local identifier. This directive is particularly useful for sites that must
support multiple languages. For example, to set the local identifier to use the French local
identifier, use the following code:
<%@LCID= 1036 %>

As with the @CodePage directive, the @LCID directive can be overwritten by the Session.LCID
property.
Part
The @Transaction Directive The transaction-processing directive is used to enable the IIS
Web server to treat the ASP script as a transaction monitored by the Microsoft Transaction
III
Server (MTS). To use the @Transaction directive on an ASP file, use the following syntax: Ch

<%@TRANSACTION=value %> 10
where value is the level of transaction support. The MTS transactions can have four levels of
support, as shown in Table 10.6.

Table 10.6 ASP Transactional Support in the Microsoft Transaction Server

Value Description
Required Script initiates a transaction.
Required_New Script initiates a transaction.
Supported Script will not initiate a transaction.
Not_Supported Script will not initiate a transaction.

Notice in Table 10.6 that although four values of transactional support are given, the transac-
tional ASP script processing is either initiated or not initiated. However, you can use these
transactional support values to add functionality to components managed by the MTS. For
more information on managing components from ASP scripts, please see Chapter 12.

To enable transactional ASP processing, use the following syntax on the first line of the ASP
file:
<%@Transaction = Required %>

14.1389-6 CH10 207 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 208 Chapter 10 The Composition of an Active Server Application

As with other directive, the @Transaction directive is only supported while the current page is
processed and is terminated after the page is finished processing. For more information on
using the Microsoft Transaction Server, please see Chapter 28, “Working with the Microsoft
Transaction Server.”

Handling Known Active Server Page Issues

As Active Server technology continues to grow, different versions of the Active Server Pages
exist. As in any emerging technology, known development issues arise that can cause potential
damage to your Web applications. Table 10.7 categorizes the different ASP versions and builds
to help identify what version of Active Server Pages you are running.

Table 10.7 ASP Version Information

ASP Version ASP Build ASP DLL Version ASP DLL Date
1.0 1.12.09 1.12.06 12/6/96
1.0A 1.13.31 1.13.09 1/31/97
1.0B 1.15.14 1.15.14 3/14/97

There have been two significant issues discovered with Active Server Pages. The first tempo-
rary setback was that the script code of the ASP could be viewed in the client browser by plac-
ing a period (.) at the end of a URL. This viewing privilege opened huge security concerns by
exposing the source code driving ASP applications, Internet Database Connector (IDC) appli-
cations, and Common Gateway Interface (CGI) applications. A hotfix to NT 4.0 SP2 has been
posted at the Microsoft FTP site at:
FTP://ftp.microsoft.com/bussys/winnt/winnt-public/fixes/usa/nt40/hotfixes-
postsp2/iis-fix/readme.1st

To obtain additional information about this problem, see article Q163485 of the Microsoft
Knowledge Base on the Microsoft support Web site.

The second issue regarding Active Server Pages is the existence of a memory leak for systems
that have upgraded to ASP version 1.0B. The rate of the memory leak is proportional to how
heavy the site is accessed. The memory leak can be demonstrated by using the NT Perfor-
mance Monitor to track the Active Server Page object. The NT Performance Monitor provides
a data collection mechanism to generate data for graph and charts for NT server objects. A
hotfix to NT 4.0 SP3 is also available at:
FTP://ftp.microsoft.com/bussys/winnt/winnt-public/fixes/usa/nt40/hotfixes-
postsp3/asp-fix/readme.1st

Note that this hotfix should only be applied to systems that have upgraded to 1.0B. For systems
that have not upgraded, the 1.0 and 1.0A version will not apply this hotfix. For more informa-
tion on this problem, see article Q150934 in the Microsoft Knowledge Base.

14.1389-6 CH10 208 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 From Here… 209

Most of the ASP updates are included in the NT service packs, but are not necessarily auto-
matically installed to update the Active Server Pages when the service packs are installed. For
example, NT 4.0 Service Pack 3 includes a separate standalone setup executable for ASP 1.0B
that is not installed automatically when you apply the NT Service Pack.

From Here…
As you can see, an Active Server Page application consists of many files, directories and
virtual directories, and six built-in ASP objects. The Metabase and Registry settings are used
to store the IIS Web ser ver configuration for your static and dynamic HTML needs. At this
point, with the lofty application architecture in place, the specifics of Active Server Page
objects can be discovered. The following chapters will provide a detailed approach on how to
use the specific ASP objects and how they all tie in together to deliver Web applications to Part
any HTML-compliant browser. III
■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of Ch
tracking a user through an ASP application.
■ Chapter 15, “Retrieving User Information: The Request Object,” demonstrates how to
10
collect information from the user using forms, hidden fields, and cookies.
■ Chapter 16, “Putting It All Together: Building an Online Catalog,” demonstrates how to
use the ActiveX components that ship with IIS 4.0.
■ Chapter 17, “Implementing Existing Server-Side Components,” illustrates how to use the
various components that ship with IIS and Active Server Pages in your applications.
■ Chapter 18, “Creating Your Own Active Server Components,” provides a walk-through to
the tools and examples on how to create tailored functionality via custom components.

14.1389-6 CH10 209 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 210 Chapter 10 The Composition of an Active Server Application

14.1389-6 CH10 210 10/26/98, 9:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH10 lp#3
 11
211

C H A P T E R

Trapping Application Events

Controlling the Discover how the Application object
is used to manage Active Server

Application: The Page applications and how to use

the beginning and ending events
to initiate and close your ASP

Application Object applications.

Controlling Application
Variables
Examine how to use the Lock and
UnLock application methods to Part
prevent multiple users from simulta-
neously changing application vari-
III
Active Server Pages has six built-in objects for program- ables. Ch
mers to use. The Application object is used to control
Using Application-Level 11
and manage all items that are available to all users of an
Components
Active Server application. The Application items can be
variables needed for all users in the application, or they Understand how to create compo-
can be instantiated objects that provide special server-side nents at the application level that are
accessible to all users and learn how
functionality. This chapter will demonstrate what the
to use application-level static and
Application object represents, how to trap its events, and
dynamic arrays.
how to use its properties and methods to manage an ASP
application. ■ Implementing Application
Collections
Learn how to use the Application
object’s Contents and Static collec-
tions to help manage
application-level variables and
objects.

15.1389-6 CH11 211 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 212 Chapter 11 Controlling the Application: The Application Object

Trapping Application Events

The Application object is a single object that is created to represent an instance of an ASP
application. From the previous chapter, you learned that an Active Server application is actually
a collection of ASP files. These files are housed in directories and subdirectories that comprise
virtual directories on the Web server. When any ASP file is requested from a virtual Active
Server directory, the Application object is instantiated to start, manage, and share informa-
tion throughout the entire ASP application. Information can then be shared to all users in the
application. As you can see, this process of starting an application is different from the tradi-
tional application startup mechanisms. Most applications are started by calling a single execut-
able file. However, with Active Server Pages, the application is started by calling any .ASP file
in the virtual directory. The Application object is the fundamental object that is used to start
and manage ASP applications. The Application object is a single object that can be accessed
by any user of that application. The individual users, represented by Session objects, can read
and update the Application object as needed. The Session objects can use the Request and
Response objects to communicate information to the browser. Figure 11.1 represents the rela-
tionship of the Application object and the Session, Request, and Response objects.

FIG. 11.1 The ASP Application

The Application The Application object
object represents an Application ("myAppVar") = "Foo"
instance of an ASP
application that can be read from
update Application object
accessed by any user. Application object

Session Object Session Object

Application.Lock
my SessionVar = Application ("myAppVar") Application("myAppVar") = "Foobar"
Application.unlock

IIS I/O: Request and Response Objects

Internet

Browser

IIS uses the Application object not only to share information among all users in the applica-
tion, but also to manage application processes and resources. IIS 4.0 enables you to run entire
ASP applications within their own process space to protect against a catastrophic failure of an

15.1389-6 CH11 212 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Trapping Application Events 213

application from bringing down the entire the entire HTTP server. Furthermore, the process
isolation enables you to wrap transaction processing directly in Active Server Pages. For more
information on configuring the ASP application, refer to Chapter 10, “The Composition of an
Active Server Application.”

The Application object, as well as the other Active Server objects, follows the familiar
object.method syntax:

Application.method

where Application is the Active Server Application object.

The Application object has two events: Application_OnStart and Application_OnEnd. The
Application object is created by the Active Server to represent a physical start up of an Active
Server application. Once the Application object is started, all subsequent users can reference
its objects and variables.

The Application_OnStart and Application_OnEnd events are good examples of event-driven

programming. Most of the examples discussed in Chapter 6, “Working with Functions, Subrou-
tines, and Dialog Boxes,” demonstrate client-side, event-driven examples, such as trapping the
Click event of an HTML button. The Application events are good examples of server-side,
event-driven programming. The server-side scripting of the Application object is controlled in Part
the GLOBAL.ASA file. The GLOBAL.ASA file is responsible for trapping the start and end events of
the Application and Session objects. These events use the similar syntax of object.event
III
that is found in the client-side examples but are not necessarily required to use the Applica- Ch
tion and Session objects. 11
The Application_OnStart and Application_OnEnd events, shown in Table 11.1, are used in
the GLOBAL.ASA file to initiate code when either of these events is triggered.

Table 11.1 The Application Object’s Events

Event Description
Application_OnStart Initiated when the ASP application is first started.
Application_OnEnd Initiated when the ASP application is terminated.

The server-side scripting of the Application object is controlled in the GLOBAL.ASA file. The
GLOBAL.ASA file is responsible for trapping the start and end events of the Application and
Session objects. The individual application events are found in the GLOBAL.ASA file and are
executed in a similar manner to any subroutine or function. As seen in Chapter 6, subroutines
and functions are executed by calling the procedure name. For example, the following
GLOBAL.ASA file only initiates the Application_OnStart event. The Application OnEnd event
will not be executed because it is commented out. For more information on the GLOBAL.ASA
file, please see Chapter 10.
<SCRIPT LANGUAGE=”VBScript” RUNAT = “Server”>

15.1389-6 CH11 213 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 214 Chapter 11 Controlling the Application: The Application Object

Sub Application_OnStart ‘=== Defines Application Object’s OnStart Event

Application(“strMessage”) = “Welcome to Active Server Pages”
End Sub

Application _OnStart ‘=== Call this procedure

‘ Application_OnEnd ‘=== Ignored, commented out

</SCRIPT>

TIP The Application_OnStart and Application_OnEnd events are not necessarily required to use
the Application object.

Declaring Application-Level Variables

As seen in Part II, “VBScript: The Foundation of Active Server Pages,” the proper use of vari-
ables forms the foundation of any programming model. The same is true when dealing with
Active Server objects. Variable declaration in the Application object is a topic that has some
interesting characteristics that need to be properly understood when building Active Server
Page applications.

Application object-level variables are accessible by all users of an application and share infor-
mation common to all users. Application-level variables can be defined as needed in any ASP
file, not just in the Application_OnStart and Application_OnEnd events of the GLOBAL.ASA file.
This means application variables can be also be created in the Session events and during page-
level script processing. Keep in mind, from the previous chapter that an ASP application is the
collection of Active Server Pages contained in a virtual directory on the Web server. Multiple
users can be navigating through a variety of different pages but all have access to a common
set of data. The Application object acts as the manager for this library of shared information.

Variables have two levels of scope in Active Server Pages: page-level variables and object-level
variables. The page-level variables are defined by a scripting language within the ASP scripts.
The page-level scripting only has scope as that page is being processed. Once the page is fin-
ished processing, the page-level variables are destroyed and the resources released. Part II of
this book focused on using VBScript to create, process, and manage page-level variables.

The second type of variable is the object variable. Object-level variables transcend page-level
variables because the scope of object variables extends beyond page-level variable processing.
Object-level variables exist in two forms: at the session level and at the application level.
Session-level variables are used to pass variables from page to page for a specific user and
application-level variables are used to share common information between all users. These
application and session variables are global to all page-level variables on Active Server Pages.

Application- and session-level variables are declared by simply using the desired variable name
with the appropriate object-level reference. To declare a variable, use the following syntax:
ObjectLevel(varName)

where ObjectLevel is the Application or Session object and varName is the name of the
variable.

15.1389-6 CH11 214 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Trapping Application Events 215

For example, to declare an application-level variable named myAppWelcomeStr that can be ac-
cessible to all users, use the following line of code in either the GLOBAL.ASA file or in any ASP
file:
Application(“myAppWelcomeStr”) = “Welcome To ASPTraining Center Online”

Notice the difference between this type of implicit declaration at the object level and how vari-
ables are declared at the script level. The script-level variables in VBScript can be declared
explicitly using the Dim keyword or can be implicitly declared by assigning a variable to a vari-
able name. Listing 11.1 uses the combination of declaring an application-level object in the
GLOBAL.ASA file to set the welcome message for the application and then uses page-level script-
ing to write the message to the page.

Listing 11.1 Declaring an application-level object to set a welcome message

for all users in the application.
‘============= The Global.ASA File Begin ===========
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>
Sub Application_OnStart
Application(“myWelcomeStr”) = “Welcome to ASP International”
Application(“myTitle”)= “ ASP International Corporate Home “End Sub Part

</SCRIPT> III
‘============= The Global.ASA File End ================
Ch

11
‘============= A Requesting ASP Page =================
<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE><% =Application(“myTitle”)%></TITLE>
</HEAD>
<BODY>

<Center><B>
<% = Application(“myWelcomeStr”) ‘=== Writes variable on page
Â%>
<% Application(“myAppConnectStr”) = “DBconnect” ’=== Creates an applica-
tion Âlevel variable
</B></Center>

</BODY>
</HTML>
‘===================== ASP End =============================

Notice how this code also uses the application variable myTitle to set the values used in the
HTML title bar. In addition, this example also writes the string ‘DBconnect’ to the application
variable named myAppConnectStr.

15.1389-6 CH11 215 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 216 Chapter 11 Controlling the Application: The Application Object

CAUTION
Take care to ensure proper spelling with application- and session-level variables. The lack of the Option
Explicit functionality found at the script level may lead to incorrect variable assignments or zero length
string retrievals.

Although you can use page-level and object-level variables in Active Server Pages, they are not
exclusive to themselves. You can utilize both types of variables at different scoping levels as
needed within any ASP. In addition, you can utilize VBScript variables and control logic in the
GLOBAL.ASA events to process your application logic. For example, the following script
builds a counter that tracks the amount of users entering a site using the application- and
session-level objects.
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>
Sub Application_OnStart ‘=== Sets an Application variable
Application(“counter”)=”1"
End Sub
</SCRIPT>

<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Session_OnStart ‘=== Using the Session Level
➥event
dim userCounter
userCounter =Application(“counter”)
userCounter = Counter + 13
Application(“Counter”)= userCounter ‘=== Update the Application
➥variable

End Sub
</SCRIPT>

In this example the Application_OnStart event is used to initiate the counter variable to 1
when the ASP application is created. As new users enter the site, the Session_OnStart event
uses VBScript to increment the counter. For more information on the Session object and its
events, please see Chapter 13, “Managing the User Session: The Session Object.”

Notice that the scripting variables declared in the Application_OnStart event are not them-
selves perceived as application-level variables by other ASP files in the application. For ex-
ample, if you try to reference the userCounter script variables declared in the Session_OnStart
event, the return of the variable would be a zero length string.

TIP Script-level variables used in the GLOBAL.ASA application and session events are not accessible from
other ASPs.

The same programming functionality can be created in the Application_OnEnd event. The
Application_OnEnd event will not usually contain information critical to a specific Web user.
Most of the processing contained in the Application_OnEnd event will pertain more to adminis-
tration functions, such as writing to a log file when the Web server is being shut down.

15.1389-6 CH11 216 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Trapping Application Events 217

In the previous sections, I have used the Application_OnStart and Application_OnEnd events
to illustrate how application-level variables are used. Now let’s take a closer look at details of
each Application event.

Application_OnStart Event
To start using the OnStart event of the Application object, the appropriate script tags must be
defined in the GLOBAL.ASA file. The Application_OnStart event requires the following syntax:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>
Sub Application_OnStart
. . . ‘=== Your application initialization code here
End Sub

</SCRIPT>

The Application_OnStart event is executed before any user or session event is triggered. Any
attempts to call a Session, Response, or Request object will generate a fatal runtime error
because these objects have not been created at this point. However, you can reference the
Server object. The Server object is used to help provide HTTP administration features. The
Server object is looked at more closely in Chapter 12, “Controlling the Server with the Server
Object and Using Transaction Processing with the ObjectContext Object.” Part
III
CAUTION Ch
Referencing the Session, Request, and Response objects in the Application_OnStart event will
create a runtime error. These objects cannot be called because they have not been instantiated at this point. 11

Application_OnEnd Event
The Application_OnEnd event requires the following syntax within the GLOBAL.ASA file similar
to the Application_OnStart event:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>
Sub Application_OnEnd
‘=== Your Script Here
End Sub

</SCRIPT>

The Application_OnEnd event occurs when the Web server terminates the Application
object. Shutting down the Web server can occur in two situations. The first situation is the
obvious closing down of the HTTP service via termination of the process or via the Microsoft
Management Console or the Internet Service Manager. The second situation automatically
occurs when the GLOBAL.ASA file needs to be recompiled and reloaded into memory. This situa-
tion occurs when changes have to be made and saved to the GLOBAL.ASA file. When the Appli-
cation object detects that the GLOBAL.ASA file has changed, the Application object starts to
shut down the HTTP service. When the Application object initiates the shutdown process, the
HTTP service refuses any additional browser requests while the new GLOBAL.ASA file is being
loaded into memory.

15.1389-6 CH11 217 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 218 Chapter 11 Controlling the Application: The Application Object

As in the Application_OnStart event, any references to the Session, Request, or Response

objects will generate an error. The Session, Request, and Response objects are closed before
the Application_OnEnd event, so trying to reference a closed object will generate an error.

Controlling Application-Level Variables with Methods

The capability to create or change application-level variables on any page of the ASP application
by any or all users creates the need to control and regulate variable access. The Application
object has two methods designed to control variable access, the Lock and UnLock methods (see
Table 11.2). These methods help to prevent simultaneous multiple users from changing
application-level variables.

Table 11.2 The Application Object’s Methods

Name Description
Lock Prevents other users from changing Application object properties.
UnLock Enables other users to modify the Application object’s properties.

The application methods require the following syntax and can be referenced from any point
within the application:
Application.Method

where Application is the Application object and Method is the Lock or UnLock method.

Lock Method
The goal of the Lock method is to tightly control Application object’s properties by preventing
concurrent users from modifying the same application-level variables at the same time. The
Lock method gives a single user session the sole administration rights to an application variable
or variables. The individual user retains control of the Application object until the
Application.UnLock method is declared. The Lock method is implemented using the following
syntax:
Application.Lock

To prevent users from simultaneously changing the same Application object properties, use
the Lock and UnLock methods in conjunction. For example, suppose that you build an intranet
application to help disperse information about the status of current project. To prevent two
managers from simultaneously trying to change the due date of a project, use the Lock method
to freeze only the Application object properties to one user session.
<%
Application.Lock
Application(“vDueDate”) = Request.Form(“txtDueDate”)

15.1389-6 CH11 218 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Controlling Application-Level Variables with Methods 219

Application.UnLock
%>
This application was processed on
<%= FormatDateTime(Application(“vDueDate”), vbShortDate) %>.

Keep in mind that the Lock and UnLock methods do not keep track of the permissions of users
to change these variables. The specific user permissions are the responsibility of the Active
Server Page code to monitor using the Session object.

UnLock Method
To release control of the locked application variables, the Application.UnLock method is used.
The Application.UnLock method does not have to be explicitly called to release control of the
application variables. The Application.UnLock method is implicitly called in two situations.
The first situation is when the processing of a page is complete and the UnLock method has not
been called. Although the Application object is available to all users and all page-level scripts,
the Application OnStart and OnEnd events must occur on the same script page. If not, the
OnEnd event is automatically triggered to release the Application object’s control. The follow-
ing code illustrates using the Lock and UnLock methods to set a database connection string and
UnLock method.

<% Part
Application.Lock
➥from simultaneously modifying values
‘=== Prevent other users
III
Application (“DataWidget_ConnectionString”) = “DSN=datawidgets;DriverId=25;FIL=MS Ch
Access;MaxBufferSize=512;PageTimeout=5;”
Application (“DataWidget_ConnectionTimeout”) = 15 11
Application (“DataWidget_CommandTimeout”) = 30
Application (“DataWidget_RuntimeUserName”) = “”
Application (“DataWidget_RuntimePassword”) = “”
Application.UnLock ‘=== Explicitly declared
%>

This application uses the

<%= Application(“ DataWidget_ConnectionString “) %> Database Connection.

If the UnLock method was not explicitly declared, the Web server would automatically release
control of the Application object from the specific user session when the page was finished
processing.

The second situation where UnLock is called is when the script processing for a page times out.
The Web server monitors the amount of time allocated to resource processing. The process of
terminating a request due to exceeding the amount of time allotted for processing is referred
to as timing out. Specifically, the server monitors the amount of time the specific user session is
inactive and the amount of time spent processing the script embedded on the current page.
When a user becomes inactive by not making any requests in a specified amount of time, the
resources assigned to the individual Session object are freed, as demonstrated in Chapter
13. Likewise, if the processing of an ASP file exceeds a given amount of time to finish process-
ing, the page processing is aborted. A Timeout error is usually an indication that it was a large

15.1389-6 CH11 219 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 220 Chapter 11 Controlling the Application: The Application Object

processing request, a connection could not be established, or something is wrong with the
current process. The server uses the Registry settings to store the timeout values and various
other values critical for the Active Server Pages to function properly. The page-level timeout
value is stored in the ScriptTimeOut of the ASP Registry settings or in the IIS MetaDatabase.
The default page-level script timeout value is 90 seconds.

Using Components at the Application Level

Variable manipulation at the application level is only half of the benefit of using the Applica-
tion object. The second benefit of using the Application object is the capability to connect to
ActiveX objects that have application-level scope. ActiveX objects enable you to connect into
registered server-side automation object, such as e-mail, workflow, and database systems. For
example, you can use a single database connection for an entire application by creating the
database connection as an application-level object.

Instantiating Application-Level Objects

Application-level objects have many of the same characteristics that are common to application-
level variables. However, you are not limited to only using variables at the application level, you
can also create automation objects that have application-level scope. This enables you to create
a global instance of the object and enable multiple user sessions to have access to the Applica-
tion object.

Object declarations can have three different scoping levels. The first two scoping levels, the
application- and session-level objects, can only be created in the GLOBAL.ASA file. You can create
a reference to an automation object by embedding the unique class identifier either by the
ProgID or the ClassID within the object tag. The ProgID represents the registered name of the
object. The ClassID refers to the registered class number of the object. To create a server-side
object, use the following syntax:
<OBJECT RUNAT = Server SCOPE = Scope ID = Identifier PROGID = “progID”|CLASSID
=”ClassID”>
</OBJECT>

where Scope identifies the object’s lifetime and ID identifies the object’s instantiated name. The
scope of the object can either be “Application” to specify an application-level object scope,
“Session” to specify a session-level object, or “Page” or left blank to specify a page-level object.
Using the existing objects that ship with the Active Server will be covered in greater detail in
Chapter 17, “Implementing Existing Server-Side Components.”
<OBJECT RUNAT = Server SCOPE = Scope ID = Identifier PROGID = “progID”|CLASSID
=”ClassID”>

To create an application-level object, two steps are necessary. The first is to specify the scope of
the object to “Application”. The second step is to embed the <OBJECT> tag in the GLOBAL.ASA
file. The <OBJECT> tag must be placed outside the <SCRIPT> and </SCRIPT> tags used to define
the GLOBAL.ASA file. If an <OBJECT> tag or any other tag is found between the <SCRIPT> tags, the

15.1389-6 CH11 220 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Using Components at the Application Level 221

script interpreter will try to process the <OBJECT> tags and consequently generate an error.
This <SCRIPT> and <OBJECT> tag layout is illustrated in the following code:
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>
Sub Application_OnStart
‘=== Application Level events and object placed here
End Sub

Application_OnStart
</SCRIPT>

<OBJECT RUNAT = Server SCOPE = Application

ID = MyAdrot PROGID = “MSWC.AdRotator”>
</OBJECT>

<OBJECT RUNAT=Server SCOPE=Application ID=MyDataConn

PROGID=”RemoteConn.DataConn”>
</OBJECT>

In the preceding example, two components are declared to have application scope in the
GLOBAL.ASA file. The first object, the AdRotator Component, is a component that ships with the
Internet Information Server that rotates advertising banners on Web pages. This component is
covered in more detail in Chapter 17. The second component is a custom created component
Part
that is used to ensure that a data connection exists between two remote offices. The ID descrip-
tor of the Object tag specifies the name of the Application object that will be referenced by III
calling ASP code. This example uses the registered name to specify the type of identifier Ch
needed to instantiate this object on the Web server. For more information on the specifics of
using objects in Active Server Pages, please see Chapter 10. 11
The application-level object can now be accessed from any ASP in the application by using the
ID name of the object. The following example displays a new banner on the page by calling the
GetAdvertisement method of the AdRotator Component and checking the existence of a live
connection between remote offices by using the CheckDataConnection method of the
MyDataConn component. This capability to call the Application object from any page prevents
the server from having to create a new instance of the component on every page where the
component’s functionality is needed.
<%
Dim ScheduleFilePath, adrot
ScheduleFilePath=”datawidgetswarehouse.txt”
Response.Write adrot.GetAdvertisement(ScheduleFilePath)

bConnectionResult = MyDataConn.CheckDataConnection(“Dallas”)
If bConnectionResult then
ret = MyDataConn.TransferData(‘Inventory’)
Else
ret = MyDataConn.CheckDataConnection(“Reston”)
End If
%>

15.1389-6 CH11 221 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 222 Chapter 11 Controlling the Application: The Application Object

Then, using VBScript’s If...Then logic, if the connection is valid, the TransferData method of
the MyDataConn method is executed to transfer inventory information to the Dallas office. To
learn more about creating your own Active components, please see Chapter 18, “Creating Your
Own Active Server Components.”

Using Arrays at the Application Level

Another useful application-level variable is the array. Arrays provide a centralized location to
help organize and contain variables that are logically grouped. Arrays can be created at either
the application, session, or page levels. Arrays are useful because they can be passed between
pages as variables provided that they are declared as an application- or session-level variable.

Using application-level arrays is a more efficient way to provide information to an entire appli-
cation. Application arrays are best used to contain commonly accessed groups of information
that would usually require constant interaction or external calls. For example, if you are design-
ing a training registration scheduling service and need to display the list of training classes
offered, an application-level array would be a good way of storing this information and making
it available to all pages in your ASP application. Because this is static information, populating an
array once from a database conserves resources and processing power and saves you from
having to make a database call for every request. The following code populates the array in the
Application_OnStart event of the GLOBAL.ASA file.

Listing 11.2 Populating an application-level array.

<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Application_OnStart
Dim aMyAppTraining(4) ‘=== Creates a
➥fixed size array
aMyAppTraining (1)=”Using Active Server Pages” ‘=== Starts
➥assigning array
aMyAppTraining (2)=”Internet Architecture: The Basics”
aMyAppTraining (3)=”Content Management”
aMyAppTraining (4)=”Internet Application WorkFlow”
Application(“aTraining”)= aMyAppTraining ‘=== Assigns
➥Application variable
End Sub

Sub Session_OnStart
End Sub

Sub Session_OnEnd
End Sub

Sub Application_OnEnd
End Sub

</SCRIPT>

15.1389-6 CH11 222 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Using Components at the Application Level 223

The Dim statement in this example declares a static length array, in this situation of size 4.
(Remember that arrays are now one-based instead of the traditional 0-based arrays.) The
aMyAppTraining array is then populated with the available training classes. After the array is
loaded with information, an application-level variable named aTraining is created and loaded
with the new populated training array.

After this array is loaded as an application variable, the array can be referenced and manipu-
lated from any ASP page. To retrieve a local reference to the training class array, simply create
a reference to the array by referring to the application variable where it was stored.
<%
Dim aLocalTrainingArray
aLocalTrainingArray = Application(“aTraining”)
%>

At this point, the variable aLocalTrainingArray is only a local array. Therefore, any changes to
this array will exist only within this page. For example, you can overwrite the local array values
by reassigning a local array element. However, that change, as well as the local reference to the
array, will be destroyed after the page is finished processing.
aLocalTrainingArray (4)=”My New Class”

Once this is accomplished, the array can be processed as any other local array. For example, to Part
determine how many elements exist in the array, use the VBScript Ubound function. The follow- III
ing code populates the variable iClass with the total amount of classes available.
Ch
<%
iClass= Ubound(aLocalTrainingArray) 11
%>

The number of classes in the array will then be used to cycle through the array and write to the
page. To cycle through the array, a For...Next loop is used to iterate through each array item.
Once the new item is reached, the array value is written to the page using the Response object
and the array index.
<%
For i = 1 to iClass
Response.Write(aLocalTrainingArray (i) & “<BR>”)
Next
%>

Notice that the For...Next loop could have used the following syntax to produce the same
results.
<%
For i = 1 to Ubound(aLocalTrainingArray)
Response.Write(aLocalTrainingArray (i) & “<BR>”)
Next
%>

However, as the development application model moves back toward the server-centric environ-
ment, optimizing your code to be efficient is very important. Using the preceding code listing

15.1389-6 CH11 223 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 224 Chapter 11 Controlling the Application: The Application Object

requires the upper bound of the array to be calculated every time to loop iterates. A better use
of processing power would only calculate the upper bound of the array once and then use that
variable in the For...Next loop.

The application-level arrays should only be used with information that is basically static. How-
ever, you can also use dynamic arrays in the Application and Session objects. For example, a
dynamic training class array could have been declared using the following code:
Sub Application_OnStart
Dim aMyAppTraining() ‘=== Creates a dynamic
➥array
Redim aMyAppTraining (4) ‘=== Allocates temporary
➥storage space
aMyAppTraining (1) = “Using Active Server Pages”
aMyAppTraining (2) = “Internet Architecture: The Basics”
aMyAppTraining (3) = “Content Management”
aMyAppTraining (4) = “Internet Application WorkFlow”
Application(“aTraining”) = aMyAppTraining ‘=== Assigns Application
➥variable
End Sub

The following code listing puts the GLOBAL.ASA file and the calling ASP code together in one
reference point:
==================================== The Global.ASA File =====================
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Application_OnStart
Dim aMyAppTraining(4) ‘=== Creates a fixed size
array
aMyAppTraining (1)=”Using Active Server Pages”
aMyAppTraining (2)=”Internet Basics”
aMyAppTraining (3)=”Content Management”
aMyAppTraining (4)=”Internet Application WorkFlow”
Application(“aTraining”)= aMyAppTraining ‘=== Loads local
array into Application object

Application(“LastDisplayedOn”) = Now ‘=== Sets an Application

variable
End Sub

</SCRIPT>
====================================== The Calling ASP code ==================
<%@ LANGUAGE=”VBSCRIPT” %>
<html>

<head>
<title>Reading the Application Array </title>
</head>

<body bgcolor=”#FFFFFF”>

<p><font face = “Arial”><b>Ask about special group rates for the

following classes:</b><br>
<%

15.1389-6 CH11 224 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 Using Components at the Application Level 225

if IsArray(Application(“aTraining”)) then
aLocalTrainingArray = Application(“aTraining”)
iClass = UBound(aLocalTrainingArray)
For i = 1 to iClass
Response.Write(aLocalTrainingArray(i) & “<BR>”)
Next
else
Response.Write (“Bad array”)
End if
%>
</font></p>
<BR>Last viewed somewhere at: <% Response.Write Application(“LastDisplayedOn”) %>
<%
Application.Lock
Application(“LastDisplayedOn”) = Now
Application.UnLock
%>
</body>
</html>

The previous code listing loads the training class array into memory and displays the results
on a Web page. The GLOBAL.ASA file also sets an application array LastDisplayedOn, which is
updated every time a user browses this site, by using the Lock and UnLock methods. Figure
11.2 demonstrates the results of this code listing. Part
III
FIG. 11.2 Ch
The Application
object stores the array 11
of classes in memory,
which enables any
client to access the
information.

Remember in the above example that the application array is only loaded on the
Application_OnStart event. If the information changes more frequently than the
Application_OnStart event, you may want to consider another object variable that is more
closely related to the needed frequency period.

You can also populate your application-level arrays from a database. The Amazing DataWidget
Web Site demonstrates this technique as shown in Chapter 32, “Putting It All Together: Creat-
ing an N-Tier Online Catalog Application.” This example chapter combines using the Applica-
tion and Session objects to load, create, and manipulate static arrays and database-driven
arrays to create an online catalog system.

15.1389-6 CH11 225 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 226 Chapter 11 Controlling the Application: The Application Object

The Application Collection

The Application object, like all the ASP objects, uses collections to help manage all objects
that have been given application scope. The Application object uses two collections, as
shown in Table 11.3.

Table 11.3 The Application Object’s Collections

Name Description
Contents Identifies Application object declared without the <OBJECT> tag.
StaticObjects Identifies Application object declared with the <OBJECT> tag.

The Contents Collection

The Contents collection is used to contain all items that have been created during a session
that have not been declared with the <OBJECT> tag. The Contents collection requires the follow-
ing syntax:
Session.Contents(Key)

where Key is the name of the property to retrieve. The Contents collection manages all
application-level variables declared through a user session across all pages. For example, in the
following code listing, application variables have been declared in the GLOBAL.ASA file.
Application(“aTraining”)= aLocalTrainingArray ‘=== Application object
Application(“LastDisplayedOn”)

The Contents collection can then reference these variables by using the required syntax, as
shown below:
<%
Response.Write(Application.Contents(“aTraining”))
Response.Write(Application.Contents(“LastDisplayedOn”))
%>

A more useful approach to using collections is to iterate through the collection to obtain a list
of all application-level variables. You can loop through the collection to determine whether a
variable has been declared using a For...Next loop. For example, the following code demon-
strates how to loop through all the Application.Contents objects:
<%
For Each Key in Application.Contents
Response.Write Key & “<BR>”
Next %>
%>

This code will iterate through the current Contents items to display the following list:
ATRAINING
LASTDISPLAYEDON

15.1389-6 CH11 226 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 The Application Collection 227

The StaticObject Collection

The StaticObject collection is similar to the Contents collection, but StaticObject is used to
manage all session-level objects that have been declared with the <OBJECT> tag. The
StaticObject collection uses the following syntax:

Application.StaticObjects(Key)

where Key is the property to retrieve.

The StaticObject collection can also be used to reference an application variable or multiple
application variables. For example, consider the following GLOBAL.ASA file, which declares ADO
and Ad Rotator application objects and loads an application-level array:
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Application_OnStart
Dim aMyAppTraining (4) ‘=== Creates a fixed size array
aMyAppTraining (1)=”Using Active Server Pages”
aMyAppTraining (2)=”Internet Basics”
aMyAppTraining (3)=”Content Management”
aMyAppTraining (4)=”Internet Application WorkFlow”
Application(“aTraining”)= aMyAppTraining ‘=== Loads Applica-
➥tion array variable Part

Application(“LastDisplayedOn”)=Now ‘=== Create Application variable III

End Sub
Ch
</SCRIPT> 11
=============== Create Application Level Database Connection ===============
<OBJECT
RUNAT = Server
SCOPE = Application
ID = MyDBConnection
PROGID = “ADODB.Connection”>
</OBJECT>

=============== Create Application Level Advertisement Rotator =============

<OBJECT
RUNAT = Server
SCOPE = Application
ID = MyAdrot
PROGID = “MSWC.AdRotator”>
</OBJECT>

To list all items in the StaticObject collection, use the For...Next loop to step through the
<OBJECT> collection, as shown below:

<%
For Each Key In Application.StaticObjects
Response.Write(Key & “<BR>”)
Next
%>

15.1389-6 CH11 227 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 228 Chapter 11 Controlling the Application: The Application Object

This iterative process will write the names of the Application objects, shown below, to the
page:
MyAd
MyConnection

From Here…
At this point, you can see that the Application object provides a manageable means to distrib-
ute information to all users in the ASP application. The Application object also provides vari-
able management by using the Lock and UnLock methods. This chapter also demonstrated how
to use the OnStart and OnEnd events to initialize and close your application.

The following chapters focus on using the Application object and VBScript knowledge to
demonstrate how to use the remaining ASP objects to create your Web applications. The next
four chapters are each dedicated to learning how to fully exploit the Session, Request,
Response, and Server objects.

■ Chapter 12, “Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object,” discusses how the Active Server encapsu-
lates the HTTP service to control administrative function on the Web server.
■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of
tracking a user through an ASP application.
■ Chapter 14, “Speaking to the Web Client: The Response Collection,” explains how the
Response object communicates with the requesting Web client.

■ Chapter 15, “Retrieving User Information: The Request Object,” describes how to gather
information from the user using forms, hidden fields, and cookies.
■ Chapter 17, “Implementing Existing Server-Side Components,” illustrates how to use the
various components that ship with IIS and Active Server Pages to provide rich functional-
ity to your applications.
■ Chapter 26, “Putting It All Together: Creating a Dynamic Database Application,”
illustrates how to use the Application object to store information extracted from a
database to provide database interaction for the Amazing DataWidget WareHouse.
■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,”
demonstrates how to build ActiveX components to create n-tier distributed applications
for the Amazing DataWidget WareHouse Catalog.

15.1389-6 CH11 228 10/26/98, 9:20 AM

Brands3/Stone3 SEU Active Server pages #1389-6 SRing 11/11/97 CH11 lp#3
 12
229

C H A P T E R

Expanding the Server

Controlling the Server Exploit one of the most exciting
features of ASP and IIS by using the

with the Server Object CreateObject method to add external

functionality to your HTML pages.

and Using Transaction Using Encoding Techniques

and Directory Utilities
Learn how to use URL and HTML
Processing with the encoding techniques to control
the text-based HTML output from
the HTTP server and implement the
ObjectContext Object MapPath method to manipulate files
on your Web server for Server-Side
Includes (SSI) or text-based capa-
bilities.

Regulating Server Properties

Part
Learn how to control the amount of
processing time allowed for Active III
The Server object is an ASP object that is used to control Server Page scripts. Ch
administrative features of the IIS Web server and actions
that deal with HTTP service. The Server object provides 12
the capability to create instances of Component Object
Model (COM) based components on the server. This
presents exciting possibilities to use the IIS Web server as
a translator between external clients and internal systems.
This capability to provide access to numerous internal
systems through ASP with additional server-side direc-
tives, such as file mappings and encoding techniques,
presents unlimited possibilities for your ASP
applications. ■

16.1389-6 CH12 229 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 230 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

Controlling the Server Object Using Methods

The Server object is one of the six built-in objects that are provided with Active Server Pages
and IIS 4.0. The Server object represents a programmable interface to the HTTP service that
provides a mechanism to administer and control the Web server that is traditionally found only
through the proprietary HTTP server native interfaces. The Server object is used to expose
the properties and methods of the HTTP server. Although the capability to control the HTTP
functions is a valuable resource, one of the most exciting and most powerful components of the
IIS server is its capability to create instances of the components on the server.

The Server object is referenced using the following syntax:

Server.Method/Property

where Server is the Active Server object.

The Server object uses four methods to control various aspects of the Web server, as shown in
Table 12.1.

Table 12.1 The Server Object’s Methods

Method Description
CreateObject Creates a Server instance of an object.
MapPath Translates the Web server’s virtual path settings to the physical
path on the server.
HTMLEncode Utilizes HTML encoding to deliver text to browser.
URLEncode Utilizes URL encoding techniques.

The CreateObject method enables the Web server to create instances of components running
on the server. This capability to create links to other systems or embedded application logic
gives tremendous programming functionality to the Web server. The MapPath method is used
to map the virtual settings of the Web server to physical path settings. The Server object can
also provide encoding techniques, using the HTMLEncode and the URLEncode methods, to ensure
that information is properly transferred between the Web server and the requesting client.

Expanding the Web Server with the CreateObject Method

Perhaps one of the most exciting features of Active Server Pages is the capability to create an
instance of an ActiveX object. This capability to tap into ActiveX objects provides a virtually
endless number of ways to deliver information to any Web client that were once only available
to proprietary client software or dedicated network connections. The Server object acts as a
translator between the in-house data and information stores and virtually any Web browser.

The capability of the Web server to provide a bridge between remote users and centralized
network stores is made possible by the CreateObject method. The CreateObject method
creates a connection to an instance of an ActiveX object. This connection mechanism is based

16.1389-6 CH12 230 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Controlling the Server Object Using Methods 231

on the Common Object Model (COM). COM provides the standard for objects to communicate
with one another. An ActiveX Automation Server refers to the capability for one COM object to
interact or activate another object’s properties, methods, and events. The IIS Server object
uses this communication mechanism to create a new instance of an ActiveX object using the
CreateObject method. The CreateObject method uses the following syntax:

Server.CreateObject(progID)

where progID is the class or type of object to be instantiated. The progID also requires the
following special format:
appname.objecttype

where appname is the application name hosting the object, and objecttype is the class or type
of the object to create. All COM-based objects are required to have one class type per applica-
tion name.

The CreateObject method provides the foundation for accessing ActiveX components. The
Active Server ships with five components that can be used immediately: Database Object, Ad
Rotator, Browser Component, Content Streaming Component, and the Server-Side Includes
(SSI) Component. These components will be fully discussed in Chapter 17, “Implementing
Existing Server-Side Components.” However, as an example here, the following code demon-
strates using the CreateObject method to create an instance of the Browser Component. The
Browser Component is used to determine what type of browser is requesting pages from a
Web site. Some sites might be specially designed to maximize the capabilities of a specific
browser type. Determining the requesting browser type has traditionally been accomplished
using client-side JavaScript, but now it can be accomplished using the Browser Component.
Set BrowserType = Server.CreateObject(“MSWC.BrowserType”) Part

Notice the use of the Set keyword in the preceding code snippet. The Set keyword is required III
because the CreateObject function returns a pointer to the created object. To use the methods Ch
and properties of the object, simply reference the object in any ASP file. For example, to deter-
mine whether the requesting browser is ActiveX compatible, use the Boolean ActiveX Con-
12
trols property of the Browser Component. In this situation, if the client is capable of using
client-side ActiveX controls, the function LoadActiveXControls is initiated. If the requesting
browser does not support ActiveX controls, the LoadHTMLOnly procedure is called.
<%
If BrowserType.ActiveXControls Then
Call LoadActiveXControls()
Else
Call LoadHTMLOnlyVersion()
End if
%>

You can also access the object’s built-in methods. For example, the Ad Rotator Component that
ships with IIS and Active Server Pages uses a method called GetAdvertisement. The
GetAdvertisement method is responsible for retrieving information from a scheduling file that
is used to display images on the Web page and route the user to the appropriate page. In this
example, the GetAdvertisement method is used to retrieve sale price information from a file

16.1389-6 CH12 231 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 232 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

located in the salesprices virtual directory in a file called specials.txt. Remember that the
virtual directory acts as a container for your ASP applications and is used as an alias to hide the
physical directory structure of your Web server. For more information on virtual directories,
please see Chapter 10, “ The Composition of an Active Server Application.”
<%= ad.GetAdvertisement(“/saleprices/specials.txt”) %>

The CreateObject method only creates objects that have page-level application scope. This
limited scope object is unlike the objects that were created in the GLOBAL.ASA that could either
have application- or session-level scope. Because the local objects only have page-level scope,
the objects are destroyed when the page is finished processing. If you need to access the same
component multiple times, in most situations you would benefit from using an application- or
session-level object. Creating an object once reduces the overhead required to constantly cre-
ate and destroy the object. If you have further questions about using application-level variables,
objects, or arrays, see Chapter 11, “Controlling the Application: The Application Object,” for
more information.

Managing Server Information with MapPath

The Server object uses the MapPath method to track and manage path information on the
server. The path information acts as a translator between the virtual or relative directories of
the Web server to the physical directories on the server. The MapPath method requires the
following syntax:
Server.MapPath(path)

where Server is the Active Server object and path is a physical or virtual directory.

The MapPath method is primarily used to help manage the underlying file system of the Web
server. It is particularly useful when you consider that the Internet itself is based on a text-
based delivery system that passes files from location to location. Organizing the massive num-
ber of files is not an easy task, and gathering and organizing the files into various directories
will help you in terms of structure and sanity. However, programming the ASP to find files in
various physical and virtual directories is often a daunting task. To eliminate broken links and
missing files, the MapPath method helps with ASP file management.

In most situations, you will use the MapPath method and other server variables to provide
virtual path information needed for hyperlinks or Include statements. But before the combina-
tion of the MapPath method and the different server variables is discussed, let’s investigate the
details of the Path statement.

The MapPath method provides the link between the virtual paths and the physical paths on the
Web server through the Path statement. The Path statement represents the directory struc-
ture, whether the file structure is virtual or physical, that you want to specify. There are two
basic rules to remember with the Path arguments.

■ Path arguments that start with a backslash (\) or a forward slash (/) are used to
represent virtual directories.
■ Path arguments that do not start with the backslash or forward slash represent relative
directories.

16.1389-6 CH12 232 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Controlling the Server Object Using Methods 233

The Path statement is used to give a directory location relative to either the virtual directory or
physical directory. The Path statement uses the backslash (\) and forward slash (/) to tell the
Server object to use virtual directories. For example, to map a path to the scripts virtual
directory, use the following statement:
<% = Server.MapPath(“\Scripts”)%> ‘=== Returns ‘C:\InetPub\Scripts’

The previous code example demonstrates the role of the MapPath to associate the virtual direc-
tory managed by the Web server with the physical directory structure managed by the operat-
ing system. In this situation, the MapPath statement explicitly defines the physical location
C:\InetPub\Scripts from the virtual directory \Scripts. When the forward slash is used in
the Path argument, the server processes the request in the same manner, treating it as a vir-
tual directory.
<% = Server.MapPath(“/Scripts”)%> ‘=== Returns ‘C:\InetPub\Script’

Furthermore, if just the backslash or the forward slash were used in the MapPath method, the
root directory of the Web server would be returned:
<% = Server.MapPath(“/”)%> ‘===Returns ‘C:\InetPub\wwwroot’
<% = Server.MapPath(“\”)%> ‘===Returns ‘C:\InetPub\wwwroot’

The MapPath statement is also used to manipulate relative paths by excluding the forward or
backslash characters. For example, if just the word “scripts” is represented as the Path argu-
ment, the server would return results of apparent creation of a directory from where the ASP
file was executed. In the following situation, the ASP file was run in the Sales directory.
<% = Server.MapPath(“Scripts”)%> ‘=== Returns
➥‘C:\InetPub\wwwroot\Sales\Scripts’
Part
The reason the word apparent is used to describe the return result of the previous listed
MapPath statement is because the MapPath statement does not create a non-existent directory III
or return an error if the directory does not exist. The MapPath statement simply finds the cur- Ch
rent execution directory and adds the Path value to it.
12
If the Path argument is excluded or contains a zero-length string, the MapPath method returns
an error:
<% = Server.MapPath( )%> ‘=== Returns a Wrong number of arguments Error
<% = Server.MapPath(“”)%> ‘=== Returns a missing path error

You are prevented from directly mapping to another machine via the UNC, machine name, or
IP address. If you need to map a drive to another machine, use the Internet Information Man-
ager to manipulate virtual directories.
<% Response.Write Server.MapPath(“\\207.96.34.41\Scripts”)%> ‘=== Returns an
➥‘Invalid Path Character’ error

In most situations, you will want to use the combination of the virtual directory and a filename
to point to a given source document, such as a text file or HTML file. These files can include
HTML files, images, or text files. For example, the following path generator produces a refer-
ence to a file called myNewLeads.txt in the mySalesCalls virtual directory.
<% Response.Write Server.MapPath(“\mySalesCalls\myNewLeads.txt”)%>

16.1389-6 CH12 233 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 234 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

Not only can you use the MapPath method to reference files, but you can also use it to generate
paths to a Server-Side Include. The following example uses the MapPath statement to set the
complete path location to retrieve information from an include file.
<% Response.Write Server.MapPath(“/mySalesApp/FooterInfo.inc”)%>

However, you do not have to use a fully qualified path with the MapPath method. The following
example uses the MapPath method to manage the server-side path needed to create an HTML
anchor link.
<a href = “<%= Server.MapPath(“Sales”)%>\Orders\OrderForm.htm” target = “frMain”>
‘=== Returns ‘a href = “C:\InetPub\wwwroot\Sales\Orders\OrderForm.htm” target =
“frMain”’

This MapPath method is often used in conjunction with server variables to assign relative and
virtual directory settings. One particular server variable is the PATH_INFO variable. The
PATH_INFO variable returns the virtual directory path information and the page that is being
called. For example, the following code extracts the path and page information from the cur-
rent executing page using the Request.ServerVariables variables.
<% = Request.ServerVariables(“PATH_INFO”)%> ‘=== Returns ‘/Sales/mySales.ASP’

In the above situation, an ASP file named mySales.asp is executed from the Sales virtual direc-
tory. (The Request object will be covered in much greater length in Chapter 15, “Retrieving
User Information: The Request Object.”) You can also combine the MapPath method with the
Request.ServerVariable information to meet your needs. The following example uses both
the Server and Request objects to produce the full path of the calling ASP file:
<% = Server.MapPath(Request.ServerVariables(“PATH_INFO”))%> ‘=== Returns
‘C:\InetPub\wwwroot\Sales\mySales.ASP’

Using the Server Encoding Methods

All Web technology is based on the transfer of simple text across the Internet or an intranet via
the TCP/IP protocol and the translation of that text into content within the Web browser. The
Web content and the information being displayed is controlled and manipulated by HTML tags.
These layout tags help arrange the content on the page. Because the browser only interprets
text, embedded non-ANSI characters get misinterpreted as text or simply are not available
from the keyboard. As a result of this difficulty in translating text into browser characters, the
Web server or browser often tries to process or write non-desired text. To prevent this confu-
sion from occurring, encoding methods use escape codes to explicitly tell the client what to
display by using a combination of the % character and the associated ASCII hexadecimal code
or reserved code. This is particularly important when dealing with blank spaces, accented
characters, punctuation, and non-ASCII characters. For example, notice the OnClick event of
the following button. When the button is pressed, the browser navigates to another Web loca-
tion. Because single quotes are needed to embed the desired site location, the reserved charac-
ters ‘&quot’ are used to ensure that the embedded quotes are not misinterpreted on some
clients.

16.1389-6 CH12 234 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Using the Server Encoding Methods 235

<html>
<HEAD>
<title>Embedded Reserved Characters</title>
</HEAD>
<BODY>
<FORM METHOD=”POST”>
<INPUT LANGUAGE=”VBScript”
TYPE=button
VALUE=” Login “
ONCLICK=”Window.location.href =
&quot;www.myWebApplication.com&quot;”
NAME=”btnLogin”
>
</FORM>
</BODY>
</html>

You can see the possible errors that may arise, particularly with ASP, due to the extensive use
of the <, >, and % characters. To help prevent the misinterpretation of characters, encoding
techniques—at both the HTML and URL levels—help to explicitly control characters on HTML
pages and within the URL.

Using the URLEncode Method

The URLEncode method of the Server object is centered on delivering explicit information from
the Web server to the client via the URL. Using the URL is one of the traditional ways to pass
information from a Web client to the Web server. This transferring of information via the URL
is usually accomplished using hyperlinks and forms. In both situations, information is passed
to a file on the Web server, which processes information contained in the URL.
Part
However, as mentioned previously, only ASCII-based characters can be sent via HTTP. If char-
acters such as blanks and punctuation are passed, erratic and unpredictable results will occur.
III
To ensure that all the characters entered by the user are properly passed to the server, the Ch
characters must be scanned to make sure that only known characters are being sent. For ex- 12
ample, to use the URL encoding methods, use the following syntax:
Server.URLEncode(string)

where Server is the Server object and string is the string to apply the URL encoding rules.

The server-side processing uses the filename in the URL to process information that precedes
the question mark (?) using the following syntax:
http://www.myApplication.com/processdata.asp?FieldName1=
➥FieldValue1&FieldName2=FieldValue2

The previous code example illustrates the calling of a file processdata.asp on the Web server
www.myApplication.com to process information in the URL fields named FieldName1 and
FieldName2. The file processdata.asp extracts the data from the named fields for its internal
processing. In this situation, this processing file is an Active Server Page, however, this pro-
cessing can also call executable or Common Gateway Interface scripts to process information.

16.1389-6 CH12 235 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 236 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

The processing of information via the URL is based on the simple field/value motif, where
fields are assigned a value and are passed as a string set through the URL, as shown in the
following snippet:
Fieldname = Fieldvalue

If multiple field values are used, the ampersand (&) character acts as the AND operator to concat-
enate the field values together.

The only problem with passing data via the URL occurs when you pass certain characters, such
as blanks or formatting characters. For example, if Fieldvalue1 contained a space, as in the
name “Data Widgets Inc.,” the processing page would only accept the “Data” section of the
value, because the HTTP text stream has been interrupted by a non-ASCII character and it
stops processing the assignment of that value. To enable a continuous text stream of informa-
tion to the browser, the plus character (+) is used to replace the blank spaces. There are sev-
eral other instances where the passed data must be modified to transfer data. However, the
URLEncode method makes data transfer via the URL much easier. The URLEncode method per-
forms the following processes to the data:

■ Spaces are transformed into plus (+) symbols.

■ Fields are left unencoded.
■ An unencoded equal sign assigns an unencoded field name with the data value.
■ Non-ASCII characters are transformed into escape codes.

One of the most common places where URL encoding is implemented is in link navigation or
forms-based processing. Link navigation is often used to list information, and it provides drill-
down capabilities on each link. This type of situation is most often implemented in database
parent–child relationships, where information such as customer names is displayed in a list.
If you click the customer name in the list, the customer name is passed via the URL to find the
customer details, such as shipping and contact information. The URLEncode method automati-
cally applies the encoding rules to ensure that information is properly passed in the URL.

For example, suppose you want to pass the string “Dave’s #1 Fan” via the URL. You can see
two potential problems with the string: the existence of formatting characters and blank
spaces. To successfully pass this entire string to a calling page when the user clicks a
hyperlink, apply the URLEncode method as shown in the following code:
<a href=”urlencodeServer.asp?txtInput=<% =Server.URLEncode(“Dave’s #1 Fan”)
➥%>”>Send info via a link</a>

The URLEncode method converts the string “Dave’s #1 Fan” into HTML source code that the
Web browser interprets to be the following:
<a href=”urlencodeServer.asp?txtInput=Dave%27s+%231+Fan”>Send info via a link</a>

Now, to complete the cycle, the requesting page can accept the entire string “Dave’s #1 Fan,”
formatting characters and blank spaces, into the field named txtInput.

16.1389-6 CH12 236 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 The Server Object Properties 237

Using the HTMLEncode Method

The HTMLEncode method is used by the server to explicitly define the characters to be dis-
played to a page. This method plays an important role in ensuring that information is properly
displayed on the Web page. To use the HTMLEncode feature, use the following syntax:
Server.HTMLEncode(string)

where Server is the Server object and string is the string to encode.

The HTMLEncode method is important to ensure that the proper characters are displayed on the
page and not processed by the server. For example, if you are preparing an online class on
Active Server Pages and need to display HTML and ASP syntax on your pages, the HTMLEncode
feature would be helpful to present the code instead of executing the code. If you want to dis-
play the syntax for creating an HTML table that was 100-percent wide, you would want to dis-
play, but not execute, the <Table Width = 100%> syntax. To accomplish this, use the
HTMLEncode method as seen below:

<% =Server.HTMLEncode(“<Table Width=’100%’>”) %>

If you view the source code at the browser, you will notice that the previous code was inter-
preted as the following:
&lt;Table Width=’100%’&gt;

The same process is directly applicable to ASP syntax. For example, if you want to display the
syntax for the Session.Abandon method and you write the correct syntax—<% Session.
Abandon %>—on the ASP source page, when the page executes your Session object will be
destroyed. Using the HTMLEncode method ensures that escape sequences are properly inserted
for any character that might be misinterpreted as execution characters. Part

<%= Server.HTMLEncode(“<% Session.Abandon % >”) %> III

The Server.HTMLEncode method translated the greater than and less than (> and <) symbols Ch
into &lt and &gt characters. 12
&lt;% Session.Abandon % &gt;

This encoding technique is more suitable when applied to ensure the proper data transfer of
information from the Web server to the browser, particularly when you are writing information
from a database. For example, you can use the same encoding technique to retrieve values
from a database RecordSet object. In the following situation, the HTMLEncode method is applied
to write the “CompanyName” field to the ASP page.
<%Response.Write Server.HTMLEncode(rs.Fields(“CompanyName”).Value)%>

The Server Object Properties

The Server object has only one property, as shown in Table 12.2. The ScriptTimeout is used to
set the amount of time the Server will process the ASP script before terminating the request.

16.1389-6 CH12 237 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 238 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

Table 12.2 The Server Object’s Property

Method Description
ScriptTimeout Sets or retrieves the maximum amount of seconds the server
will process the ASP.

The ScriptTimeout property is a property of the Server object that prevents a process from
running endlessly. Without a limitation to control how long a process is allowed to stay alive,
run-away threads can consume large amounts of system resources and leave the requesting
browser waiting endlessly for a response. A ScriptTimeout event indicates that something has
gone wrong within the ASP code or something has gone wrong with an external call to an
instantiated object.

In either case, the Server object relies on the ScriptTimeout property to prevent the situation
from happening. The ScriptTimeout property sets the timeout values at both the server level
and the page level. The default script timeout value for all Active Server Pages is stored in the
ScriptTimeout Registry in IIS 3.0 or the Metabase in IIS 4.0. The default timeout value for
script processing is 90 seconds. If a ScriptTimeout event occurs, a Timeout event is written to
the event log of the Web server.

The script timeout setting can also be controlled on the individual ASP. This page-level control
is used to override the default timeout value indicated in the Registry. Just as the default script-
ing language can be overridden at the page level, the same capability exists for the timeout
value. The ScriptTimeout value requires the following syntax:
Server.ScriptTimeout = Seconds

where Seconds indicates the number of seconds allotted for page-level scripting to process.

After this amount of time has been reached, the Server object stops processing the script and
writes an event to the event log. To determine the timeout value for your server, use the follow-
ing code:
<% Response.Write Server.ScriptTimeout %>

If the default settings are still in place, the code should return 90.

For example, the following code overrides the default timeout value by setting the page-level
timeout to 20 seconds and generating an infinite loop to trigger the ASP Timeout event.
<%
Server.ScriptTimeout = 20
dim iCount
iCount = 0

Do While iCount < 10

If iCount = 9 Then
iCount = 1
End If
iCount = iCount +1
Loop
%>

16.1389-6 CH12 238 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Using the ObjectContext Object for ASP Transaction Processing 239

After the script times out, an ASP 0113 Script Timed Out error displays the following mes-
sage to the user:
error ‘ASP 0113’

Script timed out

/TimeOutExample.asp

The maximum amount of time for a script to execute was exceeded. You can change
this limit by specifying a new value for the property Server.ScriptTimeOut or by
changing the value for ScriptTimeout in the registry.

At the same time, the Web server log file also records and describes the error:
207.96.34.41, -, 7/3/97, 21:58:52, W3SVC, [Server Info] /TimeOutExample.asp,
|ASP|-|ASP 0113|Script timed out

The location of your Web server log can be found on the Logging tab of the Internet Server
Manager. The logs can be written to a text file or written to an ODBC-compliant database.

Using the ObjectContext Object for ASP Transaction

Processing
With the Internet Information Server 4.0, Active Server Pages have a direct interface to con-
trolling components in the Microsoft Transaction Server. This interface is made possible
through the ObjectContext object. The ObjectContext is used to help manage and define the
conditions of an instantiated object managed by the Microsoft Transaction Server (MTS).
Part
These conditions are used to help identify properties of the object, such as what process cre-
ated the object or when is the unit of work complete. The context properties are used by the III
Microsoft Transaction Server (MTS) to allocate resources to the instantiated object. To sup- Ch
port transaction processing in your ASP, both of the components managed by MTS and the
ASP must be properly configured.
12

Configuring the MTS components

The MTS manages components by setting four transaction properties, as shown in Table 12.3.
These properties enable you to configure different levels of transaction support.

Table 12.3 The MTS Component Properties

Property Description
Requires a transaction Sets the component object to execute within the
scope of the transaction.
Requires a new transaction Sets the component object to execute only within its
own transaction.

continues

16.1389-6 CH12 239 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 240 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

Table 12.3 Continued

Property Description
Support transactions Sets the component object to execute within the
scope of the client’s transactions.
Does not support transactions Sets the object, does not run within a transaction.

In order to use transaction-level processing, the Components property must be set to either
Requires a transaction, Requires a new transaction, or Support transactions. For more
details on how to create your own customized components, please see Chapter 18, “Creating
Your Own Active Server Components.” For more information on setting up and configuring
MTS to manage your components, please see Chapter 28, “Working with the Microsoft Trans-
action Server.”

Setting Up ASP to Support Transactions

Active Server Pages use the Transaction directive to initiate page-level transactional process-
ing. When the scripting engine processes this directive, the Active Server Page is executed as
a transaction on the Transaction Server. To initiate your page as a transaction, use the follow-
ing directive syntax in the first line of your ASP:
<%@ Transaction = Required %>

This interaction between the Web server and the MTS enables direct commit and rollback of
an MTS object directly from an Active Server Page. Now your application logic can be consoli-
dated in an Active Server Page, while core-processing functionality is left to the MTS compo-
nents.

N O T E A transaction cannot occur over multiple Active Server Pages; as a result, MTS components
should not be stored in Application or Session objects. ■

Controlling Success and Failure with Methods

The ObjectContext is used to identify when that object transaction is complete. In addition,
you can control the object’s completed task by immediately accepting its results, temporarily
delaying the results, or completely rejecting the object’s work. This capability to control
whether or not the object’s work is accepted is made possible through the SetComplete and
SetAbort methods, as shown in Table 12.4.

Table 12.4 The ObjectContext Methods

Method Description
SetComplete Sets the work of an object as a success and permanently accepts
the changes to the resource.

16.1389-6 CH12 240 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Controlling Success and Failure with Methods 241

Method Description
SetAbort Sets the work of an object as a failure and returns the resource to
it original state, neglecting any changes made to the resource.

The SetComplete Method

The SetComplete method is used to commit an MTS object’s transaction as a success. The
changes made by the object method on a resource are permanently committed. To accept
changes made by object, use the following syntax:
ObjectContext.SetComplete

For example, assume that an MTS object is used to transfer funds from a credit table into a
debit table, as portrayed in the following example:
<%@Transaction = Required
Set AccountTransfer = Server.CreateObject(“myCompany.AccountProcessing”)
If AccountTransfer.MoveFunds(“Credit”, “Debit”, “300000”) then
‘=== Transfer of funds was successful
ObjectContext.SetComplete
Else
‘=== Failure, Send transaction fee for wasting packets
ObjectContext.SetAbort
End if
%>

In this example, if the transfer of $300,000 is successful from the Credit account to the Debit
account, the SetComplete method finalizes this transfer of funds. However, if the funds were
not available, the transaction is rolled back to its original state. Part

After the SetComplete method is executed, the OnTransctionCommit event, if available, is pro-
III
cessed on the script. If the subroutine is not available, the Active Server Page continues pro- Ch
cessing the script.
12
The SetAbort Method
The SetAbort Method is used to roll back changes as a result of an object method. When the
SetAbort method is called, the object rejects the completed work and returns the transaction
back to its original state. To reject changes made by the MTS object, use the following syntax:
ObjectContext.SetAbort

The previous example demonstrated the use of the SetAbort method to roll back transactions
made by an MTS object.

After the SetAbort method is executed, the OnTransactionAbort event, if available, is pro-
cessed on the script. If the subroutine is not available, the Active Server Page continues pro-
cessing the script.

16.1389-6 CH12 241 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 242 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

Trapping the ObjectContext’s Events

The ObjectContext has two events that are triggered after the ObjectContext’s methods are
executed, as seen in Table 12.5. The OnTransactionCommit event is triggered after the
SetComplete method. The OnTransactionAbort event is triggered after the SetAbort method
is called.

Table 12.5 The ObjectContext’s Events

Event Description
OnTransactionCommit Event triggered after the ObjectContext.SetComplete
method is executed.
OnTransactionAbort Event triggered after the ObjectContext.SetAbort
method is executed.

OnTransactionCommit Event
The ObjectContext object SetComplete method triggers the OnTransactionCommit event.
When the SetComplete method is processed, the script will process the OnTransactionCommit
subroutine. The simplest example using the OnTransactionCommit event is demonstrated
below:
<%@TRANSACTION = Required %>
<%
Sub OnTransactionCommit()
Dim strMessage
strMessage =”This was generated by executing the SetComplete method “
strMessage = strMessage & “to trigger the OnTransactionCommit event”
Response.Write strMessage
End Sub
%>
<HTML>
<HEAD>
<TITLE>Set Complete</TITLE>
</HEAD>
<BODY>

<%
ObjectContext.SetComplete
%>

</BODY>
</HTML>

The example initiates transaction processing using the TRANSACTION = Required directive and
executes the SetComplete method to trigger the OnTransactionCommit() event. In this ex-
ample, the OnTransactionCommit subroutine writes the following text to the page:
This text message was generated by executing the SetComplete method to trigger
the OnTransactionCommit event

16.1389-6 CH12 242 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 Trapping the ObjectContext’s Events 243

Of course, you are not limited to just writing messages to the page. For example, the following
example uses a custom-created ActiveX control to transfer account data.
<%@TRANSACTION = Required LANGUAGE=”VBScript”
Sub OnTransactionCommit()
Response.Redirect “CreateVirutalReceipt.ASP”
End Sub
Sub OnTransactionAbort()
Response.Redirect “InsufficientFunds.ASP”
End Sub

Set AccountTransfer = Server.CreateObject(“myCompany.AccountProcessing”)

If AccountTransfer.MoveFunds(“Credit”, “Debit”, “300000”) Then
‘=== Transfer of funds was successful
ObjectContext.SetComplete
End if
%>

In this code example, if funds were properly transmitted, the ObjectContext.SetComplete

method triggers the OnTransactionCommit() event of the code to route the user to an ASP
page called CreateVirutalReceipt.ASP.

OnTransactionAbort Event
Similar to its counterpart, the OnTransactionAbort event is triggered when the ObjectContext
object SetAbort method is executed. When the SetAbort method is processed, the script will
process the OnTransactionCommit event. To illustrate using the OnTransactionAbort() event,
consider the following example:
<%@TRANSACTION = Required %>
Part
<%
Sub OnTransactionAbort() III
Dim strMessage
strMessage =”This text message was generated by executing the SetAbort Ch
➥method “
strMessage = strMessage & “to trigger the OnTransactionAbort event” 12
Response.Write strMessage
End Sub
%>
<HTML>
<HEAD>
<TITLE>Set Abort</TITLE>
</HEAD>
<BODY>

<%
ObjectContext.SetAbort
%>

</BODY>
</HTML>

16.1389-6 CH12 243 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 244 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

The example initiates transaction processing using the TRANSACTION = Required directive and
executes the SetAbort method to trigger the OnTransactionAbort() event. In this example,
the OnTransactionAbort subroutine writes the following text to the page:
This text message was generated by executing the SetAbort method to trigger the
OnTransactionAbort event

Just as you saw in the OnTransactionCommit event, you can expand the role of your ActiveX
components to now handle success and failures for transferring account information. The
following example demonstrates using both the OnTransactionAbort() and
OnTransactionCommit() events.

<%@Transaction = Required LANGUAGE=”VBScript”

Sub OnTransactionCommit()
Response.Redirect “CreateVirutalReceipt.ASP”
End Sub
Sub OnTransactionAbort()
Response.Redirect “InsufficientFunds.ASP”
End Sub

Set AccountTransfer = Server.CreateObject(“myCompany.AccountProcessing”)

If AccountTransfer.MoveFunds(“Credit”, “Debit”, “300000”) then
‘=== Transfer of funds was successful
ObjectContext.SetComplete
Else
‘=== Failure, Send transaction fee for wasting packets
ObjectContext.SetAbort
End if
%>

The previous code example fails to properly transfer account information, the
ObjectContext.SetAbort methods explicitly rolls back the transaction and executes the
OnTransactionAbort() subroutine.

From Here…
In this chapter, we have covered the ASP Server object, which is used to control objects on the
Web server. The CreateObject method provides capabilities to instantiate objects on the Web
server to add functionality to Active Server Pages. These components, such as the Browser
Capability Component and the Database Access Component, are discussed in further detail in
Chapter 17, “Implementing Existing Server-Side Components.” The MapPath method is used to
provide an interface between the virtual and physical directory paths mappings. The Server
object also uses encoding techniques to ensure that information from the Web server is prop-
erly transferred between the client and server.

For related information, please see the following chapters:

■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of
tracking a user through an ASP application.
■ Chapter 15, “Retrieving User Information: The Request Object,” depicts how to gather
information from the user using forms, hidden fields, and cookies.

16.1389-6 CH12 244 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 From Here… 245

■ Chapter 17, “Implementing Existing Server-Side Components,” discusses the various

components that ship with IIS and Active Server Pages.
■ Chapter 18, “Creating Your Own Active Server Components,” walks you through the
tools and examples on how to create tailored functionality via custom-created components.
■ Chapter 19, “Putting It All Together: Creating an Online Registration System,” illustrates
how to implement the different ActiveX components that ship with the Internet Informa-
tion Server to create a Web site for the Amazing DataWidgets WareHouse online catalog.

Part
III
Ch

12

16.1389-6 CH12 245 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 246 Chapter 12 The Server Object and Transaction Processing with the ObjectContext Object

16.1389-6 CH12 246 10/26/98, 9:21 AM

Brands SEU Active Server Pages #1389-6 SRing 11/13/97 CH12 lp#3
 13
247

C H A P T E R

Tracking User Workflow with

Managing the User Session Variables
Discover how session variables are

Session: The Session used to store information that can

be accessed from any page in your
application.

Object Implementing the Session

Object’s Variables, Properties,
and Events
Learn the specifics of the Session
object and how to use its properties
to manage the user session, and use
the OnStart and OnEnd events to
One of the biggest shortcomings of deploying applications initialize ASP code when a new user
over the Web is the inability to track application workflow. enters your application.
The connectionless protocol, HTTP 1.0, is used only to
quickly send and receive ASCII-based information. Conse- Managing User Resources on
quently, Web-based applications can rely either on continu- the Server
ously updating an application tracking database on the Learn how to set session resources
Web server or writing excessive amounts of information to needed to control processing and
client-side cookies. With the rollout of Active Server resources on the Internet Informa-
tion Server.
Pages, a server-side Session object combines the func-
tionality of both of these user-tracking methods. The Using Session-Level Part
Session object gives cookie-based browsers a method to
track user-specific information across different Active
Components III
Discover how to use session-level Ch
Server Pages. The user-specific information can be stored
objects to refer to an instantiated
and accessed from any page in the ASP application by the
component across different Web 13
individual user. This tracking mechanism stores an pages that are limited only to a
equivalent session fingerprint in a client-side cookie that is specific user.
used to reference the actual data stored in memory on the
Web server. The data, accessible to the user through the Understanding the Session
entire application, is accessible only to the specific user Collection
that created the information. This chapter discusses how Learn how to use the Session collec-
to implement the different aspects of the Session object, tion to help manage session-level
as well as how to track user state information on browsers objects within your ASP application.
or networks that do not support cookies. ■

17.1389-6 CH13 247 10/26/98, 9:22 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 248 Chapter 13 Managing the User Session: The Session Object

Using Session Variables to Track User Workflow

The Session object tries to overcome the shortcomings of developing workflow applications
over a stateless protocol. By their very nature, the Web server and HTTP are used only to
transfer and process discrete requests for information. Once the request is processed and the
information delivered, the connection between the client and server is released. Although this
“accept and release” policy provides great scalability for delivering information to a large
number of requests, it presents a fatal flaw in the application development arena. The Session
object provides the capability to track user information throughout an ASP application. The
tracking functionality is largely based on setting a unique user ID when the Web client first
enters the ASP application. That ID is then transferred back and forth between the Web client
and the Web server during additional requests to monitor and track the specific user informa-
tion. The stored session information is unique to the specific user and cannot be shared or
accessed between the requesting users. If information needs to be shared between different
users, that information should be assigned to an Application object. For more information on
using the Application object, please see Chapter 11, “Controlling the Application: The Appli-
cation Object.”

Identifying the User

The capability to uniquely identify a user and track the user location in a Web application has
been a sore spot for Web application developers. Excruciating tricks and workarounds were
implemented in the past to try to overcome Web process management and browser-based
navigation techniques. Even though there have been great advances in Web application devel-
opment tools, the same processes of passing values through URLs, hidden fields, cookies, or
databases still prevail in today’s Web development environment.

The unique identification of the user and tracking user information based on a unique ID is the
foundation of the ASP Session object. The Session object presents many benefits that help
decrease the amount of development time required to build Web-based applications for cookie-
compliant browsers. The Session object, like the Application object, enables developers to
trap when the user enters and leaves an application. The Session object is created when a new
user enters the ASP application. A new user is a user that does not have a Session object cur-
rently created for him.

The Session object acts as a subset of the Application object. In Chapter 11, you saw that the
Application object is responsible for managing information and resources that are needed in
an ASP application. As a result, when the ASP application is shut down the dependent Session
object is also destroyed.

The Session object is based on using cookies to store and transfer a unique user ID between
the browser client and the Web server. A cookie is a text file used by the Web browser to store
information. This unique ID, the SessionID, is used to create and reference server-side objects
specific to a particular user. If cookies are not permitted on the client browser, because of
firewall issues, browser incompatibility, or desktop/network security concerns, the Session

17.1389-6 CH13 248 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Using Session Variables to Track User Workflow 249

object is rendered useless. If cookies are not permitted, ASP applications can still be developed
and deployed. However, most of the state tracking will have to be done through more cumber-
some methods, such as using HTML hidden fields and passing information via the URL.

When a user enters the application, the ASP Server object first checks to see if the requesting
user has a valid SessionID. If the SessionID is found in the requesting HTTP header, the user
is identified as an active user and is able to continue in the application. The term “active” user
is used to describe a user that has not left the application, either by logging out of the applica-
tion or triggering a Session Timeout or Abandon method.

If a valid SessionID is not found in the HTTP header, the Server object generates a unique
identifier and sends it to the browser to be stored in a temporary cookie. This identifier is
needed to create a unique Server object that represents that specific user session. The purpose
of the identifier is to generate an exclusive identification to tie the actions of the browser client
to corresponding objects on the Web server. After the SessionID is assigned, the browser
passes a unique token to identify itself and its associated server-side variables. For more infor-
mation on the Server object, please see Chapter 12, “Controlling the Server with the Server
Object and Using Transaction Processing with the ObjectContext Object.”

CAUTION
Session objects are only supported on Web browsers and networks that support cookies.

The Problem with Cookies

In a cookie-friendly world, the Session object greatly helps manage tracking user state in an
ASP application. However, when dealing with non-cookie–compliant browsers or networks, it is
not possible to track user information via the Session object. Typically this capability to
uniquely identify a user is of importance when the possibility of client impersonation might
occur. This counterfeiting of a user can occur in two situations. The first is if the SessionID is
captured while the data is in transit between the client and server. The second possible secu-
rity violation occurs when the cookie file itself is copied from the user machine to another Part
machine. With a valid SessionID, a potential hacker could temporarily clone himself as a valid
user.
III
Ch
To prevent these possible security infractions, three levels of security must be addressed:
access to the Web server, in-transit data security, and user verification. 13
N O T E To prevent security lapses:

● Use SSL to encrypt in-transit data.

● Control access and permissions using NTFS on the Web server.

● Issue digital certificates to ensure user authentication. ■

17.1389-6 CH13 249 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 250 Chapter 13 Managing the User Session: The Session Object

Secure socket layers (SSLs) can be implemented to ensure data integrity when the information
is being passed between the Web server and browser client. SSL encrypts the in-transit data
between the Web server and client to prevent the packets from being easily interpreted for
valuable information.

Server-side security can also be implemented by using the Windows NT File System (NTFS)
security. NTFS enables you to implement security on the Web server by using user or group
permissions. Furthermore, SSL client certificate authentication can be used to make the re-
questing client prove its identity using a client certificate. Once the client is proved to be
trusted, a digital identification is assigned to the user. Therefore, you can require the digital
certificate before allowing clients access to your application. For more information on setting
security access and settings for your Web applications, see Chapter 31, “Using the Internet
Information Server 4.0 to Manage Your Web Applications.”

In addition to having to having to worry about firewall and network security considerations, the
individual browser has the capability to accept or reject cookie information.

Active Server Pages also enable you to read and write information to a cookie file. For more
information on how to use cookies to store information on the client’s browser, please see
Chapter 14, “Speaking to the Web Client: The Response Collection.” To extract information
from the client’s browser, see Chapter 15, “Retrieving User Information: The Request Object.”

Implementing Session Variables and Properties

The Session object is responsible for managing user information. To start tracking and ma-
nipulating user workflow, you can start using the Session object’s methods and properties
using the following syntax:
Session.property|method

Where Session is the Session object. The Session object has four properties, as listed in Ta-
ble 13.1.

Table 13.1 The Session Object’s Properties

Property Description
SessionID Returns the unique user session identifier.
TimeOut Returns or sets the user timeout value in minutes.
CodePage Sets the language attribute for the deployed pages.
LCID Sets the local identifier used to set local date, currency, and time formats.

17.1389-6 CH13 250 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Implementing Session Variables and Properties 251

Declaring Session Variables

Session variables are declared in a manner similar to declaring application variables. To create
or reference a session variable, use the Session object and name of the variable with the follow-
ing syntax:
Session(varName)

where varName is the name of the session-level variable. Remember that session-level variables
are only available to the user session that created the object and cannot be shared among sepa-
rate user sessions.

If you want to create a session-level variable, simply assign a value to a session variable name.
There is no declaration statement that exists for the session- or application-level variables.
Therefore, you do not have to use the Dim, Redim, Public, Private, or Const declared state-
ments that are found in VBScript to explicitly create variables. For example, to create a new
session variable called SalesUserName and assign a value of Wynn to that variable, use the fol-
lowing code:
<%
Session(“SalesUserName”) = “Wynn” ‘=== Creates and assigns session variable
Session(“SalesToDate”) = 24
%>

This same syntax that is used to create a session variable is also used to reference an existing
variable name. For example, to change the session variable SalesUserName value to Lois, use
the same syntax.
<%
Session(“SalesUserName”) = “Lois” ‘=== Reassigns value of session variable
Session(“SalesToDate”) = 28
%>

Because the session variables are only available to the specific user session, no Lock and Un-
lock methods are needed to prevent simultaneous updates to the same variable. However,
great care should be taken to ensure the correct spelling of session- and application-level vari-
ables. If a session or application variable is misspelled, a new variable with that misspelling is Part
created.
III
One of the most common uses of the Session object is to set database connection properties. Ch
The database connection properties are recommended to be made at the session level to allow
you to tap into the new features of ODBC 3.5 and component management using Microsoft 13
Transaction Server. The Session_OnStart event of the GLOBAL.ASA file is a convenient location
to store all the variables needed to connect to the database, as shown in the following code:
Sub Session_OnStart
Session(“ConnectionString”) = “DSN = myIntraNet”
Session(“ConnectionTimeout”) = 15
Session(“CommandTimeout”) = 30
Session(“RuntimeUserName”) = “”
Session(“RuntimePassword”) = “”
End Sub

17.1389-6 CH13 251 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 252 Chapter 13 Managing the User Session: The Session Object

Notice in the previous code that the actual database connection is not established in the
Session_OnStart event. Instead, the connection should be opened and closed only as needed
in the ASP code on the individual ASP pages. The reason for this “connect only as needed”
policy is to enable ODBC 3.5 and the Transaction Server to manage your database connection
and pooling capabilities. Opening the database connection only when you need to gives you
more scalability in terms of overall connection resources.

This approach is fundamentally different from the traditional database connection environment
where connecting to and disconnecting from a database required significant resources. Even if
ODBC 3.5 is not used, dedicating a database connection from the time a user enters the appli-
cation is not the most efficient database management technique. For example, how are you to
know that the requesting client will use the dedicated database connection? How do you pre-
vent running out of database connections? Connecting and optimizing your database connec-
tion will be discussed in Chapter 28, “Working with the Microsoft Transaction Server.” In addi-
tion, Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,” will
demonstrate building ActiveX database components that use the Transaction Server to manage
database connections.

The SessionID Property

The Session object uses a SessionID to keep track of user information from page to page
within the ASP application. The Web server generates the SessionID when a new session is
started. The SessionID is available using the following syntax:
Session.SessionID

where Session is the Session object.

A new session is started if no existing Session objects exist for the requesting user. Only one
Session object exists per user per Application object. To display the SessionID, use the
SessionID property as illustrated below:

<% = Session.SessionID %>

or
<% Response.Write(Session.SessionID) %>

For example, the following script displays the user’s SessionID:

Your SessionID is: <% Response.Write(Session.SessionID) %>

This script would return the following results with the individual user’s ID:
Your SessionID is 24214

The Timeout Property

The Timeout property sets or returns the amount of time, in minutes, that the Session object
can remain inactive before the user’s Session object resources are released. This property
is necessary because the stateless HTTP protocol prevents the application from knowing
whether the user has actually finished the application or has simply left the application
and navigated to another site. The Session object tries to overcome this lack of a persistent

17.1389-6 CH13 252 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Implementing Session Variables and Properties 253

connection by using the Timeout property. To prevent an unlimited number of Session objects
from being created on the server, the Timeout property helps to conserve server resources by
destroying unattended user sessions. To set the Timeout property, use the following syntax:
Session.Timeout [ = nMinutes]

where Session is the Session object and nMinutes is the timeout value specified in minutes.

To display the default Session.Timeout value for your Web server, use the following code:
<% = Session.Timeout %>

Unless otherwise stated, IIS 3.0 uses the Registry to regulate the session timeout value. IIS 4.0
stores all configuration information in a new storage location called the Metabase. To set the
Timeout value in the Metabase, use the Property tab of the specified Web site in the Microsoft
Management Console.

The default value of the SessionTimeOut Registry setting is 20 minutes. After the specified
amount of time has passed and no activity has occurred with the Session object, the server will
destroy and release all resources related to that session.

To temporarily override the Timeout setting, you can specify a new timeout value by assigning
a value to the Timeout property directly in the ASP file. This capability to set the individual
timeout values was important in the IIS 3.0 world when individual Web applications could not
have their own unique settings. IIS 3.0 only enables settings to be applied at the HTTP server
level, not at the individual application levels. Therefore multiple applications running from the
same HTTP server all had to use the same application settings. However, with IIS 4.0, indi-
vidual applications can have their own unique settings. In any case, the Timeout value enables
the developer to specify the appropriate timeout value without relying on the Web site adminis-
trator to make the change. The following example sets a new Session Timeout value of five
minutes:
<% Session.Timeout = 5 %>

The reason the word “temporary” is used to describe the change in the Session.Timeout
property is because when a new Session object is created, the Session object pulls the default
Part
session timeout value from the Registry. Therefore, unless reassigned either in the
Session_OnStart or at the page level, the new Session object uses the timeout value specified III
in the Registry. Ch
To illustrate this point, the following code writes the current session timeout value and current 13
SessionID value to a page. Then the ASP changes the Session.Timeout property from the
default value to one minute.
<%
Response.Write “Existing Session ID: “ &Session.SessionID & “<BR>”
‘=== Writes SessionID
Response.Write “Existing Session Timeout: “ &Session.Timeout & “<BR>”
‘=== Writes TimeOut value
Session.Timeout = 1 ‘=== Sets New Timeout value to 1 minute
Response.Write “New Session Timeout: “ &Session.Timeout & “<BR>”
‘=== Writes the new TimeOut value
%>

17.1389-6 CH13 253 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 254 Chapter 13 Managing the User Session: The Session Object

The previous code returns the following results:

Existing Session ID: 35441
Existing Session Timeout: 20
New Session Timeout: 1

After this code is executed the first time, the Session Timeout value is set to one minute. If the
code is executed again after one minute, a new SessionID will be written to the page and the
default session timeout value will be assigned.

To further illustrate the destruction of the Session object and its resources, the previous code
is modified to include a session variable named myUserName.
<%
Response.Write(Session(“myUserName”)) & “<BR>” ‘=== Writes blank first time
➥page is called
Response.Write(Session.SessionID) & “<BR>” ‘=== Writes the SessionID
Response.Write(Session.Timeout) & “<BR>” ‘=== Writes the default
➥timeout value
Session(“myUserName”) = “FlintStone” ‘=== Assigns value to variable
Session.Timeout = 1 ‘=== Re-assign new timeout
➥value
%>

The first time the code is processed, the session variable myUserName has no value and conse-
quently returns a zero length string. After writing the SessionID and timeout values to the
page, the session variable myUserName is assigned the value FlintStone. If the page is refreshed
or called again within one minute, the first line of code now writes FlintStone to the page. The
remaining code writes the same SessionID and the newly assigned timeout value to the page.
However, if the page is refreshed after the allotted timeout value of one minute, the old Session
object is destroyed and replaced by a new Session object. The results now write a zero length
string for the session variable myUserName, a new SessionID, and the default timeout value.
These various effects of destroying the Session object can be seen in Table 13.2.

Table 13.2 Result from the Session.Timeout Sample Code

Refresh Before Refresh After
ASP Code Initial Execution Timeout Timeout
Session(“myUserName”) - FlintStone -
Session.SessionID 13530 13530 16384
Session.Timeout 20 1 20

The LCID Property

The LCID property is used to set the local identifier properties for an ASP. The local identifier
is used to control the display formatting that is specific to a localized location or region. The
LCID property requires the following syntax:

Session.LCID(=LCID)

17.1389-6 CH13 254 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Implementing Session Variables and Properties 255

where LCID is the valid local identifier. For example, to set the local identifier to use the French
local identifier, use the following code:
<%
Session.LCID = 1036
%>

The LCID is a standard abbreviation that identifies localized formatting issues, such as time,
date, and currency formats. If you want to take advantage of different regional settings to dis-
play formatted information, you would have to change the regional settings on the Web server
itself. For example, if the Regional Settings for the Web server are set to French (Luxembourg),
the assigned time, date, and currency formatting issues would look like the following script:
<%
Dim amtDue
amtDue = 30
Response.Write(“The amt Due was printed on “ &FormatDateTime(Now) &”<BR>”)
Response.Write(“Your total amount is “ &FormatCurrency(amtDue)&”<BR>”)
%>

This script would produce the following result:

The amt Due was printed on 30/09/97 13:43:55
Your total amount is 30,00 F

However, instead of limiting the Web server to displaying information to only one language
type, the LCID property enables you to set this LCID for the ASP file. For example, the follow-
ing ASP uses the French LCID to display information in the stand French LCID without chang-
ing the display format for the entire Web server.
<%
Session.LCID =1036
Dim amide
amide = 30
Response.Write(“The was printed on “ &FormatDateTime(Now) &”<BR>”)
Response.Write(“Your total amount is “ &FormatCurrency(amtDue)&”<BR>”)
%>

The previous example writes the current date, time, and currency values to the page using the Part
French regional setting. For example, the French date formatting is used to display the current III
date and time, 30/09/97 13:43:55. If the regional settings were set on English (United
States), and the Session.LCID setting were removed, the previous code example would pro- Ch
duce and format results specific to that region, as follows: 13
The was printed on 9/30/97 1:54:03 PM
Your total amount is $30.00

Keep in mind that you can also set the LCID using the LCID directive. The syntax for the LCID
directive is the following:
<%@ LCID = a local identifier %>

For more information on using directives, please see Chapter 10, “The Composition of an Ac-
tive Server Application.”

17.1389-6 CH13 255 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 256 Chapter 13 Managing the User Session: The Session Object

The CodePage Property

Active Server Pages also provide additional support for multi-lingual sites by supporting mul-
tiple character maps. A specific language’s character map consists of that language’s letters,
numerals, and punctuation characters. The CodePage property of the Session object is used to
assign the specific character map for an ASP. To implement the CodePage feature, use the fol-
lowing syntax:
Session.CodePage = CodePage

where CodePage is a valid code page for the scripting engine. For example, to set the CodePage
property to Japanese, use the Japanese CodePage setting:
<% Session.CodePage = 950 %>

Similar to the LCID, CodePage can also be set using the CodePage directive with the following
syntax:
<%@ CodePage = valid code page %>

For more information on using directives, please see Chapter 10.

Trapping Session Events

The Session object uses two events, shown in Table 13.3, to symbolize when a user enters and
exits the ASP application: the Session_OnStart and Session_OnEnd events. A new Session
object is created by the Active Server to represent when a new user has started to use the ASP
application. After the Session object is started, information pertaining to that user’s session
can be stored on the Web server.

Table 13.3 The Session Object’s Events

Event Description
Session_OnStart Triggered when a new user enters an ASP application.
Session_OnEnd Initiated when the ASP Session object is terminated.

Session_OnStart
The Session object, like the Application object, uses two events in the GLOBAL.ASA file to
execute code on the session starting and ending events. The code contained in these events is
processed before the requested page is processed. The biggest difference between the two
objects is scope. The Application object is used to share information that is global to all users
in the ASP application, whereas the Session object is used to store and track information spe-
cific to the individual user. Also, remember that the GLOBAL.ASA file is an optional file that does
not have to be used in order to create Session and Application objects. The GLOBAL.ASA sim-
ply represents a central location to help administer application- and session-level objects and
events.

17.1389-6 CH13 256 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Trapping Session Events 257

To execute code at the start of a new Session object, you must use the GLOBAL.ASA file. The
Session_OnStart event requires a layout similar to the following GLOBAL.ASA syntax:

<SCRIPT LANGUAGE = ScriptLanguage RUNAT = Server>

Sub Session_OnStart
‘=== The Code to be executed when a new user enters the app
End Sub

</SCRIPT>

The Session_OnStart is often used to set properties that will be used throughout the user’s
stay in the application. One of the most common uses of the Session_OnStart event is to define
settings for the user session. For example, in most applications, you need to define constants
that are used in the application. You can set these constants by placing the variable declarations
in the Session_OnStart event, as shown below:
Sub Session_OnStart
const MaxOrders = 25
const MaxRequests = 10
const MaxAmount = 15
End Sub

Session events can reference any other ASP object in the OnStart and OnEnd events. The fol-
lowing example demonstrates the Session_OnStart event’s capability to access application-
level variables to create a counter to track the amount of users who browse the application.
Further examples in this section will demonstrate using other objects, such as the Response
object, in the Session_OnStart event.
Sub Session_OnStart
Application.Lock
Application(“iUserCounter”)= application(“iUserCounter”) + 1
Application.UnLock
End Sub

Because the Session_OnStart event starts when a user initiates a user session, this is often the
position where you will want to read cookie information from the client machine. Cookie infor-
mation is often used to provide a temporary storage location to store user-specific information,
Part
such as a customized display or available options within an ASP application. For example, you
may want to determine the last time that a user entered an application so you can display any III
new information or news. This process of retrieving information from a cookie is made possible Ch
by using the Cookie collection of the Response and Request ASP objects. The Response object
is responsible for writing information to the client browser and the Request object is respon- 13
sible for retrieving information from the browser. Chapters 14 and 15 focus on the details of
client interaction in more detail. For example, the following code extracts the user’s favorite
search engine from a cookie file:
Sub Session_OnStart
SearchEngine = Request.Cookies(“SearchEngine”)
End Sub

17.1389-6 CH13 257 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 258 Chapter 13 Managing the User Session: The Session Object

Furthermore, you can utilize the Session_OnStart event to force the user into a particular
page when they enter an application. This is a very important feature when you consider the
potential workflow nightmares that browser bookmarks present to developers. Bookmarks are
dangerous to application workflow because they provide a manner to bypass required steps to
reach a certain point in the application. This is important not only for fulfilling process require-
ments, but also in terms of possibly bypassing application security. In most situations, you want
to limit who can access your ASP application as well as control the rights and permissions of
the user within the application. Using the Session_OnStart event prevents a user from enter-
ing an application via a bookmark or from entering a specific URL address. Because the
Session_OnStart event is executed every time a new user enters the application, you can force
the user to always enter a logon page to enter the application. The routing of the user to the
logon page is accomplished by using the Redirect method of the Response object. The Re-
sponse object is covered in much greater detail in Chapter 14. The following example forces a
user into a login screen regardless of where he tries to enter an application:
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Session_OnStart
dim startPage
startPage = “Login.asp”
Response.Redirect(startPage)
End Sub

Session_OnStart

</SCRIPT>

The previous example uses the Response object to force the user to the Login.asp page every
time he enters the application.

Session_OnEnd
The Session_OnEnd event, also located in the optional GLOBAL.ASA file, is triggered when the
current Session object is closed. The Session_OnEnd event requires the following syntax:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>

Sub Session_OnEnd
‘=== Closing code here
End Sub

Session_OnEnd

</SCRIPT>

where ScriptLanguage is any script-compliant language.

The Session_OnEnd event is triggered when the Session object is abandoned or times out.
During this pre-shutdown of the Session object, all interaction with the client browser is pro-
hibited. When the Session_OnEnd event is triggered, all new HTTP requests are terminated

17.1389-6 CH13 258 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Controlling User Session Resources 259

while the existing requests in the queue are processed. Any ASP code requests to the Response
or Request objects in the Session_OnEnd event will generate an error. References to all objects
internal to the Web server, such as the Server, Application, or Session objects, are still valid.
However, references to the Server.MapPath method would cause a ‘type mismatch’ error to
occur. The valid object calls for the Session_OnEnd events are illustrated in Table 13.4.

Table 13.4 Session_OnEnd Object Calls

Valid Objects Invalid Object Calls
Server Request

Session Response

Application

Controlling User Session Resources

The Session object uses ASP settings to help regulate the amount of resources consumed by a
specific user’s session. If session information is being stored, calling the Abandon method de-
stroys the session resources.

The Abandon Method

The Abandon method is used to destroy and release the resources consumed by Session ob-
jects. The Abandon method can be implicitly or explicitly called. The implicit execution of the
Abandon method is automatically called when the Session.Timeout value is exceeded. The
explicit destruction of the Session object is called using the following syntax:
Session.Abandon

where Session is the current Session object.

In most situations, the Abandon method is executed implicitly when the Session.Timeout value Part
is passed. Usually the timeout value passes after the user has finished using the application and III
has navigated to another Web application. The Web application has no way to determine
whether the user is simply inactive, is finished with the application, or has browsed to another Ch
site. 13
The Session.Abandon method has some special scoping characteristics that need to be ad-
dressed. First, when the Session.Abandon method is called, the Web server continues to pro-
cess the remaining ASP code on the page. This is unlike the Server.Timeout event and the
Response.Redirect methods, in which code execution on that page is instantly terminated.
The following code demonstrates how the ASP continues to write the welcome message to the
page after the Abandon method is called.

17.1389-6 CH13 259 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 260 Chapter 13 Managing the User Session: The Session Object

<%
Response.Write(Session.SessionID) &”<BR>” ‘===
Writes ID to page
Session(“myUserName”)=”Rubble” ‘===
Assigns value to variable
Response.Write(Session(“MyUserName”)) &”<BR>” ‘===
Write variable
Session.Abandon ‘===
Call to abandon current
session
Response.Write(“Welcome to BrainWare International”) &”<BR>” ‘=== Writes
➥message
%>

This previous example writes the SessionID and variable name Rubble to the page and then
executes the Abandon method. At this point, the processing of the script does terminate, but it
does write the string “Welcome to BrainWare International” to the page.

Keep in mind that this capability to continue processing ASP script varies from object to object.
For example, consider the Server object. If the Server.Timeout is exceeded, as shown in the
following example, the script processing is instantly terminated and scripting control is re-
leased from the current page. For more information about the Server.Timeout property, see
Chapter 12.
<%

dim iCount
iCount = 0

DO WHILE iCount < 10 ‘=== Start of infinite loop

IF iCount = 9 THEN
iCount = 1
END IF
iCount = iCount +1
LOOP

Response.Write(Server.ScriptTimeOut) ‘=== Never is executed because script

➥timesout
%>

The second Session object scoping issue to remember is that all of a Session object’s proper-
ties that were created before the Session.Abandon method call are now destroyed . The
Abandon method destroys all the properties of the Session object without actually destroying
the Session object itself. For example, if you try to reference a session variable that existed
before the Session object was abandoned, no values will be returned from the Session object:
<%
Session(“myUserName”)=”Rubble” ‘=== Sets Session variable
Response.Write(Session(“MyUserName”)) &”<BR>” ‘=== Writes variable to page
Session.Abandon ‘=== Destroys pointers to
➥properties
Response.Write(Session(“MyUserName”)) &”<BR>” ‘=== Writes a zero length string
%>

Third, because the Session object itself is not yet destroyed, the Session object can be used to
temporarily store user properties. Only after the page has completely finished processing is the
current Session object placed in the queue for destruction. For example, the following code

17.1389-6 CH13 260 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Using Session-Level Objects 261

calls the Session.Abandon method, but still uses the Session object to create another session
variable before the Session object is destroyed.
<%
Response.Write(Session.SessionID) &”<BR>” ‘=== Writes 94734 to page
Session(“myUserName”)=”Rubble” ‘=== Assigns session value
Response.Write(Session(“MyUserName”)) &”<BR>” ‘=== Writes ‘Rubble’ to page
Session.Abandon ‘=== Calls the abandon method
Session(“myUserName”)=”FlintStone” ‘=== Assigns session variable
Response.Write(Session(“MyUserName”)) &”<BR>” ‘=== Writes ‘FlintStone’ to page
Session(“myBossName”)=”Slate” ‘=== Creates new Session variable
Response.Write(Session(“myBossName”)) &”<BR>” ‘=== Writes ‘Slate’ to page
Response.Write(Session.SessionID) &”<BR>” ‘=== Writes 94734 to page, same
➥as above
%>

When the Session.Abandon method is called, the closing of the Session object triggers the
Session_OnEnd event in the GLOBAL.ASA file.

Configuring the Session Object

As previously mentioned, IIS 4.0 relies on the Metabase to store configuration settings for the
Web server, and the Microsoft Management Console (MMC) is used to configure the
Metabase, whereas IIS 3.0 uses the Registry as its primary configuration information store.
The AllowSessionState Registry value is used to indicate whether Active Server Pages should
track user state information. If AllowSessionState is set to 1, session information will be
recorded. As a result, any code in the Session_OnStart and Session_OnEnd events will be
initiated, as well as writing the SessionID to a client-side cookie. If the AllowSessionState
Registry setting is set to 0, the Session object is not created to monitor user workflow informa-
tion. Consequently, the Web server will not send cookie information to the client and any refer-
ence to a Session object, including the actual Session_OnStart and Session_OnEnd events, will
generate an error. If you are now using the Session object, turning the session information off
can improve performance as a result of eliminating the need to process additional HTTP
header requests, to allocate and de-allocate memory space, and to send and receive cookie
information. Part
III
Using Session-Level Objects Ch
The Session object can also be used to create references to other instantiated objects, which 13
are then available throughout the user’s specific session. However, because individual objects
must be created on the Web server for every Session object, the threading model of the com-
ponent must be considered to avoid locking out any users.

Instantiating Session-Level Objects

As with page- and application-level objects, objects can be instantiated at the session level.
Creating an object with session-level scope enables that object to be referred to during the
lifetime of that user session. Session-level objects give you the ability to share information

17.1389-6 CH13 261 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 262 Chapter 13 Managing the User Session: The Session Object

particular to a user from page to page within an application. The instantiated session-level
object is destroyed whenever the user Session object is destroyed. In order for reallocation of
session resources to occur, the Session.Abandon method must be called or the
Session.Timeout value must be exceeded.

You can create a reference to an automation object by embedding the unique class identifier
either by the ProgID or the ClassID within the object tag. The ProgID represents the registered
name of the object. The ClassID refers to the registered class number of the object. To create a
server-side object, use the following syntax:
<OBJECT RUNAT = Server SCOPE = Scope ID = Identifier PROGID = “progID”|CLASSID
=”ClassID”>
</OBJECT>

Instantiating Session-Level Objects

One consideration when using components is the component threading model. The component
threading model dictates how components will run on the server. A single-threaded component
prevents multiple users from initiating requests on the same component. This single-serve
approach prevents information from being mixed across multiple users. However, this presents
a large obstacle when implementing Web-centric solutions. For applications with a large num-
ber of users, a single instance component can cause significant delays while the users are
sequentially given access to the component. Keep in mind that the threading is relative to the
processor, not the specific user of application sessions. Therefore, managing your component’s
threading models will play a critical role in the performance of your site. For this reason, only
objects that use the apartment and free threading models can be stored within session vari-
ables.

N O T E Only session-level objects that are free threading and apartment threaded can be stored in
Session variables. ■

For more information on component threading issues, please see Chapter 18, “Creating Your
Own Active Server Components.”

Storing Arrays in the Session Object

One of the most common uses of Session object is to store a session array that contains infor-
mation about the user’s progress in an application. As you saw in Chapter 10, the Application
object is used to store an array of training classes that is loaded into memory when an ASP
application begins. Using the Application array is suitable when the required information
does not change often. However, in general, your applications will contain a wide variety of
dynamic information that depends on interaction with individual users.

Implementing a session array is done in the same manner as declaring a session variable. For
example, a local array called myArray is added to a session variable named mySessionArray in
the following example:
<% Session(“mySessionArray”) = MyArray %>

17.1389-6 CH13 262 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Using Session-Level Objects 263

If a session variable named mySessionArray does not exist, the session variable is created. If a
variable of that name already exists, the existing data is overwritten by the assignment of the
array.

To create a session-based array, the first step is to create a page-level array in VBScript using
the Dim keyword. The following example demonstrates creating a local array, assigns values to
the array, and then transfers the local array into a session variable.
<%
Dim aLocalTraining(4) ‘=== Creates a fixed size array
aLocalTraining (1)=”Using Active Server Pages”
aLocalTraining (2)=”Internet Basics”
aLocalTraining (3)=”Content Management”
aLocalTraining (4)=”Internet Application WorkFlow”
Session(“aTraining”)= aLocalTraining ‘=== Assigns local array in the
Session object
%>

Now the user can access this information from any other page by referring to the session vari-
able aTraining.
<%
If IsArray(Session(“aTraining”)) then ‘=== Validates array
dim aTraining, iclass
aTraining = Session(“aTraining”) ‘=== Creates local array
iClass = Ubound(aTraining) ‘=== Finds upper limit for array
For i = 1 to iClass ‘=== Loop through results
Response.Write(aTraining(i) & “<BR>”) ‘=== Write array to page
Next
Else
Response.Write (“Error in processing array”)
End If
%>

You can also utilize the Session_OnStart event in the GLOBAL.ASA file to populate a session-
level array. The same approach is used to create, fill, and transfer the local page array to the
session variable. The following example uses a session-level array to hold user information. In
this situation, information about the requesting client is stored when the user first enters the Part
application.
III
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>
Ch

Sub Session_OnStart
13
Dim myUserInfoArray (4) ‘=== Creates a fixed size array
myUserInfoArray (1) = “myScreenResolution” ‘=== Assigns text to first
➥element
myUserInfoArray (2) = “myBrowserType” ‘=== Assigns text to second
➥element
myUserInfoArray (3) = “myOperatingSystem” ‘=== Assigns text to third
➥element
Session(“myUserInfoArray “) = myUserInfoArray
End Sub

17.1389-6 CH13 263 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 264 Chapter 13 Managing the User Session: The Session Object

Sub Session_OnEnd
End Sub

</SCRIPT>

To use this information, create a local instance of an array and assign the session array to the
local array. For example, to automatically route the user to pages designed for a specific resolu-
tion, the second element of the session array myUserInfoArray is used to check the user reso-
lution set in the Session_OnStart event.
<%@ LANGUAGE=”VBSCRIPT” %>
<%
dim aUserInfo
aUserInfo = session(“myUserInfoArray”) ‘=== Create Local Array
Select Case aUserInfo(2) ‘=== Check value
Case “800x600”
Response.redirect “/VGA/Intro.asp” ‘=== Redirect to this page
Case else
Response.redirect “/LowRes/Intro.asp”
End Select
%>

Notice how this ASP script is a non-visual Web page that simply processes information. This
technique can be used throughout your applications to provide processing logic without having
to embed all your code in pages that are displayed to the browser.

This concept of using session arrays to store a collection of user information is used exten-
sively in the online catalog examples illustrated in Chapters 16, “Putting It All Together: Build-
ing an Online Catalog,” 19, “Putting It All Together: Creating an Online Registration System,”
and 26, “Putting It All Together: Creating a Dynamic Database Application.” These chapters
demonstrate using application and session arrays to display information throughout a site and
keep track of what the user has ordered.

Managing Session Variables Using Collections

The Session object also utilizes collections to help manage session-level variables. Session-
level variables can be categorized into two classes based on how the variable is declared, as
shown in Table 13.5. Variables can be declared with or without the <OBJECT> tag. If the session-
level object is declared as an object using the <OBJECT> tag, the object is managed by the
StaticObject collection. If the variable is not declared using the <OBJECT> tag, the session-
level variable is managed by the Contents collection.

Table 13.5 The Session Collection

Name Description
Contents Contains all session-level variables that have not been created with the
<OBJECT> tag.

StaticObject Contains all session-level objects declared with the <OBJECT> tag.

17.1389-6 CH13 264 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 Managing Session Variables Using Collections 265

The Contents Collection

The Contents collection contains all items that have been created during a session and that
have not been declared with the <OBJECT> tag. The Contents collection requires the following
syntax:
Session.Contents( Key )

where Key is the name of the property to retrieve. The Contents collection manages all session-
level variables declared through a user session across all pages. For example, assume that the
following session-level variables have been declared in separate ASP.
<% Session(“myAcctNum”) = “IR-404-3305-S” %> ‘=== Declared in one ASP
<% Session(“myDepVersion”) = “50-TT-50” %> ‘=== Declared in another ASP

The Contents collection can then reference these variables by using the required syntax, as
shown below:
<%
Response.Write(Session.Contents(“myAcctNum”))
Response.Write(Session.Contents(“myDepVersion”))
%>

Furthermore, you can iterate through the collection to obtain a list of all session-level variables
by using a For...Next loop:
<%
For Each Key In Session.Contents
Response.Write(Key & “<BR>”)
Next
%>

The StaticObject Collection

The StaticObject collection is similar to the Contents collection, but it is used to manage all
session-level objects that have been declared with the <OBJECT> tag. The StaticObject collec-
tion uses the following syntax:
Session.StaticObjects( Key ) Part

where Key is the property to retrieve. The StaticObject collection can also be used to refer- III
ence a single session variable or multiple session variables. For example, consider the follow- Ch
ing GLOBAL.ASA file that declares an ADO and an Ad Rotator Session object as well as loads a
session-level array: 13
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Session_OnStart
Dim aTraining(4) ‘=== Creates a fixed size array
aTraining(1)=”Using Active Server Pages”
aTraining(2)=”Internet Basics”
aTraining(3)=”Content Management”
aTraining(4)=”Internet Application WorkFlow”
Session(“aTraining”)=aTraining
End Sub

17.1389-6 CH13 265 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 266 Chapter 13 Managing the User Session: The Session Object

Session_OnStart
</SCRIPT>
<OBJECT RUNAT = Server SCOPE = Session ID = MyConnection PROGID =
➥“ADODB.Connection”>
</OBJECT>
<OBJECT RUNAT = Server SCOPE = Session ID = MyAd PROGID = “MSWC.AdRotator”>
</OBJECT>

To list all items in the StaticObject collection, use the For...Next loop to step through the
<OBJECT> collection:

<%
For Each Key In Session.StaticObjects
Response.Write(Key & “<BR>”)
Next
%>

This iterative process will write the names of the Session objects, shown below, to the page:
MyAd
MyConnection

From Here…
In this chapter, the versatility of the Session object is demonstrated to track user-specific
information for cookie-based browsers. Your ASP applications can still be written if cookies are
not supported on the browser’s clients, however, the Session object will be rendered useless.
However, in situations where cookies are supported, the Session object can be used to store a
variety of user-specific information using variables, arrays, and even ActiveX components. The
Session object uses the Session.Timeout and Session.Abandon properties and methods to
help manage resources on the server.

For related information, please see the following chapters:

■ Chapter 15, “Retrieving User Information: The Request Object,” depicts how to gather
information using forms, hidden fields, and cookies.
■ Chapter 17, “Implementing Existing Server-Side Components,” demonstrates how to use
the various components that ship with IIS Active Server Pages.
■ Chapter 18, “Creating Your Own Active Server Components,” walks you through
the tools and examples on how to create tailored functionality via custom-created
components.

17.1389-6 CH13 266 10/26/98, 9:23 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/12/97 CH13 lp#4
 14
267

C H A P T E R

Utilizing the Response Cookies

Speaking to the Web Collection
Discover how to create client-side

Client: The Response cookies to store user or application

information, control the lifetime of
the cached cookie, and manipulate

Collection data within the cookie.

Using the Response Object’s

Properties
Learn how to control the data prop-
erties of information delivered to
the browser by regulating how
information is sent to the browser,
Controlling how information is distributed from the Web controlling the lifetime of the page
server to the requesting browser forms half of the founda- on the client’s browser, and other
tion needed to deliver Active Server applications. The page management essentials.
Active Server uses a built-in object called the Response
Controlling the Response
object to control and manage the data sent to the browser.
Object Using Methods
The Response object is responsible for controlling the
delivery of data, writing HTTP header information, writing Tap into the power of the Response
object by learning how to control
text, HTML, scripting variables, and non-textual informa-
the information flow from the server
tion, and controlling cookies on the client browser.
to the browser by using buffering,
The Response object is separated into three sections that logging, and Web page redirection
are used to properly display information between the capabilities.
client and server. The Cookies collection provides a valu-
able resource in administering and controlling client-side
cookies. The Cookies collection enables you to create,
control, and manipulate information stored in cookies.
Part
The Response object’s properties enable you to set many
of the physical properties of pages, such as the lifetime of III
the page in the client’s cache. The Response object’s meth- Ch
ods are used to explicitly control the flow of information
from the server to the browser, and enable you to buffer 14
pages on the server, write textual, non-textual, and HTML
information to pages, and write information to the Web
server log files. ■

18.1389-6 CH14 267 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 268 Chapter 14 Speaking to the Web Client: The Response Collection

The Response Collection

The Response object, in addition to the other built-in Active Server Page objects, uses collec-
tions to help group common indexes. A collection represents a logical storage unit that helps
manage information. The Response object only uses one collection, the Cookies collection. The
Cookies collection manages and controls both cookie files and the data stored within the cookies.
This temporary storage location is used in a variety of ways in Web application development.
One of the most common uses of cookies is to store the last visited Web page in an application
or site and the last time and date a particular page or Web site was accessed. This same con-
cept of tracking user activity throughout a site can be applied to the application workflow pro-
cess. Cookies provided a mechanism to track a user’s location within the application.

The Cookies collection enables single or multiple variables to be stored and manipulated in
temporary text files on the client’s browser. To write variables to cookies, use the Response
object in conjunction with the Cookies collection. The Cookies collection requires the following
syntax:
Response.Cookies(Cookie)[(key)|.attribute] = value

where Response is the built-in Response object, Cookie is the name of the cookie file, key identi-
fies a dictionary element, attribute is a specific characteristic of the cookie, and value is the
value being assigned to the cookie.

To assign a value to a cookie, use the Cookies collection syntax as described above. For ex-
ample, to create a cookie named myLowFatCookie and assign it a value of A banana, use the
following syntax:
<% Response.Cookies(“myLowFatCookie”) = “A Banana” %>

Notice that you do not have to manually trap and replace non-ANSI characters when extracting
or writing the information to the cookie. The encoding process is built into the Response and
Request Cookies collections. As a result, the encoding process automatically filters out any
non-ANSCI characters. In the previous example, the HTTP header string that is sent from the
Web server converts the string A Banana to A+Banana. For more information about URL and
HTML encoding methods, see Chapter 12, “Controlling the Server with the Server Object and
Using Transaction Processing with the ObjectContext Object.”

Keep in mind that you cannot write information to a cookie file after HTTP header information
has been sent to the requesting browser. In other words, you cannot send cookie information
after to the browser after any HTML tags are sent to the browser without generating an error.
For example, to successfully write the cookie information to a browser, you must follow the
code syntax like this:
<%@ LANGUAGE=”VBSCRIPT” %>
<% Response.Cookies(“myLowFatCookie”) = “A Banana” %>
<HTML>
<HEAD>
<TITLE>Write a simple cookie</TITLE>
</HEAD>
<BODY>

18.1389-6 CH14 268 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 The Response Collection 269

A cookie was just written to your system.

</BODY>
</HTML>

However, if you try to write data to the cookie after the Web server has delivered any HTML,
a Header error will occur. The following script indicates the improper placement of the
Response.Cookies placement in the ASP script:

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>Write a simple cookie</TITLE>
</HEAD>
<BODY>
<% Response.Cookies(“myLowFatCookie”) = “A Banana” %>
A cookie was just written to your system.
</BODY>
</HTML>

This improper placement of the Cookies collection will generate the following error:
Response object error ‘ASP 0156’
Header Error
/DemoCookie/writecookievalues.asp, line 7
The HTTP headers are already written to the client browser. Any HTTP header
modifications must be made before writing page content.

To correct this error, place the Response.Cookies attribute before the <HTML> tag.

Writing information to cookies is worthless if the information is never extracted again. To

extract the data from the cookie, the Request object’s Cookie collection is used. For example,
to retrieve the data from the newly written myLowFatCookie, use the following code:
<% Response.Write(Request.Cookies(“myLowFatCookie”)) %>

In this situation, the text A Banana is extracted from the myLowFatCookie cookie file. To find
out more about the details of retrieving information from the browser, see Chapter 15, “Retriev-
ing User Information: The Request Object.”

Notice that the cookie acts as a single storage unit for information. If you write additional infor-
mation to the cookie, all previously held information in the cookie is overwritten. For example,
if you assign the value An Apple to myLowFatCookie, the previously held value of A Banana is
overwritten.
<% Response.Cookies(“myLowFatCookie”) = “An Apple” ‘=== Over writes
➥previous value %>
Part
In most situations, if your application implements cookies, you will want to use this temporary
storage location to house multiple values. These multiple values can be accessed using keys.
III
Keys enable the conversion of a single unit storage location into smaller, identifiable units Ch
within an object. This key approach is similar to the concept of using indexes to identify sub- 14
units within an array.

18.1389-6 CH14 269 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 270 Chapter 14 Speaking to the Web Client: The Response Collection

To store multiple variables within the same cookie, use the key attribute of the Cookies collec-
tion. The following example creates three keys: Breakfast, Lunch, and Dinner, and assigns the
corresponding Banana, Apple, and Orange values.
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Banana”
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “Apple”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “Orange”
%>

Remember: If the cookie is assigned a new value, the key values are destroyed. If you re-assign
the Dinner key to Orange, as shown in the following code example, the previously assigned key
values, Breakfast, Lunch, and Dinner, would be destroyed.
<% Response.Cookies(“myLowFatCookie”)(“Dinner”) = “Orange” %>

The Cookies collection also uses attributes to help manage cookies on the browser. Table 14.1
displays the cookie attributes that can be set by the Response collection.

Table 14.1 The Attributes of the Response Cookies Collection

Name Description
Expires Sets the date when the cookie will expire.
Domain Specifies cookie delivery to only members specified by this domain.
Path Determines the delivery path information.
Secure Specifies whether the cookie is secure.
HasKeys Returns whether the cookies contain multiple values.

The Expires Attribute

To create a cookie that exists until a specific date, use the Expires attribute. The following
code demonstrates setting the lifetime of a cookie to exist until December 30, 1999.
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Banana”
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “StrawBerry”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “Orange”
Response.Cookies(“myLowFatCookie”).Expires = “December 30, 1999”
%>

One of the easiest ways to verify the cookie expiration date is to turn on the cookie warning
within your browser. After this warning is activated, the browser will display the contents and
expiration date of the cookie trying to be sent to your browser. You can explicitly control
whether or not to accept the cookie. If you choose to accept the cookie, that cookie, the data
within it, and the cookie attributes are written to your browser. If you do not choose to accept
the cookie, no information is written to the browser. In Figure 14.1, the contents of the cookie
and the expiration date of December 30, 1999 are displayed as a result of activating the Cookie
warning indicator in Internet Explorer.

18.1389-6 CH14 270 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 The Response Collection 271

FIG. 14.1
Displaying the cookie
data and expiration
date using the browser
Cookie warning flag in
IE 4.0.

The ability of the user to reject cookies demonstrates another shortcoming with relying on
cookies for Web application development. The ability of the user to reject information that your
application may deem necessary presents a drawback in controlling and monitoring user state
information.

When using the attributes of the Response Cookies collection, make the effort to ensure that
valid data is being used. For example, if an incorrect date were entered for the cookie expira-
tion date, a Response runtime error would occur. For example, if an expiration date of Decem-
ber 32, 1999 were used for myLowFatCookie, the Response object would generate a runtime
error:
<% Response.Cookies(“myLowFatCookie”).Expires = “December 32, 1999” ‘=== Run
➥time error %>

If no expiration date is assigned to the cookie, the cookie will expire at the end of the user’s
session. For more information about using the Session object, please see Chapter 13, “Manag-
ing the User Session: The Session Object.”

The Domain Attribute

The Response object also enables you to limit the distribution of cookie files to specific do-
mains. The Domain attribute of the Cookie collection is used to identify what domains cookies
will be issued to. For example, to only distribute cookies to the domain MyCompany.Com,
use the following code:
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Another Banana”
Part
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “One StrawBerry”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “A Orange”
III
Response.Cookies(“myLowFatCookie”).Expires = “December 30, 1999” Ch
Response.Cookies(“myLowFatCookie”).Domain = “MyCompany.Com”
%> 14

18.1389-6 CH14 271 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 272 Chapter 14 Speaking to the Web Client: The Response Collection

The Path Attribute

The Path attribute extends the Domain attribute by specifying a path location in the domain that
cookies will be sent to. For example, instead of issuing cookies to all requests in the MyCompany
domain, the requests can be limited to requests from the Diet virtual directory:
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Another Banana”
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “One StrawBerry”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “A Orange”
Response.Cookies(“myLowFatCookie”).Expires = “December 30, 1999”
Response.Cookies(“myLowFatCookie”).Domain = “MyCompany.Com”
Response.Cookies(“myLowFatCookie”).Path = “/Diet”
%>

The Secure Attribute

The Secure attribute of the Cookies collection is used to set whether or not the cookie is deliv-
ered over a secure channel. The Secure attribute accepts a Boolean value of True if False. For
example, to deliver our myLowFatCookie over a secure channel, set the Secure attribute to True,
as shown below:
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Another Banana”
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “One StrawBerry”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “A Orange”
Response.Cookies(“myLowFatCookie”).Expires = “December 30, 1999”
Response.Cookies(“myLowFatCookie”).Domain = “MyCompany.Com”
Response.Cookies(“myLowFatCookie”).Path = “/Diet”
Response.Cookies(“myLowFatCookie”).Secure = True
%>

If a secure channel is not detected in the HTTP header, the cookie information will not be sent.

The HasKeys Attribute

The HasKeys attribute is used to determine whether multiple values are stored within the
cookie. A key is the term used to identify the variable name within a collection. One of the best
ways to prevent keys from being overwritten within cookies is to use the HasKeys attribute of
the Cookies collection. The HasKeys attribute can be used as an indicator to determine whether
a cookie has keys:
<%= Response.Cookies(“myLowFatCookie”).HasKeys %>

Remember that you can utilize the looping functionality found in the For...Each...Next loop
to search for existing key values from a cookie or reset all keys to a value. The following ex-
ample determines whether the cookie contains multiple values and the reads from the cookie
and lists all the stored values:
<%
For Each key in Request.Cookies(“myLowFatCookie”)
Response.Write Request.Cookies(“myLowFatCookie”)(key) &”<BR>”
Next
%>

18.1389-6 CH14 272 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using the Response Object’s Properties 273

Notice that to run the previous example, you can also utilize the For...Each...Next to resets
all the keys in the cookie to the string “Empty Value”.
<%
For Each key in Response.Cookies(“myLowFatCookie”)
Response.Cookies(“myLowFatCookie”)(key) = “Empty Value”
Next
%>

Again, the cookie warning indicator in IE3.0 displays that the key values have been all updated
to “Empty Value” in the requested cookie, as shown in Figure 14.2.

FIG. 14.2
Use keys to reset values
in cookies to prevent
destroying other cookie
keys.

Notice that the word key in the For Each...Next is not a keyword but is used to represent an
element or variable in a valid object collection. The previous example can be re-written using a
non-descriptive variable name to accomplish the same looping functionality.
<%
For Each FatCell in Request.Cookies(“myLowFatCookie”)
Response.Write Request.Cookies(“myLowFatCookie”)( FatCell) &”<BR>”
Next
%>

Using the Response Object’s Properties

The Response object has various properties that are used to set the characteristics of informa-
tion delivered by the server. These characteristics are used between the browser and the
server to provide a communication mechanism between the information provider and informa-
tion consumer. Using the properties of the Response object controls this “hand-shake” capabil-
ity between the browser’s client and Web server. Table 14.2 illustrates the Response object’s
properties.

Part
Table 14.2 The Response Properties III
Property Name Description
Ch
Buffer Sets whether or not page output is buffered.
14
CacheControl Enables dynamic output of ASP to be cached by proxy servers.

continues

18.1389-6 CH14 273 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 274 Chapter 14 Speaking to the Web Client: The Response Collection

Table 14.2 Continued

Property Name Description
CharSet Appends the name of the character set to the content-type header.
Expires Sets the amount of time in which the page cached on the browser
expires.
ExpiresAbsolute Sets the date and time in which a page cached on a browser
expires.
IsClientConnected Determines whether the browser has disconnected from the
server.
Status Indicates the status line returned by the server.
PICS Adds the value of a PICS label to the pics-label field of the
response header.

The Buffer Property

The Buffer property is used to control when information is sent to the requesting browser.
The Web server can either stream information to the user as the server is processing the
script, or it can wait to release all the data after the entire script is finished processing. The
Buffering property enables you to process information on the server before sending any
HTTP Header information to the browser. To utilize this preprocessing approach, the Response
buffering requires the following syntax:
Response.Buffer [= flag]

where Response is the Response object and flag can be set to either True or False (see
Table 14.3).

Table 14.3 The Buffering Flags

Flag Description
True Server buffering is cached until the entire ASP page has completed process-
ing or the Flush or End methods have been called.
False Server buffering is disabled, enabling output to be streamed to the browser as
it is processed.

The Buffer property is used to override the BufferingOn Registry setting at the ASP level. For
example, to ensure that buffering is disabled on an ASP even if server buffering is enabled in
the Registry, the following ASP code can be used:
<% Response.Buffer = False %>

18.1389-6 CH14 274 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using The Response Object’s Properties 275

If the buffering flag is set to True, the Web server output is cached until all the ASP code has
finished processing or the Flush or End Response methods have been called. If buffering is set
to False, server buffering is disabled and information is sent to the browser as the server
processes it.

Enabling buffering on the Web server increases server performance because the server does
not have to continuously create and release HTTP connection resources for every item deliv-
ered by the ASP. Although this streamlining approach can enhance server performance, care
must be taken to not alienate the user by not displaying any output. The user may assume that
the Web site is down or is slow because no information is being presented to the browser.

Keep in mind that the Buffering property only refers to controlling the release point where
information is sent to the requesting browser and does not influence the Session object or any
HTTP Keep-Alive request. Also, enabling buffering presents a presentation delay to the users.
Consequently, your application or network may appear to be slow or congested.

Buffer activating code should be placed within the first lines of code on the page. Because
buffering cannot be set after output is sent to the browser, the Buffer tag must be placed be-
fore any text is streamed to the client. Trying to change the buffering characteristics after data
has been sent to the client will produce a runtime error. For example, the following code will
generate a runtime error because information is already being streamed to the client.
<%@ LANGUAGE=”VBSCRIPT” %>
<% Session(“UserName”) = “Harvey West” %>
<HTML>
<HEAD>
<TITLE>Response Error</TITLE>
</HEAD>
<BODY>

Hello <% = Session(“UserName”) %>

<% Response.Buffer = True ‘=== Produces runtime error %>
</BODY>
</HTML>

As a result of the incorrect placement of the Buffer property, the following error is produced:
Response object error ‘ASP 0156’
Header Error
/Demo/bufferingerror.asp, line 10
The HTTP headers are already written to the client browser. Any HTTP header
modifications must be made before writing page content

To correct this code, place the buffering tag after the LANGUAGE directive: Part
<%@ LANGUAGE=”VBSCRIPT” %>
<% Session(“UserName”)=”Harvey West” %>
III
<HTML> Ch
<HEAD>
<TITLE>Response Error</TITLE> 14
</HEAD>
<BODY>

18.1389-6 CH14 275 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 276 Chapter 14 Speaking to the Web Client: The Response Collection

Hello <% = Session(“UserName”) %>

<% Response.Buffer = True ‘=== Produces runtime error %>
</BODY>
</HTML>

The buffering property is very useful in situations where information must be processed be-
fore sending any HTTP header information to the requesting browser. This is particularly
useful considering that the HTTP header information can only be sent to the requesting
browser once. As you have seen with different ASP objects, such as the Cookies.HasKeys
property and the Response.Redirect method, trying to re-send any HTTP header information
after the initial HTTP header information has already been sent will generate a runtime error.
The Response.Buffer property is useful to prevent these errors. For example, consider the
previous Cookie collection example where all the values are stored in a cookie. To determine
whether the cookie contained multiple values, the HasKeys property must be used before send-
ing any HTML to the browser to prevent an error. To accomplish this in one seamless step,
enable buffering to allow the HasKeys property to be utilized after HTML is called in the ASP,
as shown below:
<%@ LANGUAGE=”VBSCRIPT” %>
<% Response.Buffer = True %>
<HTML>
<HEAD>
<TITLE>Read Cookies information and Buffer Output</TITLE>
</HEAD>
<BODY>
<%
If Response.Cookies(“myLowFatCookie”).HasKeys Then ‘=== Would have
➥normally generated an error
For Each key in Request.Cookies(“myLowFatCookie”)
Response.Write Request.Cookies(“myLowFatCookie”)(key) &”<BR>”
Next
End If
%>
</BODY>
</HTML>

The CacheControl Property

The Response object uses the CacheControl property to enable a proxy server to cache output
from Active Server Pages. The CacheControl requires the following syntax:
Response.CacheControl [= Cache Control Header ]

where Cache Control Header can be True or False. If the Cache Control Header is set to
False, the Cache Control header is considered private, meaning the results of the ASP page
will pass through a proxy server without caching the individual users request. To enable cach-
ing of ASPs on the proxy server, set the Cache Control Header to True. Now the proxy server
will cache the ASP output and consider the results as public to all users requesting that page:
Response.CacheControl = True

18.1389-6 CH14 276 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using The Response Object’s Properties 277

The CharSet Property

The ContentType, another property of the Response object, is used to regulate how a request-
ing browser displays the textual information that is deliverer from the Web server. The
CharSet property of the Response object is used to set the character set of the ContentType
that is displayed on the requesting browser. The CharSet property controls the ContentType
header information by using the following syntax:
Response.Charset(CharSetName)

where CharSetName is the character set designated for a particular page. For example, to
change the character set for a page to ISO-LATIN-1, use the following example:
<% Response.Charset(“ISO-LATIN-1”) %>

The content-header delivered from the Web server would be:

content-type:text/html; charset = ISO-LATIN-1

Notice that, if the Response.Charset property is not specified, the CharSet label is not added to
the content-header string. Furthermore, the CharSet property does not check to validate if the
CharSet name is a valid standard.

Setting the CharSet value is often done using the META HTML tags. For example, consider the
default META tag that uses the International Organization for Standards (ISO) 8859-1 standard:
<META HTTP-EQUIV = “Content-Type” content = “text/html; charset=iso-8859-1”>

If an ASP were written in Japanese, you would use the following CharSet:
<META HTTP-EQUIV = “Content-Type” CONTENT = “text/html; charset=shift_jis”>

However, now you can set the character set using the ASP Response.CharSet property. Table
14.4 is a list of valid character set tags and associated code page values.

Table 14.4 Valid Character Set Values and CodePage Values

CharSet Tag CodePage
us-ascii 1252

iso8859-1 1252

ascii 1252

iso_8859-1 1252
Part
iso-8859-1 1252
III
ANSI_X3.4-1968 1252
Ch
iso-ir-6 1252
14
ANSI_X3.4-1986 1252

continues

18.1389-6 CH14 277 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 278 Chapter 14 Speaking to the Web Client: The Response Collection

Table 14.4 Continued

CharSet Tag CodePage
ISO_646.irv:1991 1252

ISO646-US 1252

Us 1252

IBM367 1252

cp367 1252

csASCII 1252

latin1 1252

iso_8859-1:1987 1252

iso-ir-100 1252

ibm819 1252

cp819 1252

Windows-1252 1252

iso8859-2 28592

iso-8859-2 28592

iso_8859-2 28592

latin2 28592

iso_8859-2:1987 28592

iso-ir-101 28592

l2 28592

csISOLatin2 28592

windows-1250 1250

x-cp1250 1250

windows-1251 1251

x-cp1251 1251

windows-1253 1253

windows-1254 1254

shift_jis 932

shift-jis 932

x-sjis 932

18.1389-6 CH14 278 10/26/98, 9:24 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using The Response Object’s Properties 279

CharSet Tag CodePage

ms_Kanji 932

csShiftJIS 932

Extended_UNIX_Code_Packed_Format_for_Japanese 932

csEUCPkdFmtJapanese 932

x-euc-jp 932

x-euc 932

csISO2022JP 932

iso-2022-jp 932

windows-1257 1257

big5 950

csbig5 950

x-x-big5 950

GB_2312-80 936

iso-ir-58 936

chinese 936

csISO58GB231280 936

csGB2312 936

gb2312 936

csKOI8R 20866

koi8-r 20866

ks_c_5601 949

ks_c_5601-1987 949

korean 949

csKSC56011987 949

euc-kr 949
Part
ISO_8859-8:1988 1255

iso-ir-138 1255
III
Ch
ISO_8859-8 1255

ISO-8859-8 1255
14
continues

18.1389-6 CH14 279 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 280 Chapter 14 Speaking to the Web Client: The Response Collection

Table 14.4 Continued

CharSet Tag CodePage
Hebrew 1255

CsISOLatinHebrew 1255

Windows-1256 1256

Windows-874 874

Windows-1258 1258

The ContentType Property

The ContentType property enables you to specifically control the HTTP content-type that is
sent to the browser. The content-type is used by the browser to interpret the information sent
from the server. The browser uses the content-type information to determine how it should
interpret the information, such as treat the information as text, HTML, or as an image file. The
ContentType functionality is implemented using the following syntax:

Response.ContentType [= ContentTypeID ]

where ContextTypeID is the browser-specific MIME content-type.

The content-type is formatted into two sections, a general content category and a sub-content
category, as seen in the following syntax:

“General Category/Sub Category”

For example, the following list displays some common content-types:

text/HTML
image/GIF
image/JPEG

Content-types are stored within the browser and usually are associated with file extensions and
helper applications. Helper applications are programs that are automatically started when the
file type is sent to the browser.

Notice that the browser controls the content-types on the client machine. If the client has an
associated content-type configured on the client and you specify a specific content-type, the
setting on the browser will be interpreted by the client. Remember that you can specific the
content-type of an item, but the client’s browser is ultimately in control of how the information
is interpreted and displayed.

The Expires Property

The Response.Expires property is an interesting feature that enables you to specify the
amount of time in minutes before the page is expired from the browser cache. The Expire
property uses the following syntax:

18.1389-6 CH14 280 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using The Response Object’s Properties 281

Response.Expires [= number]

where number is the number of minutes the page will remain active in the cache.

After this time expires, the browser will be forced to retrieve information from the hosting site.
If the client requests the same page before the expiration timeout value, the page will be re-
trieved from the client cache. For example, to force new requests for a page from the Web
server, set the Expire property to 0, as seen in the following example:
<% Response.Expires = 0 %>

However, if you are designing your site for proper scalability and load considerations, this
constant requesting to the Web server might not be optimal. Not all situations require constant
data retrieval mechanisms back to the Web server. For example, consider sale items in an
online catalog. This information would contain product descriptions, images of the products,
and sale price information. If this was mostly static information displayed on commonly refer-
enced pages, multiple requests to the Web server would be redundant and a waste of process-
ing power and bandwidth.

Remember that the Response.Expires property controls the expiration time for a delivered
page in the browser and the previously discussed Cookies.Expires attribute controls the life-
time of a cookie file. However, although you can programmatically control the page settings,
you should not rely on this information to be present in the cache. The cache settings in the
browser override the page programmatic settings. For example, if the client’s browser does not
cache Web pages, setting an expiration time for an ASP is useless.

The ExpiresAbsolute Property

The ExpiresAbsolute is an expansion of the Expire property. The ExpiresAbsolute property
enables you to specify the exact time and date a page will expire in the browser’s cache. The
ExpiresAbsolute property requires the following syntax:

Response.ExpiresAbsolute [= [date] [time]]

where date is the date on which the page will expire, and time indicates the time the page will
expire. However, notice the expiration date and time is converted to Greenwich Mean Time.

To enable the page to expire on December 31, 1999 at 11:59:59 PM GMT, the following syntax
should be used:
<% Response.ExpiresAbsolute = #Dec 31,1999 23:55:59 # %>

If you try to set the page ExpiresAbsolute property to a time in the past, the current date and
Part
time stamp will be used as the expiration date and time values for the page.
III
The PICS Property Ch
The Recreational Software Advisory Council on the Internet (RSACi) is a nonprofit organiza- 14
tion that provides an advisory council and rating system for software games and Web sites.
This rating system sets a standard for monitoring levels of sex, nudity, violence, and offensive
language. RSACi works with the Platform for Internet Content Selection (PICS) to provide a
standard rating system to indicate labeling information for Internet-based content.

18.1389-6 CH14 281 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 282 Chapter 14 Speaking to the Web Client: The Response Collection

ON THE WEB
For more information on these rating systems, visit the RSACi Web site at www.rsac.org or the PICS
site at www.w3.org.

The PICS property of the Response object enables you to control rating values to the PICS-label
field in the response header. This functionality can be controlled by using the following syntax:
Response.PICS(Picslabel)

where PicsLabel is the formatted PICS label.

To implement the PICS Response property, the following example adds the required PICS
rating labels to the response header field:
<% Response.PICS(“(PICS-1.1 <http://www.rsac.org/ratingv01.html> labels on “ &
➥chr(34) & “1997.01.05T08:15-0500” & chr(34) & “ until” & chr(34) &
➥“1999.12.31T23:59-0000” & chr(34) & “ ratings (v 0 s 0 l 0 n 0))”) %>

The delivered header information is sent to the requesting client to enable client-side rating
systems to help control content. If content advising has been enabled on the client’s browser,
the browser will determine whether the delivered ratings are acceptable according to the rat-
ing guidelines set on the client’s browser. Figure 14.3 illustrates setting the ratings on the
Internet Explorer 4.0 browser.

FIG. 14.3
You can set ratings for
your site with this dialog
box.

The previous Response.PICS code example would deliver the following HTML to the client’s
rating systems.
pics-label:(PICS-1.1 <http://www.rsac.org/ratingv01.html> labels on
➥“1997.01.05T08:15-0500” until “1999.12.31T23:59-0000” ratings (v 0 s 0 l 0 n 0))

18.1389-6 CH14 282 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Using the Response Object’s Properties 283

The Status Property

The Response object uses the Status property to control the status line returned by the Web
server. The status line is used to determine the results of the Web server request and is speci-
fied by HTTP specifications. To implement this functionality, use the following syntax:
Response.Status = StatusDescription

where StatusDescription is the status code and status code description. The following is a list
of common HTTP status lines that can be returned by the Web server.

■ 400 Bad Request

■ 410 Unauthorized - Login Failed
■ 404 Not Found
■ 406 Not Acceptable
■ 412 Precondition Failed
■ 414 Request-URL Too Long
■ 500 Internal Server Error
■ 501 Not Implemented
■ 502 Bad Gateway

One of the most common uses of the Status property is to force authenticate the requesting
user. For example, to force user validation for a page, use the following code to prompt the user
for a username and password:
<% Response.Status = “401 Unauthorized” %>

When this page is requested, the username and password window is displayed to solicit login
information from the user, as shown in Figure 14.4.

FIG. 14.4
Use the Status
property to authenti-
cate user access to
pages.

The username and password can then be examined to determine whether access is available to
the requested pages. However, keep in mind that the username and password are being sent to Part
the server in clear text. This unsecure communication mechanism provides the potential for III
malicious individuals to obtain critical information by filtering or examining the IP packets. To
Ch
guard against this potential security pitfall, only transmit critical information across a Secure
Socket Layer (SSL) connection. 14

18.1389-6 CH14 283 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 284 Chapter 14 Speaking to the Web Client: The Response Collection

The IsClientConnected Property

The IsClientConnected property is used to expand the capability of the Session object to
track the user. The IsClientConnected tracks whether the requesting browser has discon-
nected from the server since the last time the server issued a Response.Write command. The
IsClientConnected property requires the following syntax:

Response.IsClientConnected()

The IsClientConnected property provides an extension of the Session.Timeout capability by

enabling the requesting browser to exceed the session timeout variable without losing server
connection, and is often implemented before processing script. The following example tests the
Response object to see whether the connection is valid before calling a function to transfer data.

If Response.IsClientConnected Then
vTransfer = TranferInfo(“AcctRetail”, “AcctWholeSale”)
Else
VTransfer = ReConnect(“UserName”, “Department”)
End If

If the user is still connected to the Web server, a script function called TransferInfo begins
transferring information from the “AcctRetail” department to the “AcctWholeSale” depart-
ment. If the connection is not valid, the user tries to reestablish the connection.

Sending Output to the Browser: The Response

Methods
The Response methods are used to explicitly control information flow from the server to the
browser. The Response methods are used to control not only the textual information that is
displayed in textual or HTML format, but also HTTP header information, non-textual content
such as binary images, and Web server log files. In addition, the Response methods control
server buffering and page redirection issues. Table 14.5 lists the available Response methods.

Table 14.5 The Response Methods

Method Name Description
Write Writes a string to the current HTTP output.
BinaryWrite Writes information to the current HTTP output without any character
conversion.
Clear Erases any buffered HTML output.
End Stops the Web server from processing the script and returns the
current result.
Flush Bypasses buffering and sends output immediately to client.
Redirect Attempts to automatically route the browser to a URL.

18.1389-6 CH14 284 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Sending Output to the Browser: The Response Methods 285

Method Name Description

AddHeader Writes a string to the HTML header.
AppendToLog Writes a string to the end of the Web server log entry.

The Write Method

The Write method of the Response object is used to send output to the browser. The Write
method requires the following syntax:
Response.Write variant

where variant is any variant datatype supported by the Visual Basic scripting engine. For
example, to display the time a page was requested by using the Write method, use the follow-
ing example:
<% Response.Write now %>

The Write method is then used to write the value to the requesting page:
12/23/97 11:48:38 PM

Textual content can also be added to the Write method by constructing an argument using
concatenation strings. For example, the following example adds a descriptive message to the
VBScript Now function.
<% Response.Write “Hello, at the tone, the time will be: “ &now %>

This code example produces the following output on any requesting browser:
Hello, at the tone, the time will be: 1/13/98 10:38:47 PM

The Write method can also use arguments to hold information in the ASP, instead of directly
embedding functions in the Write method. The parentheses are optional, but help to provide a
clear logical grouping for the input arguments, as shown in Listing 14.1.

Listing 14.1 Using the Write method to display the time.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>The Time</TITLE>
</HEAD>
Part
<BODY>
III
<%
dim myTimeStamp, myMsgStr Ch
myTimeStamp = Now
myMsgStr = “Hello, at the tone, the time will be: “ 14
myMsgStr = myMsgStr & myTimeStamp

continues

18.1389-6 CH14 285 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 286 Chapter 14 Speaking to the Web Client: The Response Collection

Listing 14.1 Continued

Response.Write(myMsgStr) ‘=== Notice the user of parenthesis
%>

</BODY>
</HTML>

This example creates a variant myTimeStamp and assigns the system time and date stamp to it.
Then the Write method is used to write the same information to the requesting page as in the
previous example. For more information about the concatenation operators in VBScript, please
refer to Chapter 4, “Working with Operators.”

Just like any other server-side script, you can integrate HTML and scripting tags together. For
example, to expand on the previous example, the HTML text, “Your virtual receipt, printed on: ”
can be added to the ASP:
<%
dim myTimeStamp
myTimeStamp = Now
%>
Your virtual receipt, printed on: <% Response.Write(myTimeStamp) %>

The code produces the following output:

Your virtual receipt, printed on: 11/23/97 11:48:38 PM

Finally, you can directly embed HTML tags within the Write method. The same concatenation
technique is used to add HTML tags to the variant string. For example, to center the previous
time message on the browser page, add the HTML tags as shown in Listing 14.2.

Listing 14.2 Printing a virtual time stamp.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>The Time</TITLE>
</HEAD>
<BODY>

<%
dim myTimeStamp, myMsgStr
myTimeStamp = Now
myMsgStr = “Your virtual receipt, printed on: “
myMsgStr = myMsgStr & myTimeStamp
myMsgStr = “<CENTER>” & myMsgStr & ”</CENTER>”
Response.Write(myMsgStr)
%>
</BODY>
</HTML>

18.1389-6 CH14 286 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Sending Output to the Browser: The Response Methods 287

This example displays a message centered on the HTML page. Furthermore, you are not lim-
ited to passing the HTML tags embedded in the argument. Just as the argument was con-
structed within the Write method, as previously shown with text and the Now function, the
same approach can be applied to HTML tag:
<%
dim myMsgStr
myMsgStr = myMsgStr & Now
myMsgStr = “<CENTER>” & myMsgStr & ”</CENTER>”
Response.Write(myMsgStr & “<BR>”)
%>

This example adds an HTML break to the centered message by concatenating the HTML
break tag <BR> directly into the Write method. The break tag creates a hard return on the
page.

As seen throughout this book, two different syntaxes are used to write information to a page:
the <% and %> scripting tags and now the Response.Write method. The Response.Write
method is actually what is called when you implement the shorthand notation to write to an
ASP page using the <% and %> delimiters. But remember, the shorthand notation of the
Response.Write method is intended only to be used for inline script processing, not within
blocks of code. Inline processing occurs when HTML text is integrated with ASP code, as seen
in the following example:
“Hello, at the tone, the time will be: “ <% = now %> <BR>

Furthermore, there are some unique behaviors that are associated with using the two display
methods. The first peculiar behavior of the <% %> tag is that this feature cannot be used within
procedures, that is, subroutines and functions, in your ASP. To write to a page from within a
procedure, the Response.Write method must be used.
<%
Sub WriteData(myArgument)
Response.Write(myArgument) ‘=== Write the argument to page
= myArgument ‘=== Generates runtime error
End Sub
%>

The runtime error is produced because VBScript is trying to associate a value to a variable
when it sees the equal sign. If you really must use the shorthand Write delimiter within a pro-
cedure, you can accomplish this by splitting the procedure with script delimiters:
<%
Sub WriteData(myArgument)
Part
Response.Write(myArgument &”<BR>”)
%> III
<% = myArgument %>
<% Ch
Response.Write(myArgument &”<BR>”)
End Sub 14
%>

18.1389-6 CH14 287 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 288 Chapter 14 Speaking to the Web Client: The Response Collection

However, you can see that this is not encouraged because the additional delimiters break up
the logical grouping of the procedure and make the procedure hard to follow and read.

The second interesting point about displaying information occurs with the Response.Write
method. Because the ASP scripting engine uses the <% and %> delimiters to indicate the begin-
ning and ending of script tags, care must be taken not to cause the scripting engine to close a
script tag as it is processing the script. This premature termination of the script engine often
occurs in two situations. The first situation is when trying to write HTML tags that use the
percent (%) sign. For example, if you are trying to display the actual HTML tags needed to
create a table using the Write method, as seen in the following, a runtime error would result:
<% Response.Write “<TABLE Border = 1 WIDTH = 100%>” %>

In this situation, the Active Server scripting engine interprets the %> tags needed to close the
Table Width tag as the ending of a script tag and stops processing the tag. The script would
start processing and displaying a table, but would produce the following error:
Microsoft VBScript compilation error ‘800a0409’
Undetermined string constant
Response.Write “

The second most common situation where the script is prematurely terminated is where the
syntax for ASP code is trying to be displayed. For example, if you are trying to display the
syntax for the Response.Write method on an HTML page, the following line of code produces
the same ‘Undetermined string constant’ runtime error shown in the previous example:
<% Response.Write(“Syntax: <% Response.Write Variant %> “) %> <!-- Run Time
➥Error -- >

To keep these errors from occurring, escape characters must be generated to ensure the Web
server properly receives the correct information from the Active Server engine. Escape charac-
ters can either be implemented by hand or automatically processed by using the Server object
encoding methods. To implement the proper escape characters by hand, the closing script tag
%> should be replaced with %\>. Proper solutions to both problems are illustrated in the follow-
ing example:
<%
Response.Write “<TABLE Border = 1 WIDTH = 100%\>”
Response.Write (Server.HTMLEncode(“Syntax: <% Response.Write Variant %\>”))
%>

For more information on the different encoding mechanisms, please see Chapter 12.

N O T E When using the Response.Write method, the variant datatype to be delivered to the
client cannot contain the string %>. This string is interpreted by the VBScript engine as an
end script delimiter. Instead, use the string %\> to write the close script delimiter tag. ■

The Write method uses the variant datatype to send information to the browser. The Variant
datatype, although the most versatile of datatypes, does have it limitations. The Variant
datatype itself can only contain 1,022 bytes of information. If you try to use a variable that is
larger than 1,022 bytes, a runtime error will occur.

18.1389-6 CH14 288 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Sending Output to the Browser: The Response Methods 289

The AddHeader Method

The AddHeader method enables you to add customized header information to the existing
HTTP header. The HTTP header information is the communication mechanism between every
request between the browser and the client. The header information is processed before any
output to the browser is displayed. To use the AddHeader method, use the following syntax:
Response.AddHeader name, value

where name is the name of the new header variable and value is the value assigned to the new
header variable.

The AddHeader method must be called before any content is sent to the browser. Failure to do
so will generate a runtime error because the HTTP protocol first sends header information to
the browser. After the header information is sent to the browser, the content is sent to the
browser. Content information is sent when any HTML or Response.Write method is called.
For example, you can use the preprocessing approach to identify a customized data encryption
technique before data is returned back to the user:
<% Response.AddHeader “DB-Encrypt”, “128-Bit” %>

In this situation, a custom header variable named “DB-Encrypt” is assigned a value of

“128-Bit” and is delivered to the browser.

The AppendToLog Method

The AppendToLog method of the Response object enables you to append information directly to
the Web server log file. This limited reporting feature is often used for error trapping and
limited reporting features. To use the Web server log functionality, use the following syntax:
Response.AppendToLog string

where Response is the built in Response object and string is an 80-byte character string.

One of the most common implementations of this method is to use the AppendToLog method as
a logging device or remote debugging tool. For example, you may want to track how many
times a custom ActiveX server component is accessed, but embedded scripting logic does not
always activate the component every time the page is accessed. Therefore, to track how many
times that component is called, the log file can be used to store that information.
<%
Response.AppendToLog “Billing Server Activated”
Call LoadBillingServer(“myMachineName”)
Response.AppendToLog “Billing Server Terminated” Part
%>
III
This code writes the following information to the Web server’s log file:
Ch
127.0.0.1, -, 11/24/97, 13:57:00, W3SVC1, SCOTDELL, 127.0.0.1, 161342, 485, 228,
➥200, 0, GET, /Logappend.asp, Billing Server Activated, 14
127.0.0.1, -, 11/24/97, 13:58:00, W3SVC1, SCOTDELL, 127.0.0.1, 161342, 485, 228,
➥200, 0, GET, /Logappend.asp, Billing Server Terminated,

18.1389-6 CH14 289 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 290 Chapter 14 Speaking to the Web Client: The Response Collection

Because the Internet Information Server log file is comma-delimited, avoid using commas in
the data string. The log file can then be exported to a variety of data sources, and customized
reporting can enable you to monitor the activity of your sites and applications. To get the most
from your Web server, use this logging capability with the NT Performance Monitor to get an
accurate representation of your site’s activity and resource management. For more information
on monitoring the IIS performance, please see Chapter 31, “Using Internet Information Server
4.0 to Manage Your Web Applications.” For an in-depth discussion on using the NT Perfor-
mance Monitor, please see Special Edition Using Windows NT Server, Second Edition
(0-7897-1388-8).

The BinaryWrite Method

The BinaryWrite method enables direct non-formatted output to be displayed to the request-
ing browser. This direct output is useful when displaying non-string information, such as vari-
ous image formats. To use the BinaryWrite method, use the following syntax:
Response.BinaryWrite data

where data is information that will be sent to the browser without any character conversion.

An example of using the BinaryWrite method is to display images on a Web page in thumbnail
format. Assuming that an ActiveX component that converts a full-size JPEG image into a
thumbnail format is installed on the server, the BinaryWrite method can be called to display
the image.
<%
Dim mySessionObj
set mySessionObj = Server.CreateObject(myActiveX.ThumbNailComposer)
myThumbNail = mySessionObj.MakeThumbNail(“Corporate Logo”)
Response.BinaryWrite myThumbNail
%>

This example instantiates the ThumbNailComposer component and calls the MakeThumbNail
method of that object. This method converts the full-size image, named “Corporate Logo”,
into a thumbnail image and stores that image temporarily in the myThumbNail image. Next, the
BinaryWrite method is used to display this binary image to the requesting page.

The Clear Method

The Clear method is used to erase any HTML that has been buffered on the server. The buffer
content is stored on the server by setting the Response.Buffer property to True. To destroy
the temporary storage location, the Clear property uses the following syntax:
Response.Clear

where Response is the built-in Response object and Clear is the Clear method. To clear any
HTML output that has been buffered, use the following line of code:
<% Response.Clear %>

There are two interesting points about the Clear method. The first point is that the Clear
method only clears the HTML output from the server and does not clear the HTML header
output. When the Response.Clear method is called, the header output is still sent to the

18.1389-6 CH14 290 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Sending Output to the Browser: The Response Methods 291

browser, but only the content is destroyed. The second point is that the Clear method only is
applicable if the buffer object has been created or turned on. If the Clear method is called and
buffering is not enabled, the following runtime error will occur:
Response object error ‘ASP 0159’

Buffering Off

Buffering must be on

To prevent this error from occurring, you can trap for the buffer property, as demonstrated in
Listing 14.3.

Listing 14.3 Using the Clear method to destroy buffered content.

<%@ LANGUAGE=”VBSCRIPT” %>
<% Response.Buffer = True %>
<HTML>
<BODY>
<%

Response.Write “This example demonstrates using buffered output that is never

➥displayed to the requesting user. The Clear method destroys the cached
➥content”
%>
</BODY>
</HTML>
If Response.Buffer Then ‘=== Checks if Buffering is enabled
Response.Clear ‘=== Clears the cached buffer and does not
➥display anything to the user
End If
%>

This example prevents a buffering runtime error from occurring if buffering was not activated.
By trapping the Buffer property using the if Response.Buffer then code, the buffer can be
properly cleared using the Response.Clear method.

The End Method

The End method also is used to manage the buffered server output. The End method returns
the current buffered output up to the point where the End method is called. To use the End
method, use the following syntax:
Part
Response.End

where Response is the built-in Response object and End activates the End method. To present
III
the buffered output to the browser up until the point where the End method is called, use the Ch
following code: 14
<% Response.End %>

The Response.End method is treated by the Active Server engine in a similar way to how the
Response.Redirect method is used. When the End method is called, the Active Server engine

18.1389-6 CH14 291 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 292 Chapter 14 Speaking to the Web Client: The Response Collection

terminates the processing of any remaining ASP scripts and sends the current output to the
client. Any remaining script is not processed, as shown in Listing 14.4.

Listing 14.4 Terminating script output using the End method.

<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<BODY>
<%
Response.Write “This information is processed and sent to the browser”
Response.End
Response.Write “But this information is never displayed”..
%>
</BODY>
</HTML>

The previous example uses the End method to display the welcome message and terminate
processing any remaining script. If you do not want display any information to the browser, you
can also combine the End and Clear methods, as shown in the following example:
<%
If CheckDataError(“StationID”) Then
If Response.Buffer Then ‘=== Checks if Buffering is enabled
Response.Clear ‘=== Clears the cached buffer
Response.End ‘=== Display nothing to browser
End if
End If
%>

The script example checks to see if the function CheckDataError returns an error message.
If an error is found, that is CheckDataError returns True, the Response buffer is cleared and
terminated using the End method. Because the Clear method is called immediately before the
End method, the End method sends no data to the client.

The Flush Method

The Flush method is used by the Response object to immediately send any buffered output to
the browser. The syntax for the Flush method is as follows:
Response.Flush

where Response is the built-in Response object and Flush calls the Flush method. The Flush
method can only be used if buffering has been activated. The following code uses the Flush
method to write the buffered output to the browser:
<% Response.Flush %>

The Flush method also requires that buffering be set to True before activating the Flush
method. Failure to set the buffering property properly will result in a runtime error. To prevent
the HTTP header runtime error from occurring and to allow the script to continue processing
throughout the page, use the following code:

18.1389-6 CH14 292 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 Sending Output to the Browser: The Response Methods 293

<%
If CheckDataError(“StationID”) Then
If Response.Buffer hen ‘=== Checks if Buffering is enabled
Response.Write (ErrorMsg) ‘=== Displays the error message
Response.Flush ‘=== Clears the cached buffer but
➥allow script to process
End If
Call WriteLogFile(“myLogFile”) ‘=== Record activity to log file
End If
%>

The previous example builds on the functionality used in the Clear and End methods, except in
this situation the Flush method is used to display the resultant error message and enable the
script to continue processing on the server. In this situation, after the error is returned to the
browser, a procedure named WriteLogFile records the success or failure of the event.

The Redirect Method

The Redirect method is another useful method of the Response object. The Redirect method
is used to route the browser to another Web page. The Redirect method is implemented by
using the following syntax:
Response.Redirect URL

where Response is the built-in Response object and URL is the Uniform Resource Locator that
indicates where the browser is routed. The Redirect method must be called before any HTML
is delivered to the browser. Failure to call the Redirect method before any HTML is sent to the
browser will generate the following header error message:
Response object error ‘ASP 0156’
Header Error
/YourASPFile.asp, line X
The HTTP headers are already written to the client browser. Any HTTP header
modifications must be made before writing page content

When the Response.Redirect method is called from an ASP, the browser is automatically
redirected to the new URL location. The Redirect method can be used to ensure that a user
who has your site bookmarked gets automatically routed to your new pages. This traditionally
occurs if a particular page is bookmarked and you change the name of the page. For example,
assume your corporate information Web page was just converted to an ASP to add functionality
to that page. The previous page, named CorpInfo.htm, is now renamed CorpInfo.ASP. How-
ever, if browsers have your page bookmarked, the requesting browsers will receive an HTTP
Object Not Found error. To prevent this from occurring, the Response.Redirect method can
Part
be used within the CorpInfo.htm file:
<%
III
Response.Redirect “CorpInfo.ASP” Ch
%>

This example now automatically routes the browser to the new and improved CorpInfo.ASP
14
page. Notice that once the server processes the Redirect command, the browser is automatically

18.1389-6 CH14 293 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 294 Chapter 14 Speaking to the Web Client: The Response Collection

routed to the new URL without processing script that is contained after the Redirect method,
as shown below:
<%
Response.Redirect “CorpInfo.ASP” ‘=== Redirects browser to new page
Dim myVar ‘=== This script is never processed
myVar = “Foo”
%>

In this example, the server never processes the script following the Redirect method, which
declares and sets the variable myVar. This script is ignored and the server immediately starts
parsing the CorpInfo.ASP page.

The Redirect method is most useful to help process and route non-visual code pages. These
non-visual code pages contain logic that is used to determine workflow through an application.
A good example of this routing feature is seen as the main page to many sites to help personal-
ize or tailor a site for a user preference. This non-visual page often checks the requestor’s
browser type, screen resolution, and for the existence of a cookie that stores user information.
<%
If Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”) Then
Response.Redirect “LoadActiveXSite.asp” ‘=== Load ActiveX site
Else
Response.Redirect “LoadHTMLSite.asp” ‘=== Load the HTML based site
End If
%>

In the previous example, the browser type is determined by using the Request object. Once the
browser type is determined, the user is automatically routed to pages that are optimized for the
specific browser. In this instance, if the browser is Internet Explorer, the user is transferred to
an ActiveX-based Web site. If the browser is any other browser, the HTML-based pages are
loaded.

You can also use the Redirect method within the GLOBAL.ASA event to ensure that all cookie-
compliant users are routed to a specific page when they enter an ASP application. For example,
the following code illustrates routing all uses to the Login.ASP page on the Session_OnStart
event.
<SCRIPT RUNAT=Server Language = VBScript>
Sub Session_OnStart
Response.Redirect(“Login.ASP”)
End Sub
</SCRIPT>

For more information on the GLOBAL.ASA file, please refer to Chapter 10, “The Composition of
an Active Server Page Application.” The Session object is covered in more detail in Chapter 13,
“Managing the User Session: The Session Object.”

The interesting point about this decision or routing process is that this automatic routing can
be applied to a variety of validation and workflow processes, or can be used to simply to access

18.1389-6 CH14 294 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 From Here… 295

a commonly referred library of ASP code. Chapter 32 demonstrates using several non-visual
ASP code pages to process application logic for a Web-based catalog system.

From Here…
In this chapter, you learned how the Response object plays a critical role in delivering a wide
range of information to the browser. The Response object uses the Cookies collection to man-
age writing information to cookies to requesting browsers. The Cookies collection use various
attributes that enable multiple values to be stored within a single cookie, explicit control over
how long a cookie will live on the browser, control over which domain and virtual directories
will cookies be issued, and whether or not to pass cookie information over secure or non-
secure channels. In addition, the Response object uses its various properties to determine how
information is sent to the browser, the length of time the delivered page will exist in the client’s
cache, and server-side buffering capabilities. Finally, the Response methods control writing
information to the HTTP header string and the delivered HTML page, the capability to write to
the Web server log files, control when information is released, and automatic routing capabili-
ties.

Please see the following chapters for related information:

■ Chapter 15, “Retrieving User Information: The Request Object,” discusses how to gather
information from the user using forms, hidden fields, and Cookies.
■ Chapter 17, “Implementing Existing Server-Side Components,” demonstrates how to use
the various components that ship with IIS Active Server Pages.
■ Chapter 18, “Creating Your Own Active Server Components,” walks you through the
tools and examples on how to create tailored functionality via custom created components.

■ Chapter 19, “Putting It All Together: Creating an Online Registration System,” illustrates
how to implement the different ActiveX components that ship with the Internet Informa-
tion Server to create a Web site for the Amazing Data Widgets Warehouse online catalog.

Part
III
Ch

14

18.1389-6 CH14 295 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 296 Chapter 14 Speaking to the Web Client: The Response Collection

18.1389-6 CH14 296 10/26/98, 9:25 AM

Brands03/stone3 SEU Active server Pages #1389-6 Sring 11/13/97 Sring CH14 lp#3
 15
297

C H A P T E R
Part
III
Ch

15

Extracting Data Using the

Retrieving User Form and QueryString
Collections

Information: The Discover how the Form and

QueryString collections are used to
gather and process information

Request Object from the browser.

Using the Certificate

Collection
Discover how to use the Certificate
collection to ensure client verification
through certificate authentication.

The Request objectis responsible for retrieving information Retrieving Cookie Information
from the Web browser. The Request object is filled with and Using the Server
various types of collections, properties, and methods that Collections
provide many ways to retrieve information from the user. Use the Cookie collection to manage
The Request object processes traditional Web-based forms and store client-side cookies, and
and provides a greater flexibility for gleaning information use the Server collection to use
from the user. This chapter will demonstrate how to use server and client variables to tailor
the Request object to retrieve information from the your application base.
browser. ■

19.1389-6 CH15 297 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 298 Chapter 15 Retrieving User Information: The Request Object

Understanding the Request Object

On the Web, extracting information from users via traditional forms processing provides basic
data-collection functionality. Data entered using input fields is submitted to an executable pro-
cess, typically through Common Gateway Interface, on the Web server. This process, which
usually runs as an out-of-process thread, receives and manipulates the user input. Most of
these processes convert form data into automated e-mail that is sent to a person to process,
write to a text file, or manipulate a database.

The transfer of information from the client to the Web server is accomplished by using variable
names and associated values embedded in the URL. Typically, the information is sent to
the Web server using a Post method. The Post method transfers the data from the client to
the Web server. The Post method is determined by the ACTION event of the HTML form. The
ACTION event is triggered when the form is activated by a SUBMIT event, as seen in the following
HTML:
<FORM Name = “frmLogin” ACTION = “ValidateLogin.ASP” METHOD = “POST”>

When the previous example is submitted, the information on the form named “frmLogin” is
posted or sent to the ASP file “ValidateLogin” for processing. The Submit event of the form is
usually triggered by a client event, such as clicking on a button or image. The following HTML
creates a Submit button with a caption displaying “Login”.
<INPUT TYPE = “SUBMIT” VALUE = “Login”>

Keep in mind that you can also trigger the Submit event of the form by executing the
Form.Submit event. For example, the following JavaScript function is executed on the client’s
browser to submit a form only if the username and password fields have been completed.
function chklogin(form)
{
if (form.txtusername.value == “”){
alert(“Please Enter a User Name”);
setfocususername(form);}
else{
if (form.txtpassword.value == “”){
alert(“Please Enter a Password”);
setfocuspassword(form);}
else{
form.submit();
}
}
}

The capability to accept information from the browser has expanded from simple URL variable
manipulation. In order to develop and deploy your application over the Web, you also need the
ability to accept information from forms, accept and reject digital certificates to identify users,
and read cookie information from the requesting browser.

19.1389-6 CH15 298 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 299

The Request object was initially created to help control and manipulate the data flowing into Part
the Web server as a result of the Post method. The Request object uses collections, properties,
and methods to retrieve information from the user. To use the Request object, use the follow-
III
ing syntax: Ch
Request[collection|property|method](variable) 15
where Request is the Request object, collection is the available collection, property is the
Request properties, method is the Request methods, and variable is a Request variable.

For example, to reference a Request variable named OpenDatabase, simply reference the vari-
able name, as shown in the following example.
<% = Request(“OpenDatabase”) %>

Now, lets take a closer look at the Request object’s collections and methods.

Accepting User Information Using the Request

Collections
The Request object uses separate objects that can be grouped together and referred to by code
as a single unit. This grouping of common objects is referred to as a collection. The Request
object uses five collections, listed in Table 15.1, to interface with the calling client.

Table 15.1 The Request Collection

Collection Name Description
ClientCertificate Retrieves the certification fields issued by the Web browser.
Cookies Retrieves the values of cookies sent in an HTTP request.
Form Retrieves the values of form elements posted to the HTTP
request.
QueryString Retrieves the values of the variables in the HTTP query
string.
ServerVariables Retrieves the values of predetermined environment variables.

The ClientCertificate is used to help verify that information is coming from a trusted client.
The digital identification is maintained by a third party to help verify not only that users are
who they say they are, but that the server is properly represented by the appropriate hosting
facility. The Cookie collection is used to access information temporarily stored in text files on
the client’s machine. The Form collection is used to directly access information from specific
field values on an HTML form. The QueryString collection is used to help manipulate variables
that are passed through the URL. The ServerVariable collection is used to identify the re-
questing browser’s environment.

19.1389-6 CH15 299 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 300 Chapter 15 Retrieving User Information: The Request Object

If a Request variable is not found, the Request object automatically starts scanning its collec-
tions for the first instance of that variable. The Request object looks for the variable in its col-
lections in the following collection order:

1. QueryString
2. Form
3. Cookies
4. ClientCertificate
5. ServerVariables

If multiple instances of the variable name exist within one collection, the first instance located
will be used.

TIP To minimize the possibility of accessing an incorrect variable, you should refrain from giving two
variables in different collections the same name.

The ClientCertificate Collection

The Request object’s ClientCertificate collection is used to provide proper security identifi-
cation across unsecured environments. Controlling and monitoring your information in this
environment consist of regulating the aspects of information dissemination. The first area,
privacy, is used to prevent unwanted parties from viewing information that has been inter-
cepted in transit. Using encryption techniques such as Secure Socket Layers helps to ensure
that your data will remain intact as it crosses unsecured environments. The second security
area is authentication. Now that the information can safely pass between points, it is necessary
to verify that the users are who they say they are.

To deal with these security concerns, the ClientCertficate collection helps ease the burden
of developing digitally secure applications. Specifically, the ClientCertificate collection is
used to retrieve certification fields issued by the Web browser. The ClientCertificate is used
when the requesting Web browser uses the SSL3.0/PCT1 protocol to create a secure connec-
tion to the Web server. If the client requests a secure connection, the browser sends the certifi-
cation fields to the server. In turn, the server retrieves the certification information via the
ClientCertificate collection.

The certification process is not a built-in component of Active Server Pages. However, IIS 4.0
ships with a Certificate Server to help create trusted user communities. To learn more about
how to configure your Web server to request client certificates, see the documentation pro-
vided with the Certificate Server. To use Certificate Server values, use the following syntax:
Request.ClientCertificate( Key[SubField] )

where Key is the name of the certification field and SubField is the individual field located
within the certification field. Table 15.2 lists the various certificate fields available.

19.1389-6 CH15 300 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 301

Table 15.2 The ClientCertificate Certification Fields Part

Key Description III

Subject Returns a list of subfield values that contain information about the Ch
subject of the certificate.
15
Issuer Contains a list of subfield values containing information about the
issuer of the certificate.
ValidForm Returns when the certificate becomes valid.
ValidUntil Specifies when the certificate expires.
SerialNumber Returns the certification serial number as an ASCII representation
of hexadecimal bytes.
Certificate Returns the binary stream of the entire certificate content in ASN.1
format.

The use of subfields enables specific information to be retrieved from the Subject and Issuer
Key fields, as mentioned above. Table 15.3 presents the available subfields for the
ClientCertificate collection.

Table 15.3 The ClientCertificate Collection Subfields

Value Description
C Specifies the name of the country of origin.
O Specifies the company or organization name.
OU Specifies the name of the organizational unit.
CN Specifies the common name of the user.
L Specifies a locality.
S Specifies a state or province.
T Specifies the title of the person or organization.
GN Specifies a given name.
I Specifies a set of initials.

If no certificate is sent, the ClientCertificate collection returns EMPTY.

Retrieving Cookie Information Using the Cookies Collection

The Cookies collection is used by the Request object to retrieve values stored in text files on
the client’s machine. The Cookies collection requires the following syntax:
Request.Cookies(cookie)[(key)|.attribute]

19.1389-6 CH15 301 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 302 Chapter 15 Retrieving User Information: The Request Object

where cookie is the name of the requested cookie, key is the index of the subkey values, and
attribute isused to identify the property of the cookie.

For example, to extract the data in a cookie named mySiteCookie and write that value to the
page, use the Cookies collection as follows:
<%= Request.Cookies(“mySiteCookie”) %>

The Cookies collection has only one property: the HasKeys property. The HasKeys property is
used to determine whether the cookie has multiple values in a dictionary format. A Dictionary
object is a term that is used to describe a set of stored data items, usually in a Key/Value format.

If the HasKeys property returns True, then multiple keys exist within the cookie. To test for
multiple keys in a cookie named myLowFatCookie, use the following code:
<%= Request.Cookies(“myLowFatCookie “).HasKeys %>

The “myLowFatCookie” cookie is the one created in Chapter 14, “Speaking to the Web Client:
The Response Collection.” To create this cookie with key values, use the following code:
<%
Response.Cookies(“myLowFatCookie”)(“Breakfast”) = “Banana”
Response.Cookies(“myLowFatCookie”)(“Lunch”) = “Apple”
Response.Cookies(“myLowFatCookie”)(“Dinner”) = “Orange”
%>

In most situations, you want to make decisions based on the value of the HasKeys property.
This is done by using the results of the HasKeys property in a conditional statement, such as an
If...Then loop, as displayed in the following example:

<%
If Request.Cookies(“myLowFatCookie”).HasKeys then
Response.Write(“You have Keys!”)
else
Response.Write(“No Keys Found”)
End if
%>

Now that the cookie is determined to have multiple key fields, the keys can be extracted and
assigned to variables to be manipulated by the ASP script. The testing and extracting of cookie
values to server-side variables is demonstrated in the following example:
<%
If Request.Cookies(“myLowFatCookie”).HasKeys then
vBreakfastMenu = Request.Cookies(“myLowFatCookie”)(“Breakfast”)
vLunchMenu = Request.Cookies(“myLowFatCookie”)(“Lunch”)
vDinnerMenu = Request.Cookies(“myLowFatCookie”)(“Dinner”)
else
vMsgStr = “No Message”
End if
%>

This example extracts and assigns vBreakfastMenu the value of Banana, vLunchMenu the value
of Apple, and vDinnerMenu is assigned the value of Orange, from the myLowFatCookie created in
the previous code example.

19.1389-6 CH15 302 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 303

You can also utilize the collections to cycle through all keys within a collection. The capability Part
to cycle through the keys is made possible by using a For...Next loop. The following example
writes the key names and the associated values to the Web page:
III
Ch
<%
If Request.Cookies(“myLowFatCookie”).HasKeys then
For Each key in Request.Cookies(“myLowFatCookie”)
15
Response.Write(key &”: “)
Response.Write(Request.Cookies(“myLowFatCookie”)(key) &”<BR>”)
Next
else
Response.Write(“No Keys Found”)
End if
%>

This example writes the following output to the Web page:

LUNCH: Strawberry
DINNER: Orange
BREAKFAST: Banana

This same concept of manipulating the keys in the Cookies collection can also be applied to
cycling through multiple cookies in the Cookies collection, as shown in the following example:
<%
For Each cookie in Request.Cookies
Response.Write(cookie & ” = “ & Request.Cookies(cookie))
Next
%>

The Form Collection

The Request object Form collection helps to facilitate data retrieval from HTML forms. The
Form collection captures the value from an HTML form that submits information via the HTTP
Post method. The Forms collection requires the following syntax:

Request.Form(parameter)[(index)|.Count]

where parameter is the name of the Form collection, index is the specific form element, and
Count identifies the number of elements that exist on a form. For example, to obtain a value
entered in a text field named txtUserFirstName from a form, use the Form collection to refer-
ence the field name, as shown in the following code:
Hello <%= Request.Form(“txtUserFirstName”) %>

The form used for this example is shown in Figure 15.1.

As you can see in Figure 15.1, the previous code example writes Brian to the Web page. Keep
in mind that hidden fields can also be referenced via the Forms collection. Hidden fields are
often used as a temporary memory location on the Web page. This temporary memory loca-
tion gives some flexibility in storing and keeping track of user information.

19.1389-6 CH15 303 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 304 Chapter 15 Retrieving User Information: The Request Object

FIG. 15.1
Use the Form collection
to retrieve values from
the requesting browser.

The Form collection itself actually contains all the Form objects and their values from the posted
information. For example, the following line of code displays the entire Form object from the
posted information:
<% Response.Write (Request.Form) ‘=== Displays the entire form object %>

This Request.Form returns the entire Form collection to the page, as displayed in the next line
of code:
FormfrmHidden = lowRes & txtUserFirstName = Brian & txtUserForm = Dawson &
txtUserForm = 23-DD & txtUserForm = Arizona & btnProfile = Add + Profile

When a particular field or parameter is referenced, the Request object parses the Form Collec-
tion and returns the value. For example, the following scripts take the user name and password
to create a SQL string:
<%
dim vUserName, SQLstr
vUserName = Request.Form(“txtUserFirstName”) & “ “ &
Request.Form(“txtUserLastName”)
SQLstr = “SELECT Accntnum FROM Activity WHERE “
SQLstr = SQLstr & “UserAcct = ‘“ & vUserName & “’”
%>

The generated SQL string from the Form collection would be the following:
SELECT Accntnum FROM Activity WHERE UserAcct = ‘Brian Dawson’

One of the most powerful utilities of the Form collection is the capability to iterate though the
collection. The iteration approach that enables you to cycle through the collection is imple-
mented by using the For...Next loop. For example, to iterate through the Form collection and
display each field name and corresponding value, each item in the Form collection is read, as
shown in the following example:
<%
For Each x In Request.Form
Response.Write(x & “ = “ & Request.Form(x) & “<BR>”)
Next
%>

This iterative parsing of the Form collection produces the following output:
BTNPROFILE = Add Profile
TXTUSERMACHINENAME = Arizona
TXTUSERID = 22-DD
TXTUSERLASTNAME = Dawson
TXTUSERFIRSTNAME = Brian
FRMHIDDEN = LowRes

19.1389-6 CH15 304 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 305

Furthermore, the same approach can be applied to iterating through a form element that has Part
multiple values. To determine if a form element has multiple values, the Count property of the
Form collection is used, as seen in the following example:
III
Ch
<% vCount = Request.Form(“myHTMLElement”).Count %>

If the Count property returns 0, the element is not found. If the Count property returns a value 15
greater than 1, the form element has multiple values. These multiple elements can then be
referred to by their index. For example, Figure 15.2 illustrates three radio boxes that enable
the user to vote for his favorite browser. The radio buttons are all named the same to create a
control array named rdoBrowser.

FIG. 15.2
Use the Form
collection to cycle
though a control array
to determine the user
input.

Determining which button the user has selected within the control array without the use of
collections can be a tedious task. However, using collections helps eliminate unnecessary
coding and enables iteration through all the items in the Form collection. Once the selected
value is located, the value of the selected radio button can be used for coding purposes. The
following code cycles through the control array determines the value that is selected, and
displays the value of the radio box:
<%
For Each item In Request.Form(“rdoBrowser”)
Response.Write item & “<BR>”
Next
%>

The previous code finds that the selected value of the radio box of the Internet Explorer radio
box is selected and passes the embedded value of IE to the Web page. This iteration technique
can also apply to multi-selected items. For example, if the Internet Explorer and Netscape
Navigator radio boxes were selected, the information stored in the HTML values for the radio
boxes would return InternetExplorer and NetscapeNavigator would be written to the
ASP page.

The QueryString Collection

The QueryString collection is used by the Request object to extract variables from the HTTP
query string. The query string is textual content that occurs after the question mark character
(?) in the URL string. For example, the following URL points to the www.planet.locator.com
domain and executes the checkplanet.asp script. This script accepts the query string of
planet = Jupiter for processing by the page:

HTTP://www.planet.locator.com/scripts/checkplanet.asp?planet=Jupiter

19.1389-6 CH15 305 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 306 Chapter 15 Retrieving User Information: The Request Object

To extract this information from the query string, the Request object requires the following
syntax:
Request.QueryString(variable)[(index)|.Count]

where Request is the Request object, variable is the name of the variable name in the query
string, index is the element index, and Count specifies the number of variables in the query
string.

To reference the variable name using the Request object, call the sent variable name. For ex-
ample, to write to a variable named myPlanet from a QueryString variable named planet, use
the following syntax:
<% myPlanet = Request.QueryString(“planet”) %>

If the URL http://www.planet.locator.com/scripts/checkplanet.asp?planet=Jupiter

were submitted to the server, the text Jupiter would be assigned to the variable. In most
situations, you will use these variables to drive workflow issues, such as constructing queries
to guide information on the resultant pages. For example, the following syntax creates a SQL
statement from input of the user that is used to find all information about the planet Jupiter.
<%
Dim strSQL ‘=== Initiates variable
strSQL = “Select * From tbPlanet ” ‘=== Creates SQL string
strSQL = strSQL & “where fldPlanet = “
strSQL = strSQL & Request.QueryString(“planet”)
%>

When this section of script is finished executing, the SQL string variable strSQL returns Se-
lect * From tbPlanet where fldPlanet = Jupiter. Notice that the QueryString collection
is not the only means to extract the information from the URL. You can also use the Request
object and directly reference the planet variable name using the following code:
<% Request(“planet”) %>

Therefore, the same SQL string can be generated using the previous Request object syntax:
<%
Dim strSQL ‘=== Initiates variable
strSQL = “Select * From tbPlanet ” ‘=== Creates SQL string
strSQL = strSQL & “where fldPlanet = “
strSQL = strSQL & Request (“planet”)
%>

Just as I introduced keys and indexes in the Cookies and Forms collections, the QueryString
collection also utilizes this functionality. The Request.QueryString(parameter).Count prop-
erty can be used to determine if a parameter has multiple values. Table 15.4 discusses the
return results of the Count property.

19.1389-6 CH15 306 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 307

Table 15.4 The QueryString Collection Count Property Part

Return Description III

0 Parameter not found. Ch

1 A single parameter is found; multiple values do not exist. 15

>1 Multiple values for the parameter exist.

If the Count property returns a 0, the QueryString parameter does not exist. If the Count prop-
erty returns a 1, only a single instance of that parameter exists. If any other result is returned,
multiple values for the parameter exist. To access a parameter with multiple values, the index
value must be used.

Consider the following URL link that contains today’s sale items:
<a href
=”querystringresults.asp?SaleItem=Dry+Cigars&SaleItem=Pink+Socks&SaleItem=
➥Narrow+Ties”>Today’s Sale Items</a><BR>

This link refers to an ASP named querystringresults.asp on the server. If the requested
page were scanning the Request object for a variable named DiscountedItem, no results would
be found because the parameter DiscountedItem does not appear in the QueryString.

You use the Count property to help manage and control the content on the page. For example,
you do not want your pages to be dependent on specific information being passed to the
page; you want to design very flexible pages that enable the content to fit into a template. If
you want to design a page to display specialty items, such as sales, discounted items, or new
items, several approaches can be taken. The simplest approach is to design a single page for
each type of specialty item. However, this requires three pages to be created and maintained,
and the calling links to these pages have to be updated. A more efficient manner in terms of
design and maintenance is to create one page that changes depending on the input. This is best
implemented by using the Count property. The following code checks to determine whether
the variable is passed to the page and if it is passed to the page, displays the data:
<%
if Request.QueryString(“DiscountItem”).Count then
Response.Write(Request.QueryString(“DiscountItem”))
End if
%>

If the previous sale item URL were passed to the Request object, the Discount Count property
would return False and no unnecessary discounted items would be displayed on the page.
However, to display the SaleItem information, the previous example can be expanded as follows:
<%
if Request.QueryString(“DiscountItem”).Count then
Response.Write(Request.QueryString(“DiscountItem”))
Elseif Request.QueryString(“SaleItem”).Count then
Response.Write(Request.QueryString(“SaleItem”))
End if
%>

19.1389-6 CH15 307 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 308 Chapter 15 Retrieving User Information: The Request Object

If the previous code were used, the Request.QueryString(“SaleItem”).Count would return 3.

If the SaleItem parameter is referenced, the entire SaleItem parameter is used. For example,
the previous example referencing Request.QueryString(“SaleItem”) would write the follow-
ing content to the page:
Dry Cigars, Pink Socks, Narrow Ties

Now the parameter array items can be accessed in two different ways. The first is to use the
Count property to iterate through the index. This approach is shown in the following example:

<%
Dim vCount
vCount = Request.QueryString(“SaleItem”).Count
For I = 1 To vCount
Response.Write Request.QueryString(“SaleItem”)(I) & “<BR>”
Next
%>

This code determines the number of items in the index and cycles through the items in the
collection to write the values to the page.

The second method to obtain all the values from the QueryString collection is to use a
For...Next loop to identify each unit in the collection. This same method demonstrated in the
Cookies and Form collection processing can be applied to the QueryString collection, as shown
below:
<%
For Each item In Request.QueryString(“SaleItem”)
Response.Write item & “<BR>”
Next
%>

Now the previous page display example can be expanded to handle multiple input types and
display all information passed to the page, as shown below.
<%
‘=== Display Discount Items ===
If Request.QueryString(“DiscountItem”).Count Then ‘=== Checks to see
➥item exists
Dim vCount
vCount = Request.QueryString(“DiscountItem”).Count ‘=== Sets item
➥count
For I = 1 To vCount
Response.Write Request.QueryString(“DiscountItem”)(I) & “<BR>”
Next
End If

‘=== Display Sale Items ===

If Request.QueryString(“SaleItem”).Count Then ‘=== Checks to see
➥item exists
For Each item In Request.QueryString(“SaleItem”)
Response.Write item & “<BR>”
Next
End if
%>

19.1389-6 CH15 308 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 309

Notice how the two different uses of the For loops are used to iterate through the QueryString Part
collection. The first loop that displays the discounted items uses the Count property to step
through the QueryString index. The second loop is responsible for iterating through the
III
SaleItem collection using the For Each statement. Ch

15
The ServerVariables Collection
The ServerVariables collection is used to obtain server environmental variables. However,
these server variables are not specific to the server and can obtain information from the re-
questing client. To use these features, the ServerVariables collection requires the following
syntax:
Request.ServerVariables (ServerVariable)

where ServerVariable is name of the server variable. Table 15.5 displays a list of common
server variables.

Table 15.5 The Server Environmental Variables

Variable Name Description
*ALL_HTTP Displays all HTTP headers sent by the client.
*ALL_RAW Retrieves all headers as they are sent by the client.
*APPL_MD_PATH Retrieves the metabase path for the ISAPI DLL.
*APPL_PHYSICAL_PATH Retrieves the physical path corresponding to the
Metabase.
AUTH_TYPE Displays the authentication method used by the server.
*AUTH_USER Retrieves the authenticated user name in raw format.
AUTH_PASSWORD Displays the value entered in the client’s authentication
dialog using basic authentication security.
*CERT_COOKIE Returns the Unique ID for client certificate.
*CERT_FLAGS Displays whether the client certificate is valid.
*CERT_ISSUER Displays issuer field of the client certificate.
CERT_KEYSIZE Returns the number of bits in Secure Sockets Layer
connection key size.
*CERT_SERIALNUMBER Displays the client certificate serial number field.
*CERT_SERVER_ISSUER Displays the of the server certificate issuer field.
*CERT_SERVER_SUBJECT Displays the server certificate subject field.
*CERT_SUBJECT Displays the client certificate subject field.
CONTENT_LENGTH Returns the length of the content.
continues

19.1389-6 CH15 309 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 310 Chapter 15 Retrieving User Information: The Request Object

Table 15.5 Continued

Variable Name Description
CONTENT_TYPE Returns the data type of the content.
GATEWAY_INTERFACE Returns the version of the CGI specifications used on the
server.
HTTP_<HeaderName> Returns the information in the HeaderName.
*HTTPS Returns whether or not the request came in through
secure channel (SSL).
*HTTPS_KEYSIZE Returns the number of bits in Secure Sockets Layer
connection key size.
*HTTPS_SECRETKEYSIZE Returns the number of bits in server certificate private
key.
*HTTPS_SERVER_ISSUER Returns the server certificate issuer field.
*HTTPS_SERVER_SUBJECT Displays the server certificate subject field.
*INSTANCE_ID Returns the ID for the IIS instance.
*INSTANCE_META_PATH Returns the Metabase path for the instance of IIS that
responds to the request.
LOGON_USER Displays the NT login account the request is made from.
PATH_INFO Displays the server path information.
PATH_TRANSLATED Returns the translated version of the PATH_INFO.
QUERY_STRING Returns the query string in the URL.
REMOTE_ADDR Displays the IP address of the requesting machine.
REMOTE_HOST Displays the name of the requesting host.
REQUEST_METHOD Returns the method that initiated the request.
SCRIPT_NAME Displays the virtual path to the executing script.
SERVER_NAME Returns the server’s host name, DNS alias, or IP
address.
SERVER_PORT Returns the server port number the request is made on.
SERVER_PORT_SECURE Returns a 1 if request is made on a secure port, 0 if
unsecured.
SERVER_PROTOCOL Returns the name and version of the requesting protocol.
SERVER_SOFTWARE Returns the name and version of HTTP server.
URL Returns the base portion of the URL.

(* indicates the new IIS 4.0 server variables.)

19.1389-6 CH15 310 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 Accepting User Information Using the Request Collections 311

The following examples are some of the possible results of scaning various server variables. Part
REQUEST_METHOD GET SCRIPT_NAME /demo/servervariables.asp
SERVER_NAME 207.96.34.07
III
SERVER_PORT 80 Ch
SERVER_PORT_SECURE 0
SERVER_PROTOCOL HTTP/1.1 15
SERVER_SOFTWARE Microsoft-IIS/4.0
URL /Demo/servervariables.asp
HTTP_ACCEPT application/vnd.ms-excel, application/vnd.ms-powerpoint, image/gif,
image/x-xbitmap, image/jpeg, image/pjpeg, application/msword, */*
HTTP_ACCEPT_LANGUAGE en-us
HTTP_CONNECTION Keep-Alive
HTTP_HOST 207.96.34.41
HTTP_USER_AGENT Mozilla/4.0 (compatible; MSIE 4.0; Windows NT)
HTTP_ACCEPT_ENCODING gzip, deflate

Listing 15.1 demonstrates the capability to display all the items in the ServerVariables collec-
tion and their respective values.

Listing 15.1 Use the ServerVariable collection to display information about

the Web server and the requesting client.
<TABLE border = “1”>
<% For Each name In Request.ServerVariables %>
<TR>
<TD> <%= name %> </TD>
<TD> <%= Request.ServerVariables(name) %> </TD>
</TR>
<% Next %>
</TABLE>

One of the most useful capabilities of the ServerVariable collection is the capability to solicit
environmental information from the requesting browser. To obtain the requesting browser’s
environmental information, capitalize on the HTTP_<HeaderName> component of the server
variables. The HTTP_<HeaderName> variables can contain a wide variety of information about
the browser client. For example, if an Internet Explorer 3.0 browser makes a request to the IIS,
the browser sends a variety of information, such as the screen resolution, operating system,
processor type, and more:
HTTP_ACCEPT
HTTP_ACCEPT_LANGUAGE
HTTP_CONNECTION
HTTP_HOST
HTTP_USER_AGENT
HTTP_PRAGMA
HTTP_COOKIE
HTTP_UA_PIXELS
HTTP_UA_COLOR
HTTP_UA_OS
HTTP_UA_CPU

19.1389-6 CH15 311 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 312 Chapter 15 Retrieving User Information: The Request Object

Keep in mind that the individual browser dictates what information is sent to the server and is
now not consistent between browsers. For example, a Netscape Navigator 3.0 browser would
only send the following variables:
HTTP_ACCEPT
HTTP_CONNECTION
HTTP_HOST
HTTP_USER_AGENT

From these headers, various information about the browser can be determined. In fact, the
Browser Capability Component uses the HTTP header information to help identify characteris-
tics of the browser. For more information about this component, please see Chapter 17, “Imple-
menting Existing Server-Side Components.”

With this information available, you can help control and customize your application based on
the user environment. The significance here is that you do not need to rely on the user to cor-
rectly fill out a preference list or worry about storing this information in cookies. For example,
you can customize the layout of your site or application depending on the requesting browser
type. Although this browser identification capability is available as an ASP component, you can
extract the information directly from the header using the HTTP_USER_AGENT server variable
without creating an instance of the Browser Capability Component:
<%
if Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”) then
‘=== IE specific tags
Elseif Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”Mozilla”) then
‘=== Netscape tags are placed here
end if
%>

This example combines the use of server variables with the VBScript InStr function. The
InStr function is used to search for a string within another string. For example, the first line of
code uses the InStr function to search the USER_AGENT header string for the substring MSIE. If
that string is found, the requesting browser is Internet Explorer and can utilize functionality
specific to that browser. You can further refine your application by examining the display and
operating system variables. The following information is a result of trapping the HTTP_<Header>
information for a client browser:
HTTP_UA_PIXELS 800x600
HTTP_UA_COLOR color8
HTTP_UA_OS Windows NT
HTTP_UA_CPU x86

Finally, The HTTP_<Header> is expandable to accept customized header information. To accept

a customized header, simply append the header name onto the HTTP_ string. For example, if the
browser sends a header named myHeader to the server, the Request object can process the
value assigned to that header by using the following syntax:
<% vMyHeader = Request.ServerVariables(“HTTP_myHeader”) %>

19.1389-6 CH15 312 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 The Request Properties and Methods 313

The Request Properties and Methods Part

Now that the different Response collections have been covered, the Request object has one III
property and one method. The TotalBytes property dictates the total amount of bytes sent in Ch
the request. The BinaryRead method is used to accept and store data sent from a browser.
15
The TotalBytes Property
The TotalBytes property is used to return the number of bytes sent by a browser. To transfer
this information to the server, the client uses the Post method to send form information in the
HTML body tags. The TotalBytes property requires the following syntax:
Request.TotalBytes

To use the total number of bytes, assign the output of the Request object to a variable, as
shown below:
<% myByteCount = Request.TotalBytes %>

This byte information is determined by examining the length of the Form collection. For ex-
ample, if the User Profile form displayed in Figure 15.1 is posted to the server, the value of the
TotalBytes equals 128. The length of the Form object is also 128.

The BinaryRead Method

The BinaryRead method is used to read binary information that is sent to the server from a
Post request. To use the BinaryRead method, use the following syntax:

myBinArray = Request.BinaryRead(count)

where count is the number of bytes to place into an array named myBinArray. To perform a
binary read from information submitted on a form, use the following example:
<%
Dim binread
binread = Request.BinaryRead(Request.TotalBytes)
%>

Notice that you cannot call or reference any item in the Form collection once the BinaryRead
method has been called. For example, if the following code is executed, a runtime error will
occur:
<%
vCache = Len(Request.Form)
Dim binread
binread = Request.BinaryRead(Request.TotalBytes)
%>

Executing the code will generate the following error:

Request object error ‘ASP 0206’

Cannot call BinaryRead

Cannot call BinaryRead after using Request.Form collection.

19.1389-6 CH15 313 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 314 Chapter 15 Retrieving User Information: The Request Object

From Here…
From this chapter, you can see that the real power of the Request object lies within the five
collections listed below. The collections can be utilized to extract information from locations on
the browser.

The Form collection helps you obtain information from the information submitted to the
server.
The Cookies collection is used to help control and manage not only information stored
within the client-side cookies, but also the properties of the cookies.
The QueryString is used to extract information passed in the URL.
The ServerVariables are used to determine both server and browser environmental
conditions.
The ClientCertificates are used to ensure the identity of the requesting browser.

Now that retrieving information using the Request object has been covered, the next step is to
discover how to use ActiveX components to extend the Web server’s functionality. These
ActiveX components can exist from components that are shipped with the Internet Information
Server or can be created using your own ActiveX development tools. To learn about related
topics, please see the following chapters:
■ Chapter 17, “Implementing Existing Server-Side Components,” demonstrates how to use
the various components that ship with IIS Active Server Pages.
■ Chapter 18, “Creating Your Own Active Server Components,” walks you through the
tools and examples on how to create tailored functionality via custom-created components.
■ Chapter 20, “Database Access with ADO,” demonstrates how to use database informa-
tion to build the Amazing Data Widget Warehouse Catalog.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.

19.1389-6 CH15 314 10/26/98, 9:26 AM

Brands03/stone3 SEU active server pages #1389-6 SRing 11/13/97 CH15 lp#3
 16
315

C H A P T E R

Part
Using Application and Session III
Putting It All Together: Arrays Ch
Learn how to use application- and
16
Building an Online session-level variables and arrays to
create power programming tools.

Catalog Controlling the Interface

Implement the Request and Response
objects to control the interaction
between the browser and the Web
server.

Managing the Server

Discover how to use the Server
This chapter will demonstrate how to use the different object to control server-side tasks
Active Server objects to collectively form a simple online and how non-visible ASPs act as non-
catalog system. The Amazing DataWidgets WareHouse compiled programming libraries.
company uses the various Active Server Page objects to
provide a simple online catalog system that enables users
to browse catalog listings and select items to order and
then process the order. For demonstration purposes for
this part of the book, this example excludes any database
manipulation that might detract from the focus on the
Active Server objects. Instead, this chapter focuses on
demonstrating the functionality of each Active Server
object to provide the framework that is needed to add
more advanced functionality in later chapters. ■

20.1389-6 CH16 315 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 316 Chapter 16 Putting It All Together: Building an Online Catalog

The ASP Online Catalog Application: An Overview

The Amazing DataWidgets WareHouse Online catalog provides online users the opportunity to
visit a Web site and order from DataWidgets’ growing online catalog. Once the user has en-
tered the online software catalog, the catalog application will display and list the three products
available for sale, as shown in Figure 16.1.

FIG. 16.1
The Active Server Pages
objects are used to
collect, process, and
return information to the
user.

From these items, the shopper can select items to place into his electronic shopping basket.
The shopping basket is used to display what items the user has selected to purchase. The user
can select just one item to purchase or he can continue shopping for other items. The Active
Server Pages remember and keep track of what items are placed in the shopping basket spe-
cific to each individual user session, as shown in Figure 16.2.

After the user is ready to purchase the items he selected in his virtual shopping session, the
application asks the user for shipping information. After the shipping information is gathered,
the order is processed and a summary page is returned to the user, as shown in Figure 16.3.

To deliver this function, the Active Server Pages use five of the six Active Server objects:
the Application object, the Session object, the Response object, the Server object, and the
Request object. The rest of this chapter takes a closer look at how each object was used in the
application.

20.1389-6 CH16 316 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 The ASP Online Catalog Application: An Overview 317

FIG. 16.2
The shopping basket is
an example of tracking
the specific actions of
an individual user
session.

Part
III
Ch

16

FIG. 16.3
The Active Server
objects process the
user’s order request
and shipping informa-
tion and return a
summary page.

20.1389-6 CH16 317 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 318 Chapter 16 Putting It All Together: Building an Online Catalog

Pre-Loading Application Variables Using the

GLOBAL.ASA
The first stage in building the Amazing DataWidgets WareHouse is to make the catalog data
accessible to the user. Effective communication of information to your browser is a combina-
tion of determining the right data access methods and how to best use the information re-
trieved. The Amazing DataWidgets WareHouse Catalog site, like most online catalog service
sites, does not require constant interaction with the database because the schedule information
is somewhat static. Of course, you need greater interaction when you register to information
pertaining to that site, but the catalog or schedule information is relatively constant.

Knowing this information about the site’s content and wanting to achieve the greatest perfor-
mance by reducing server input/output, the software catalog section is loaded as an applica-
tion-level variable. As you saw in Chapter 11, “Controlling the Application: The Application
Object,” application-level variables are accessible across all users of the site. Because multiple
users have access to this information, the Lock and UnLock methods are used to control up-
dates to all application variables.

Because the catalog can be formed in logical groups, you are going to use application-level
arrays to provide information about the different products and to track how many of the prod-
ucts the user wants to purchase. In addition, you are going to utilize the application-level arrays
for preloading other information on the Web server. For example, the Amazing DataWidgets
WareHouse also has an online computer hardware catalog. Therefore, to provide a system that
can easily be expanded into other catalog areas, you are going to create a master array called
aCatalog to list the current catalog services.

Creating Application-Level Variables and Arrays

Although an application-level variable can be created anywhere in the application, the logical
choice is to load the variable when the application starts up in the Application_OnStart event
of the GLOBAL.ASA file. For example, consider that the Amazing Web site is going to contain two
catalogs within the same site. The catalogs are contained in the aCatalog array within the
Application_OnStart event, as shown in GLOBAL.ASA file in Listing 16.1.

Listing 16.1 Loading application-level arrays in the GLOBAL.ASA file.

<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Application_OnStart
Dim aCatalog(2,2)
aCatalog(1,1)=”Software Components”
aCatalog(1,2)=”SoftwareCatalog.asp”
aCatalog(2,1)=”Hardware Components”
aCatalog(2,2)=”HardwareCatalog.asp”
Application(“aCatalog”)=aCatalog

20.1389-6 CH16 318 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Pre-Loading Application Variables Using the GLOBAL.ASA 319

Application_OnStart

</SCRIPT>

In this example, a two-dimensional array is created that stores the catalog name and the cata-
log starting page. When the ActiveX Server processes this page, the Application_OnStart
event will trigger the Sub Application_OnStart event. When this event is triggered, the
Part
Dim aCatalog(2,2) statement creates and allocates a local space for a two-by-two memory
space. You will increase the performance of your site by eliminating the REDIM statement III
throughout your site. REDIM, although important for dynamically allocating array space, re- Ch
quires extensive resource to dynamically create, reduce, and resize arrays. By explicitly con-
trolling the size of the array, you do not have to use resources to resize the array. This local 16
array can then be converted into an application-level object by using the application keyword
name, as shown below:
Application(“aCatalog”) = aCatalog

This array can be processed from anywhere in the Active Server scripts. Likewise, you can
take the same approach to load the individual software components into an application-level
variable. In particular, you are going to load three components into an array, and each compo-
nent will have four properties that you will store in the array. To accomplish this, you need to
declare a three-by-four array: the first element will store the name of the component; the sec-
ond element will store the description of the element; and the third element will store the unit
price per component. The first item is a placeholder for how many items are ordered. The
creating and setting of the aSoftware array is shown in Listing 16.2.

Listing 16.2 Loading the online catalog in the GLOBAL.ASA file.

<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>

Sub Application_OnStart
Application(“iTimeOut”) = 30

‘=== Load the two available catalogs

Dim aCatalog(2,2)
aCatalog(1,1)=”Software Components”
aCatalog(1,2)=”SoftwareCatalog.asp”
aCatalog(2,1)=”Hardware Components”
aCatalog(2,2)=”HardwareCatalog.asp”
Application(“aCatalog”)=aCatalog

‘=== Load the currently available products

Dim aSoftware (3,4)
aSoftware(1,1)=”Mind-Bender Control”
aSoftware(1,2)=”The MIND-Bender Active Server Design-Time Control reads
➥your mind to generate your Active Server Page Code!”
aSoftware(1,3)= “1999”
aSoftware(1,4)= “0”

continues

20.1389-6 CH16 319 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 320 Chapter 16 Putting It All Together: Building an Online Catalog

Listing 16.2 Continued

aSoftware(2,1)=”Year 2000 Printer Control”
aSoftware(2,2)=”The Year 2000 Printer Control scans all your printer
➥interfaces for possible year 2000 printing problems!”
aSoftware(2,3)=”999"
aSoftware(2,4)=”0"

aSoftware(3,1)=”Fish from Anywhere”

aSoftware(3,2)=”The Fish From Anywhere Control will allow you to fish at
➥your favorite fishing hole from anywhere in the world!”
aSoftware(3,3)=”1549"
aSoftware(3,4)=”0"
Application(“aSoftware”)=aSoftware

End Sub

Application_OnStart

</SCRIPT>

This example uses the GLOBAL.ASA file to load the available products for the online catalog
application. The GLOBAL.ASA file was used to stay within bounds of the ASP objects covered in
this part of the book. Part V of this book, “Database Management with Active Server Pages,”
will cover the details of how to connect your Active Server Pages to database. Chapter 26,
“Putting It All Together: Creating a Dynamic Database Application,” will demonstrate how the
ActiveX Data Object is used to drive the Amazing DataWidget’s WareHouse online catalogs.
Furthermore, Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Applica-
tion,” demonstrates how to build your own database components and manage them with the
Microsoft Transaction Server.

Creating the User’s Shopping Basket Using Session-Level Arrays

and Variables
Now that the application-level variables are taken care of, you need to create session-level
variables to track the activities of individual user sessions. This capability to track individual
user settings is made possible using the Session object. The Session object is covered in
greater detail in Chapter 13, “Managing the User Session: The Session Object.” The Session
object is necessary to create individual shopping baskets to keep track what items the user
wants to buy. To use this approach you will use the Session_OnStart event to create an indi-
vidual copy of the software array stored in the Application object to represent a list of avail-
able products, as shown below:
Sub Session_OnStart
Dim usershoppingbasket
usershoppingbasket = Application(“aSoftware”)
session(“basket”) = usershoppingbasket
End Sub

20.1389-6 CH16 320 10/26/98, 9:27 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Creating and Managing Active Server Pages and Workflow 321

In this case, when a new user enters the application, the Session_OnStart event creates a
session array named basket. This user-specific array can be manipulated at the session level
without having to worry about inadvertently changing the application-level variables. Keep in
mind that the Session object is only created on browsers and networks that support cookies.

Creating and Managing Active Server Pages Part

and Workflow III

At this point, all the data for the application has been loaded into memory. The main screen of Ch
the Amazing DataWidgets WareHouse site uses the application-level array named aCatalog to 16
list the currently available catalog for online viewing, as shown in Figure 16.4.

FIG. 16.4
The main catalog page
uses the application-
level catalog array to
display a list of
available catalogs.

Accessing Application-Level Arrays

To use the application-level array at the page level, create a page variable using the Dim state-
ment. To copy the application-level variable onto the page, simply set the new variables you
create equal to the desired application-level variable, as shown in the following line of code:
Dim aLocalCatalog
aLocalCatalog = Application(“aCatalog”)

Now that a local version of the array has been created, use a For...Next loop to write the
names of the catalogs and the appropriate HREF locations to the Web page, as shown in
Listing 16.3.

20.1389-6 CH16 321 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 322 Chapter 16 Putting It All Together: Building an Online Catalog

Listing 16.3 Writing the information to the Web page.

<TR>
<TD>
<%
Dim aLocalCatalog, iCount
aLocalCatalog = Application(“aCatalog”) ‘=== Create local copy of
➥array
If IsArray(aLocalCatalog) Then ‘=== Test to see if it is
➥an array
For iCount = 1 to Ubound(aLocalCatalog,1)
Response.Write(“<P><Font Face = ‘Arial’><A Href =’” &aLocalCatalog(iCount,2)
&”’> “ &aLocalCatalog(iCount,1) & “</A></Font></P><BR>”)
Next
End If
%>
</TD>
</TR>

In this example, after the local variable is created, the IsArray function verifies that the newly
assigned variable is, in fact, an array. If the variable is an array, the ASP will start iterating
through the different catalogs. The For...Next loop uses the Ubound function to determine the
upper element count of the array. In the example syntax, the statement
Ubound(aLocalCatalog,1) determines the amount of elements in the first dimension of the
aLocalCatalog array. Notice how the Response.Write statement is a mix of Active Server
script and HTML tags. When this loop is processed, the following HTML is produced.
<TR><TD>
<P><Font Face = ‘Arial’><A Href=’SoftwareCatalog.asp’> Software Components</A></
Font></P><BR>
<P><Font Face =’Arial’><A Href=’HardwareCatalog.asp’> Hardware Components</A></
Font></P><BR>
</TD></TR>

This same logic is applied to displaying the list of available software components on the Soft-
ware Catalog page. Listing 16.4 demonstrates how to write the available ActiveX components
that are for sale using an HTML table.

Listing 16.4 Writing the product list to the Web page.

<TR>
<TD colspan=”3">
<%
Dim aSoftwareCatalog, iCompTitle, iComrDesc
aSoftwareCatalog = Application(“aSoftware”)
If IsArray(aSoftwareCatalog) Then
For iCompTitle = 1 to Ubound(aSoftwareCatalog,1)
Response.Write “<TR>”

20.1389-6 CH16 322 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Creating and Managing Active Server Pages and Workflow 323

For iComrDesc = 1 to 2
Response.Write(“<TD Valign=’top’><Font Face=’Arial’>”
&aSoftwareCatalog(iCompTitle,iComrDesc))
Response.Write(“</Font></TD>”)
Next
Response.Write(“<TD valign=’top’ width=’25%’ align=’center’><font
Face=‘Arial’><a href=’additem.asp?product=” &Server.HTMLEncode(aSoftwareCatalog
(iCompTitle,1))& “‘>Order</A></Font></TD></TR>”)
Next Part
End If
%> III
</TD>
</TR> Ch

16
When the ASP is interpreted, the following HTML is generated by the Internet Information
Server to produce the component software catalog listing.
<TR>
<TD colspan=”3">
<TR>
<TD valign=’top’><font Face=‘Arial’>Mind-Bender Control</font></TD>
<TD valign=’top’><font Face=‘Arial’>The MIND-Bender Active Server Design-Time
Control reads your mind to generate your Active Server Page Code!</font></TD>
<TD valign=’top’ width=’25%’ align=’center’><font Face=‘Arial’><a
href=’additem.asp?product=Mind%2DBender+Control’>Order</a></font></TD>
</TR>
<TR>
<TD valign=’top’><font Face=‘Arial’>Year 2000 Printer Control</font></TD><TD
valign=’top’><font Face=‘Arial’>The Year 2000 Printer Control scans all your
printer interfaces for possible year 2000 printing problems!</font></TD>
<TD valign=’top’ width=’25%’ align=’center’><font Face=‘Arial’><a
href=’additem.asp?product=Year+2000+Printer+Control’>Order</a></font></TD>
</TR>
<TR>
<TD valign=’top’><font Face=‘Arial’>Fish from Anywhere</font></TD>
<TD valign=’top’><font Face=‘Arial’>The Fish From Anywhere Control will allow
you to fish at your favorite fishing hole from anywhere in the world!</font></TD>
<TD valign=’top’ width=’25%’ align=’center’><font Face=‘Arial’><a
href=’additem.asp?product=Fish+from+Anywhere’>Order</a></font></TD>
</TR>
</TD>
</TR>

This is the HTML that displays the product listing seen Figure 16.2. Note that because only
HTML was generated by the ASP code list, the catalog results can be displayed in both the
Internet Explorer browser and the Netscape Navigator 3.0 browser, as shown in Figure 16.5.

20.1389-6 CH16 323 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 324 Chapter 16 Putting It All Together: Building an Online Catalog

FIG. 16.5
The software component
catalog listing as seen
with Netscape
Navigator.

Using the Server Object’s Encoding Techniques

Writing the available products names, descriptions, and HREF locations to the Web page was
made possible using the Server object’s URLEncode method. The Server.URLEncode method is
used to apply URL encoding rules to the desired string. URL encoding is used to translate non-
ANSI characters into escape characters that can be properly transported over HTTP. The URL
encoding techniques translate spaces into plus signs (+) and use the ampersand (&) to concat-
enate variables together. This is an important step to include when writing information to an
HTML page. For example, the Server.URLEncode method was used in the following code list-
ing to generate the HREF links:
Response.Write(“<TD valign=’top’ width=’25%’ align=’center’><Font Face=’Arial’><A
Href=’additem.asp?product=” &Server.HTMLEncode(aSoftwareCatalog(iCompTitle,1))&
“‘>Order</A></Font></TD></TR>”)

This code example, extracted from Listing 16.4, would generate the following hyperlink:
<A Href=’additem.asp?product=Fish+from+Anywhere’>Order</A>

When this link is activated, only the entire string Fish from Anywhere is assigned to the vari-
able product. If the Server.URLEncode method were not applied, the following HTML would
be produced:
<A Href=’additem.asp?product=Fish from Anywhere’>Order</A>

If this hyperlink is selected, only the string Fish would be assigned to the product variable.

For more information on the Server object, see Chapter 12, “Controlling the Server with the
Server Object and Using Transaction Processing with the ObjectContext Object.”

20.1389-6 CH16 324 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Creating and Managing Active Server Pages and Workflow 325

Using ASPs as Non-Visual Code Libraries

If you noticed from the previous HTML output, all the hyperlinks for the catalog listings point
to the same Additem.Asp file. The Additem.Asp is a non-visual ASP that accepts variables and
performs an action depending on the value of the variable passed to it. This method enables
you to treat Active Server Pages as code libraries to process application logic and workflow.

Using non-visual ASP files is important to prevent unnecessary application logic from being Part
repeated if the user decides to refresh the browser display. For example, when the user had
ordered a catalog item, the user’s shopping basket needs to be updated with the selected item. III
Traditionally, you may put the update functionality in the calling page that is used to display the Ch
order shopping basket items. However, if the user refreshes the browser while on the shopping
basket page, the page will add another item to the shopping cart. By using non-visible Active
16
Server Pages, you can prevent this type of behavior from occurring. For instance, consider the
situation where the user clicks the hyperlink to order a software component. The Additem.Asp
file, as seen in Listing 16.5, processes the request and adds the appropriate item to the shop-
ping basket array.

Listing 16.5 Using non-visual ASP to perform application and workflow

processing.
<%@ LANGUAGE=”VBSCRIPT” %>
<%
Dim product, iproducts
product = Request.QueryString(“product”)
Dim aLocalbasket
If InStr(product, “Mind”) Then ‘=== Search for Mind-Bender
aLocalbasket =Session(“basket”)
aLocalbasket(1,4)=aLocalbasket(1,4) + 1 ‘=== Adds counter to Mind Bender
Session(“basket”)=aLocalbasket
Elseif InStr(product, “Year”) Then
aLocalbasket =Session(“basket”)
aLocalBasket(2,4)=aLocalBasket(2,4) + 1 ‘=== Adds counter to Year 2000
Session(“basket”)=aLocalbasket
Elseif InStr(product, “Fish”) Then
aLocalbasket =Session(“basket”)
aLocalBasket(3,4)=aLocalBasket(3,4) + 1
Session(“basket”)=aLocalbasket
Else
Response.Redirect “Error.Asp” ‘=== Route user to Error page
End If
Response.Redirect “Order.Asp” ‘=== Route to the order display
page
%>

This code illustrates how to use the Request object to extract information from variables em-
bedded in the URL string. The Request object, as discussed in Chapter 15, “Retrieving User
Information: The Request Object,” is responsible for retrieving information from the user. After
the product variable is extracted using the QueryString property, it is compared to existing

20.1389-6 CH16 325 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 326 Chapter 16 Putting It All Together: Building an Online Catalog

products using the VBScript InStr function, which returns True if the string is found and re-
turns False if the string could not be found in the argument.

After an existing product is found, the user’s shopping basket, which is located in
Session(‘Basket’) variable, is copied to a page-level variable. Now the local page array can be
manipulated as needed. In this situation, you are updating the fourth array element, which is
used to store the amount of items ordered of the product name in the first array element. After
the local array has been updated, it is reassigned back to the session level, where other Active
Server Pages can access it. After the user’s shopping basket has been updated, the
Response.Redirect method routes the browser to the Order.Asp page. The Response object is
responsible for sending output to the browser. The Redirect method is used to push the
browser to another page. For more information regarding the Response object, see Chapter 14,
“Speaking to the Web Client: The Response Collection.”

You can see the powerful effect that non-visual Active Server Pages and the
Response.Redirect method can add to the programming functionality of your Active Server
Page applications. This functionality enables you to gain some control over the nuances of
using the browser’s refresh button. Because the update shopping basket functionality is not
attached to a specific visual Active Server Page, the use of the browser’s navigation buttons is
eliminated from processing application logic. For example, the SoftwareCatalog.asp is only
responsible for displaying available software items. The Order.Asp page is only used to display
information in the user’s shopping basket. Therefore, using the browser’s navigate forward,
back, and refresh buttons between these pages will not corrupt the workflow of the application.

The Order.Asp file follows the same coding style of the CatalogSoftware.Asp page. The
Order.Asp page is used to display the currently selected items in the user’s shopping basket
and enable the user to continue shopping, pay for the order, or cancel the order. The user’s
selections are sent to the same ASP file for processing.

Examining the Shopping Basket

To find out whether the user placed any items placed in his shopping basket, his shopping
basket is searched for all items that have an item count greater than zero. If the item count is
greater than zero, the total is calculated by multiplying the unit count by the unit price, as
shown in Listing 16.6.

Listing 16.6 Examining and displaying the user’s shopping basket.

<%
Dim iCounter, itemsordered, aBasket, name, description, unitprice, amtordered,
subtotal
ABasket = Session(“basket”)
For iCounter = 1 to ubound(aBasket,1) ‘=== Iterate through user basket
Itemsordered = aBasket(iCounter,4) ‘=== Find if any items are ordered
If itemsordered > 0 Then ‘=== If any items are ordered
name = aBasket(iCounter,1) ‘=== Name of item
description = aBasket(iCounter,2) ‘=== Description of item

20.1389-6 CH16 326 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Creating and Managing Active Server Pages and Workflow 327

unitprice = aBasket(iCounter,3) ‘=== Unit Price

subtotal = FormatCurrency(subtotal+unitprice*itemsordered) ‘===
Calculate subtotal
%>
<TR>
<TD Width=”15%” Bgcolor=”#F7EFDE” valign=”top”><Font Size=”2" Face=”Arial”><%
Response.Write name %></Font></TD>
<TD width=”50%” bgcolor=”#F7EFDE” valign=”top”><Font Size=”2" Face=”Arial”><%
Response.Write description %></font></TD> Part
<TD Align=”center” bgcolor=”#F7EFDE” valign=”top”><font size=”2"
Face=”Arial”><% Response.Write itemsordered %></font></TD> III
<TD align=”center” width=”10%” bgcolor=”#F7EFDE” Valign=”top”><Font
size=”2"face=”Arial”> Ch
<% Response.Write FormatCurrency(unitprice) %></font></TD>
<TD align=”right” width=”10%” bgcolor=”#F7EFDE” Valign=”top”><Font size=”2"
16
face=”Arial”>
<% Response.Write FormatCurrency(unitprice*itemsordered) %></Font></TD>
</TR>
<%
End If
Next
%>

This ASP code is used to display what items the user has ordered, as shown in Figure 16.6.
Notice the use of the VBScript FormatCurrency function. The FormatCurrency function is used
to return a formatted currency expression. The returned value uses the currency symbols that
are defined in the system control panel. These formatting features can be covered in greater
depth in Chapter 7, “Datatype Conversion Features.”

FIG. 16.6
The user-specific
shopping basket is
maintained by an array
variable stored by the
Session object.

20.1389-6 CH16 327 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 328 Chapter 16 Putting It All Together: Building an Online Catalog

Using the Request Object in a Multi-Button Form

After the user views the contents of his shopping basket, he has the option to continue shop-
ping, cancel the order, or proceed with the order. To make this functionality possible, the same
approach you used to pass arguments via the QueryString in the hyperlinks to a non-visual
code page is used. Listing 16.7 illustrates using three submit buttons on the bottom of the
order form to call the GoTo.Asp file.

Listing 16.7 Posting multiple values using the same HTML form.
<FORM NAME=”frmbutton” Action=”GoTo.ASP?” Method=”Post”>
<TR>
<TD align=”center” colspan=”5" width=”100%”>
<INPUT TYPE=”submit” NAME=”btngoto” VALUE=”Continue Shopping”>
<INPUT TYPE=”submit” NAME=”btngoto” VALUE=”Check Out”>
<INPUT TYPE=”submit” NAME=”btngoto” VALUE=”Cancel Order”>
</TD>
</TR>
</FORM>

When any of the navigation buttons are pressed, as shown in Figure 16.6, the button’s value is
submitted to the GoTo.ASP file. The ASP then processes the value by using the Request object
to determine what button is pressed and using the Select Case statement to decide what
action to take (see Listing 16.8).

Listing 16.8 Accepting and processing multiple values from an HTML form.
<%@ LANGUAGE=”VBSCRIPT” %>
<% OPTION EXPLICIT
Dim gotopage
Select Case Request(“btngoto”)
Case “Continue Shopping”
gotopage=”softwarecatalog.asp”
Case “Cancel Order”
Session.Abandon
gotopage=”softwarecatalog.asp”
Case “Check Out”
gotopage=”customer.asp”
Case Else
gotopage=”softwarecatalog.asp”
End Select
Response.Redirect gotopage
%>

In this code, the browser is redirected to the software catalog page if the user decides to shop
for more items. If the user decides to pay for the items in the shopping cart, the browser is
routed to the customer information page. If the user decides to cancel his order, the

20.1389-6 CH16 328 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 Creating and Managing Active Server Pages and Workflow 329

Session.Abandon method is called and the user is routed to return shopping at the software
catalog page. The Session.Abandon method destroys all Session objects for that user session,
effectively erasing all information about that session. The Session objects are destroyed after
the page is finished processing. However, because the user is routed to another page within the
web, a new Session object is created and the user is treated as if they entered the site for the
first time.

If the user decides to cancel the order, the Session.Abandon method destroys that user’s Part
Session object. As a result, all information about the user is destroyed. For example, if the same III
user decides to cancel the order, the Session.Abandon method purges the order items. When
Ch
the user orders a new item, the item is placed in the shopping basket, as shown in Figure 16.7.
16
FIG. 16.7
The Session.
Abandon method
destroys all settings for
the specific user and
creates an empty
shopping basket for
new items.

Extracting Information Using the Request Object

If the user is ready to “check out,” the browser is directed to the customer page. The customer
page enables the user to enter user profile and shipping information. After the user has cor-
rectly entered the required information, the information is posted to the SaveUser.ASP page.
This non-visual ASP is used to read the values entered in the form and create session variables.
Typically, and as shown in later chapters, this would be the location of rules-based, server-side
processing such as address verification, credit card validation, or inventory processing. In
Listing 16.9, the SaveUser.Asp file creates session variables by extracting values from HTML
form fields.

20.1389-6 CH16 329 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 330 Chapter 16 Putting It All Together: Building an Online Catalog

Listing 16.9 Extracting information from an HTML form and storing the user
information.
<%@ LANGUAGE=”VBSCRIPT” %>
<%
Server.ScriptTimeout = Application(“iTimeOut”)
Session(“UserFirstName”)=Request.Form(“txtfirstname”)
Session(“UserLastName”)=Request.Form(“txtlastname”)
Session(“UserCompany”)=Request.Form(“txtCompany”)
Session(“Useremail”)=Request.Form(“txtemail”)
Session(“UserAddress1”)=Request.Form(“txtaddress1”)
Session(“UserAddress2”)=Request.Form(“txtaddress2”)
Session(“UserCity”)=Request.Form(“txtcity”)
Session(“UserState”)=Request.Form(“txtstate”)
Session(“UserPostal”)=Request.Form(“txtpostal”)
Session(“UserCountry”)=Request.Form(“txtcountry”)
Response.Redirect “summary.asp”
%>

In previous examples, the Request object was used to extract variables from the URL
QueryString and to determine the value of a button when it was pushed. The same concept is
used to gather HTML form data using the Request object. The Request.Form property identi-
fies the name of the individual HTML form element to retrieve the information from. Keep in
mind that the Request object does not extract information directly from the HTML form, but
parses the information sent to the Web server via an HTTP Post. After the processing is com-
plete, the browser is shifted to the summary page.

The summary page, as shown in Figure 16.3, is used as a virtual receipt to confirm to the user
the amount of items ordered, the total cost, and to whom and where the items are being sent.
The exact same logic that is applied to display the items in the shopping cart is used to display
information in the summary page; the only difference is the HTML formatting.

From Here…
In this chapter, you have seen the integration of Active Server Pages built-in objects used to
build this simple online ordering system. This application demonstrated how to extract infor-
mation from hyperlinks and forms. The Application object is used to pre-load data that can be
accessed from any user within the site. For information that is relatively static, pre-loading this
information once can save a considerable amount of resources from retrieving the same infor-
mation on a frequent basis. The Session object provides the core functionality of the online
ordering system by enabling a user-specific virtual shopping basket. The shopping basket
consists of transferring and maintaining application-, session-, and page-level arrays and vari-
ables. The Server object is used to apply encoding techniques to string information, reset the
information about the current user, and set the ScriptTimeOut value for pages. The Response
object is used to control output sent to the browser by writing information to the ASP with the
Write method. The Response object also plays an important role by using the Redirect method
to route the user to the desired page. The Request Object is used for retrieving information
from the user. For related information, see the following chapters:

20.1389-6 CH16 330 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 From Here… 331

■ Chapter 18, “Creating Your Own Active Server Components,” walks you through
the tools and examples on how to create tailored functionality via custom created
components.
■ Chapter 20, “Database Access with ADO,” teaches you how to use the Data Access
Component to connect to any ODBC-compliant database.
■ Chapter 26, “Putting It All Together: Creating a Dynamic Database Application,”
illustrates how to use the Application object to store information extracted from a Part
database to provide database interaction for the Amazing DataWidget WareHouse. III
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,” Ch
demonstrates how to use the various features of Internet Information Server to help
manage your sites and applications. 16
■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,”
demonstrates how to build ActiveX database components managed by the Microsoft
Transaction Server to create an n-tier distributed catalog application for the Amazing
DataWidget WareHouse site.

20.1389-6 CH16 331 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 332 Chapter 16 Putting It All Together: Building an Online Catalog

20.1389-6 CH16 332 10/26/98, 9:28 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing CH16 11/13/97 lp#3
 P A R T

IV
Active Server Components
17 Implementing Existing Server-Side Components 335

18 Creating Your Own Active Server Components 383

19 Putting It All Together: Creating an Online Registration

System 409

21.1389-6 Part IV 333 10/26/98, 9:29 AM

Brands SEU Active Server Pages #1389-6 Part IV SRing 10/28/97 lp#1
21.1389-6 Part IV 334 10/26/98, 9:29 AM

Brands SEU Active Server Pages #1389-6 Part IV SRing 10/28/97 lp#1
 17
335

C H A P T E R

The Browser Capability and

Implementing Existing Advertisement Rotator
Components

Server-Side Learn how to use the Browser

Capability Component to determine
Part
requesting browsers’ characteristics

Components and how to use the Advertisement

Rotator Component to dynamically
IV
Ch
generate scrolling advertisement
bars. 17
The Counter Component and
Page Counter Component
Investigate how the Counter Com-
One of the most powerful features of the Internet Informa- ponent is used to create incremental
variables within your ASPs and how
tion Server and Active Server Pages is the capability to tap
to use the Page Counter Component
into COM-based components. These server-side objects
to track the number of times a page
greatly extend the functionality of the Web server by is requested.
providing a virtually limitless warehouse of programming
functionality. Furthermore, the server-side components The Content Rotator and Con-
help transition traditional application developers into tent Linking Components
creating Web applications by enabling ActiveX compo- Utilize the Content Rotator to
nents to be created in their native COM-based develop- change the content on a page every
ment environments, such as Visual C, C++, Visual Basic, time the page is requested or re-
Delphi, and PowerBuilder. This chapter demonstrates the freshed and learn how to use the
various server-side components that are shipped and Content Linking Component to
create a systematic Web page naviga-
installed with the Internet Information Server and Active
tion technique for pages within a site.
Server Pages. ■
The Active Messaging/SMTP
Component and Tools
Component
Discover how to tap in to the power
of Microsoft Exchange and other
e-mail–based systems using the
Active Messaging/SMTP Compo-
nent and how the Tools Component
brings additional functionality to
server-side processing.

22.1389-6 CH17 335 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 336 Chapter 17 Implementing Existing Server-Side Components

The Ad Rotator Component

The Ad Rotator Component is designed to display and rotate different advertisements on Web
pages. This rotating billboard approach is tailored to selling advertising space on commercial
Web sites. The Rotator Component can display a new advertisement every time based on the
frequency specified in a schedule file. The displayed banner can be programmed to route the
browser to a destination site when the user clicks the banner, or can be used simply to display the
banner without forwarding to another site. For example, the Ad Rotator Component is respon-
sible for managing the banner display mechanisms for an ASP page, as shown in Figure 17.1.

FIG. 17.1
The Ad Rotator displays
rotating banner images
based on a scheduling
file.

When the site is revisited or refreshed, the Ad Rotator can display another banner based on the
frequency set in the scheduling file, as shown in Figure 17.2.

Let’s take a closer look at how the Ad Rotator uses methods and properties to display the rotat-
ing banners.

The GetAdvertisement Method

To activate the Ad Rotator Component, the Server.CreateObject method is used to instantiate
an instance of the Ad Rotator Component:
Set ad = Server.CreateObject(“MSWC.AdRotator”)

In this situation, the Set command is used in conjunction with the Server.CreateObject tag to
create a reference to the MSWC.AdRotator object. The Ad Rotator Component ships with the
Internet Information Server and is automatically registered on the Web server when IIS is
installed. The Ad Rotator Component is created from an instantiated instance of the Adrot.dll.

22.1389-6 CH17 336 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Ad Rotator Component 337

Typically, the Adrot.dll file is installed in the \WINNT\System32\InetSrv directory on the NT

Web server. The Ad Rotator has one method: the GetAdvertisement method. The
GetAdvertisement method manages and retrieves information from the scheduling file. When
the GetAdvertisement method is called from a new page request or a page refresh request, the
GetAdvertisement method loops through the schedule file to display the next advertising
banner. To use the GetAdvertisement method, use the following syntax:
GetAdvertisement(ScheduleFilePath)

where ScheduleFilePath is the virtual path to the schedule file.

FIG. 17.2
When the page is
revisited or refreshed,
Part
the Ad Rotator
Component displays a IV
new banner image.
Ch

17

The Ad Rotator Component Required Files

As mentioned in the GetAdvertisement method, the Ad Rotator references other files to com-
plete its tasks. The Ad Rotator Component requires two other files to complete its functionality,
as shown in Table 17.1.

Table 17.1 The Ad Rotator File References

Name Description
Rotation schedule file Text file that manages the frequency or schedule of the
displayed banner.
Redirection file Redirects the requesting browser to the destination site for a
given banner.

22.1389-6 CH17 337 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 338 Chapter 17 Implementing Existing Server-Side Components

The Rotator Schedule File

The Rotator schedule file controls the frequency and the properties of the Ad Rotator Compo-
nent. The schedule file determines the frequency or schedule of the advertisements, manages
the properties of the advertisements, and is also used to set the various properties of the ban-
ner object. These properties are shown in Table 17.2.

Table 17.2 The Schedule File Properties

Name Description Scope
Redirect (URL) Sets the default All images in rotator
location to route schedule.
the browser (optional).
Width (imgWidth) Sets the default All images in rotator
pixel width of the banner schedule.
image (optional).
Height (imgHeight) Sets the default All images in rotator
pixel height of the banner schedule.
image (optional).
Border (imgBorder) Sets the default All images in rotator
border width of the schedule.
banner image (optional).
* Separates the default
values (shown above)
from the individual
banner detail specifications
in the schedule file.
AdURL Sets the location to Individual image in
retrieve the image file. rotator schedule.
AdHomePageURL Sets the location of Individual image in rotator
the advertiser home page. schedule.
Text The textual display Individual image in
for text-based browsers rotator schedule.
or if displaying images
are deactivated on the
browser.
Impressions Sets the percentage of Individual image in
time that the image rotator schedule.
will be displayed.

For example, Listing 17.1 demonstrates a sample scheduling file.

22.1389-6 CH17 338 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Ad Rotator Component 339

Listing 17.1 A sample rotation schedule file.

---DecemberInfo.TXT---
REDIRECT /scripts/adredir.asp ‘=== Sets redirect
➥location for logging
WIDTH 400 ‘=== Sets the default
➥width
HEIGHT 50 ‘=== Sets the default
➥height
BORDER 0 ‘=== Sets the default
➥border
* ‘=== Separator
http://myDataWidgets/DecemberSales/mywidget.gif '=== First Advertisement
➥banner image
http://www.myDataWidgets.com/ '=== Home page location
Free Demo of the new Data Widget Deluxe for ODBC ‘=== Textual Content
20 ‘=== Frequency displayed
Part
http://DataCom/DataComObject.gif
➥Advertisement banner
‘=== Second IV
- ‘=== No Home page Ch
➥location
It’s Coming, DataCom Inc, the new way of doing business ‘=== Textual Content 17
80 ‘=== Frequency Displayed

This scheduling file, named DecemberInfo.text, is used to display two advertisements. The first
advertisement points to a company at www.myDataWidgets.com, and the advertisement is to be
displayed 20 percent of the time. The second advertisement is used to display an image from the
DataCom virtual directory. This image, however, does not have a home page location, so a dash
(-) character is used. This image is activated 80 percent of the time when this site is refreshed.

The Redirection File

The redirection file is responsible for routing the browser to the destination site, as determined
in the scheduling file. This redirecting approach is not directly needed to shift the user to the
advertiser site when the banner is clicked. Simple HTTP can accomplish this. However, the
redirection file is needed to keep track of the number of hits generated from an Ad Rotation
banner. For example, in the aforementioned scheduling file, the redirection file /scripts/
adredir.asp is designated in the first line of the DecemberInfo.txt file. This redirection file is
not only used to push the requesting client to the requested advertisement, but also to store a
variety of information about the requesting client, such as the browser type or IP address. The
following code demonstrates using the Response.Redirect method to guide the browser to the
page specified in the banner scheduling file.
--- adredir.asp ---
<% Response.Redirect(Request.QueryString(“myAdUrl”)) %>

At this point, you can use various Web analysis logging tools or the Page Counter Component
to determine the number of hits the redirection page received. The Page Counter Component
is another built-in component that is discussed later in this chapter. For more information
about the Response object, refer to Chapter 14, “Speaking to the Web Client: The Response
Collection.”

22.1389-6 CH17 339 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 340 Chapter 17 Implementing Existing Server-Side Components

The Ad Rotator Properties

The Ad Rotator banner has three properties that control the Ad Rotator Component, as dis-
played in Table 17.3.

Table 17.3 The Ad Rotator Properties

Name Description
Border Sets the width of the border around the advertisement.
Clickable Sets whether the rotator banner is a hyperlink or simply an image.
TargetFrame Sets the name of the target frame to display the banner.

The properties of the Ad Rotator all follow the same syntax of requiring an object to act upon:
Object.Property

Table 17.4 illustrates a quick guide to the syntax for the Ad Rotator properties.

Table 17.4 The Ad Rotator Property Syntax

Property Syntax
Border Border (size)

Clickable Clickable (value)

TargetFrame TargetFrame (frame)

The Border property is used to set the size of the border around the banner image. To set the
border size, use the following syntax:
Border (size)

where size is an integer specifying the thickness of the border. To display a banner with no
border, similar to an image with no border, set the border size to 0, as shown in the following
example:
ad.BorderSize(0)

In this code example, the border size of the Ad Rotator object named ad is set to 0. The result
of setting the border width is shown in the following figures.

Figure 17.3 displays a banner with border size set to 2.

The Clickable property determines whether the displayed image acts as a hyperlink or simply
acts as a random image generator. To set the Clickable property, use the following syntax:
Clickable (value)

where value is True or False. The values of the Clickable property are shown in Table 17.5.

22.1389-6 CH17 340 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Ad Rotator Component 341

FIG. 17.3
Adjusting the border
size of the banner to
size of 2.

Part
IV
Ch

17

Table 17.5 The Ad Rotator’s Property Values

Value Description
True Sets the displayed banner to act as a hyperlink.
False Disables the hyperlink property of the banner.

If the value of the Clickable property is set to True, the display banner acts like a hyperlink. If
the value is set to False, the Ad Rotator simply displays the banner and the hyperlink effect of
the banner is disabled. To use this display-only banner effect, use the following code:
<% Set Ad = Server.CreateObject(“MSWC.AdRotator”) ‘=== Create Ad Rotator
Object %>
<% Ad.Clickable(False) ‘=== Disables the
hyperlink property of the banner %>
<%= Ad.GetAdvertisement(“/ASPSamp/Samples/adrot.txt”) ‘=== Display a new banner
%>

This code would not just present rotating images on ASP pages. Because these images are not
hyperlinks anymore, the default hyperlink border color will not surround the image and the
standard image border is used. The TargetFrame property is used to indicate the name of the
frame in which the Ad Rotator object is to be displayed. The TargetFrame property behaves
similarly to the HTML TARGET tag, where you can name the destination location within a
browser window for the results to be displayed. To use the TargetFrame property, use the
following syntax:
TargetFrame (frame)

22.1389-6 CH17 341 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 342 Chapter 17 Implementing Existing Server-Side Components

where frame is the name of an HTML frame. The name property supports all native HTML
Frame tags, including _TOP, _NEW, _CHILD, _SELF, and _PARENT.

For example, to display the results of the Ad Rotator Component in a frame named AdWindow,
use the following syntax:
<%
Set myAdObj = Server.CreateObject(“MSWC.AdRotator”) ‘=== Create AdRotator
Object
FrName = ”AdWindow” ‘=== Set Frame Name
myAdObj.TargetFrame (FrName) ‘=== Set target frame
%>
<%= myAdObj.GetAdvertisement(“/June/specials.txt”) ‘=== write the Ad to page
%>

The Browser Capabilities Component

The Browser Capabilities Component retrieves browser information from the requesting
browser. The Browser Capabilities Component retrieves information from the USER Agent
HTTP header string to determine the type and version number of the browser that is requesting
the page. The Browser Capabilities Component then references an external file that stores
additional information about the requesting browser.

To use the Browser Capabilities Component, use the following syntax:

Set BrowserType = Server.CreateObject(“MSWC.BrowserType”)

where BrowserType is the reference to the instantiated MSWC.BrowserType object from the
BROWSCAP.DLL. Usually the BROWSCAP.DLL is located in the \WINNT\System32\InetSrv of the NT
Web server.

The BROWSCAP.INI File

The Browser Capabilities Component uses a text file named BROWSCAP.INI to act as a reference
card to store information about the capabilities of various browsers. The Browser Capabilities
Component retrieves the User Agent Header string from the requesting browser and uses that
value to look up additional features about the requesting browser in the BROWSCAP.INI file. If a
match cannot be found, the default browser settings from the BROWSCAP.INI are used.

The BROWSCAP.INI file must be located in the same directory as the BROWSCAP.DLL file. The
BROWSCAP.INI file is initially installed to the Web server when IIS is installed. The
BROWSCAP.INI file can be created, edited, and maintained via any ANSI editor.

The BROWSCAP.INI structure is shown in Table 17.6.

22.1389-6 CH17 342 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Browser Capabilities Component 343

Table 17.6 The BROWSCAP.INI File Structure

Symbol Description
;; Comments.
[ header info] Sets a key field defining a HTTP User Agent Header request.
property=value Sets the various properties of the specific browser type.
[Default Browser Sets the default property settings.
Capability Settings]

Chapter 15, “Retrieving User Information: The Request Object,” demonstrated how to use the
Server variable to extract information from the HTTP header string. For example, the follow-
ing Server variables extract browser information. Part

<% myBrowserType = Request.ServerVariables(“HTTP_USER_AGENT”) %>

IV
Ch
If an America Online browser is making a request on your site, the variable myBrowserType
would return the following string information: 17
Mozilla/2.0 (compatible; MSIE 3.0; AOL; Windows 95)

To determine the additional properties of this browser that are not located within the HTTP
header string, reference the header string in the BROWSCAP.INI file. The header string must be
included in the hard brackets ([]) within the BROWSCAP.INI file, as shown below:
;;AOL Browser ‘=== Comment to help identify AOL browser
[Mozilla/2.0 (compatible; MSIE 3.0; AOL; Windows 95)]
parent=IE 3.0
version=3.01
minorver=01
platform=Win95

Now these additional features can be referenced by the Browser Capabilities Component to
help identify characteristics of the specific browser type.

The interesting thing about the identification fields in the BROWSCAP.INI file is that you can
include wildcard (*) placeholders to group common types of browsers. For example, to group
all Internet Explorer-compatible browsers running on Windows 95, use the following code to
identify the browser identification field in the BROWSCAP.INI file:
;;IE 3.01
[Mozilla/2.0 (compatible; MSIE 3.0;* Windows 95)]
parent = IE 3.0
version = 3.01
minorver = 01
platform = Win95

22.1389-6 CH17 343 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 344 Chapter 17 Implementing Existing Server-Side Components

The addition of the wildcard in the identification field now groups the following browsers together:
[Mozilla/2.0 (compatible; MSIE 3.0; Windows 95)]
[Mozilla/2.0 (compatible; MSIE 3.0; AK; Windows 95)]
[Mozilla/2.0 (compatible; MSIE 3.0; SK; Windows 95)]
[Mozilla/2.0 (compatible; MSIE 3.0; AOL; Windows 95)]

You can further expand the different browser types within the BROWSCAP.INI file by using the
parent tag within the browser identification tags. The parent tag enables inheritance from
previous browser versions, which enables previous browser properties to be built on each
other without having to maintain full property lists for every version of a browser that is re-
leased. If existing property values are detected, the new values overwrite existing values and
new properties are appended to the existing property list.

Listing 17.2 A sample BROWSCAP.INI file.

;; IE 3.0 ‘=== Internet Explorer 3.0 version
[IE 3.0]
browser = IE
Version = 3.0
Majorver = 3
Minorver = 0
Frames = TRUE
Vbscript = TRUE

;; IE 3.01 ‘=== Internet Explorer 3.1 version

[Mozilla/2.0 (compatible; MSIE 3.01*; Windows 95)]
parent = IE 3.0 ‘=== Inherit value from the IE 3.0
version = 3.01 ‘=== Overwrites version info
minorver = 01 ‘=== Overwrites minor version info
platform = Win95 ‘=== Appends the platform property

; Default Browser
[Default Browser Capability Settings]
browser = Default
frames = FALSE
tables = TRUE
cookies = FALSE
backgroundsounds = FALSE
vbscript = FALSE
javascript = FALSE

In Listing 17.2, if an IE 3.01-compatible browser is detected by the Browser Capabilities Com-

ponent, the parent tag of the IE 3.01 header refers to the IE3.0 version defined earlier in the
INI file. In this situation, the version is overwritten now to 3.01, the minorver tag is now over-
written to 01, and the platform property of Win95 is appended to the property list.

22.1389-6 CH17 344 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Browser Capabilities Component 345

The Browser Capabilities Component Properties

Once the Browser Capabilities Component has been created, information about the browser
can be determined by examining its properties. By using the properties you can ensure that
the requesting clients of your application and site meet specific criteria. The properties of the
Browser Capabilities Component are illustrated in Table 17.7.

Table 17.7 The Browser Capabilities Component Properties

Property Description
ActiveXControls Specifies whether the browser supports ActiveX controls.
Backgroundsounds Specifies whether the browser supports background sounds.
Beta Specifies whether the browser is beta software. Part

Browser Specifies the name of the browser.

IV
Ch
Cookies Specifies whether the browser supports cookies.
Frames Specifies whether the browser supports frames.
17
JScript Specifies whether the browser supports JScript.
Platform Specifies the platform that the browser runs on.
Tables Specifies whether the browser supports tables.
VBScript Specifies whether the browser supports VBScript.
Version Specifies the version number of the browser.

After the Browser Capabilities Component has been created, you can access these properties
to ensure a specific browser functionality level. For example, Listing 17.3 displays information
about a requesting browser:

Listing 17.3 Detecting and displaying information about the browser

capabilities.
<%
Dim myBrowser
Set myBrowser = Server.CreateObject(“MSWC.BrowserType”) %>
<table border = 1 width = 100% CellPadding = 2 CellSpacing = 2>
<tr><td>Browser Type</td><td> <%= myBrowser.browser %> ‘=== Writes
Browser Name
<tr><td>Version</td><td> <%= myBrowser.version %> </td></TR> ‘===
display Version
<tr>
<td>Frames</td>

continues

22.1389-6 CH17 345 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 346 Chapter 17 Implementing Existing Server-Side Components

Listing 17.3 Continued

<td>
<%
if (myBrowser.frames = TRUE) then ‘=== Test for Frames
Response.Write(“TRUE”)
else
Response.Write(“False”)
end if
%>
</td></TR>
<tr><td>VBScript</td>
<td>
<%
if (myBrowser.vbscript = TRUE) then ‘=== Test for VBScript Support
Response.Write(“TRUE”)
else
Response.Write(“False”)
end if
%>
</td></TR>
<tr>
<td>ActiveX Controls</td> ‘=== Test for ActiveX support
<td>
<%
if (myBrowser.ActiveXControls = TRUE) then
Response.Write(“TRUE”)
else
Response.Write(“False”)
end if
%>
</td></TR>
</table>

Knowing information about the browser will help you control how your application behaves in
various browsers. In real-world applications, you will need to build off the Browser Capabilities
Component to test for capability and display issues. For example, the following code uses both
the Browser Capabilities Component and the Request object to ensure that requesting brows-
ers are either Internet Explorer 3.0 or Netscape Navigator 3.0 compatible.
<%
dim isBrowserOK
Set browserobj = Server.CreateObject(“MSWC.BrowserType”) ‘=== Create
➥Browser Component
if Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”) then ‘=== Check
➥for Browser info using ‘=== the Request Object
if browserobj.MajorVer = 3 then ‘=== Uses Browser Object to get
➥version number
isBrowserOK = true
else
isBrowserOK = false
end if

22.1389-6 CH17 346 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Database Access Component 347

else
if Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”Mozilla”) then ‘===
Check for Netscape
isBrowserOK = true
end if
end if
%>

The previous example creates a variable named isBrowserOK to determine whether the re-
questing browser is either IE 3.0 or Netscape. If the requesting browser is either Netscape or
Internet Explorer 3.0, the isBrowserOK variable is set to True. Now you can use this value to
route the user to the appropriate location, as shown in the following code:
<% If browserOK Then %>
‘=== Proceed to next step in client validation
<% Else %>
Part
‘=== Warn user to upgrade browser
<% End If %> IV
Ch

The Database Access Component 17

Active Server Pages can access databases through the use of the Active Data Object (ADO).
ADO is an optimized database connection based off the Data Access Object (DAO) and Remote
Data Object (RDO) used in traditional application development tools such as Visual Basic. The
ADO provides critical features that are needed to transform existing client-server applications
into Web-enabled applications. The ADO object supports various cursor types, batch updating,
extended recordset management techniques, and advanced stored procedure support.

The details of the ADO object are fully described in Part V, “Database Management with Active
Server Pages.” Those seven chapters are dedicated to demonstrating how to manage your
database connection in Active Server Pages.

The ADODB object is used to add a database connection to your ASP application. To create the
ADO connection, use the following syntax:
Set myDataConn = Server.CreateObject(“ADODB.Recordset”)

where myDataConn is the instantiated reference to the ADODB object. Once the database connec-
tion object exists, a database connection can be made by using the Open method. The Open
method can accept query strings or stored procedure names. For example, the following code
uses the recordset object created in the previous example to store the result from the SQL
statement Select * From CustomerInfo:
myDataConn.Open “SELECT * FROM CustomerInfo”, “DATABASE = pubs;UID = sa;PWD=;DSN
= Publishers”

Now that the results have been stored in the myDataConn recordset object, the information can
be written to the page by using a Do...While loop and checking for the beginning of file (BOF)
and end of file (EOF) markers. If both BOF and EOF values are True, no records are returned.
The following code example demonstrates this iterative technique to write information re-
turned from the database onto an HTML page:

22.1389-6 CH17 347 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 348 Chapter 17 Implementing Existing Server-Side Components

<%
Set myDataConn = CreateObject(“ADODB.Recordset”) ‘=== Creation Connection
Object
myDataConn.Open “SELECT * CustomerInfo”, “DATABASE = pubs;UID = sa;PWD=;DSN =
Publishers”
If myDataConn.EOF and myDataConn.BOF Then ‘=== Check to see if data is
returned
‘=== No Results returned
Else
Do While Not myDataConn.EOF ‘=== Iterate through result set
Response.Write (myDataConn(“CustomerID”) & “<BR>”) ‘=== write the
customerID field ‘ ===
and an HTML break to the page
Loop ‘=== End of While Loop
End If
%>

The Active Messaging/SMTP Component

The Active Messaging Component is an exciting component that opens the object model of
various Simple Mail Transfer Protocol (SMTP) messaging systems. These messaging systems
enable you to directly tap into the power of Internet-based mail and groupware software solu-
tions such as Microsoft Exchange. Now you can not only create and manage e-mail messages,
but also other resources such as discussion databases. The Active Messaging Component uses
the Microsoft Exchange Server to store and access mail-based information. In this way, Ex-
change can be used as a gateway to other SMTP interfaces, as well as expose its calendar and
scheduling features such as displayed Microsoft Outlook to any HTML client.

As with other programming models, the Messaging object model follows a hierarchical format
that is used to manipulate its programmable objects. The Messaging/SMTP Messaging object
model is shown in Figure 17.4.

FIG. 17.4 Application

The Active Messaging
object model exposes Sessions *
the functionality found
in Microsoft Exchange Folder (Inbox or Outbox)
and Outlook.
Items *
Mail Items

Attachments *
Attachment

Recipients *
Recipient

22.1389-6 CH17 348 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Active Messaging/SMTP Component 349

The SMTP interface enables you to manage your messaging system with relative ease by in-
stantiating the ACTMSG32.DLL file. Similar to the other installed server components, the
ACTMSG32.DLL file is typically installed in the \WINNT\System32\InetSrv directory on the NT
Web server.

Although, I could dedicate a several chapters to discussing the details of the SMTP interface,
this section only will briefly highlight the capabilities of the Active Messaging Component. For
example, Listing 17.4 sends a mail message to [email protected] with the subject line
of MAPI Subject Line and a textual message of This message was generated via the Ac-
tive Messaging Component.

Listing 17.4 Send e-mail via ASP and the Active Messaging Component.
<% Part
Set myMAPISession = Server.CreateObject(“MAPI.Session”) ‘=== Creates
➥MAPI Object IV
myServerProfile = “myServer” ‘=== Creates
Ch
➥Login String
myMailboxProfile = “myUserProfile”
myMailProfile = myServerProfile + chr(10) + myMailboxProfile
17
myMAPISession.Logon “”, “”, False, True, 0, True, myMailProfile ‘=== Session
➥Logon
Set myMessageObj = myMAPISession.Outbox.Messages.Add ‘=== Open New
➥OutBox message
Set myDestinationObj = myMessageObj.Recipients.Add ‘=== Address
➥new message
myDestinationObj.name = (“[email protected]”) ‘=== Send to
➥this user
myDestinationObj.Resolve ‘=== Resolve
➥user name
myMessageObj.subject = “MAPI Subject Line” ‘=== Set the
➥Subject Line
strMsgText = “This message was generated ”
strMsgText = strMsgText & “via the Active Messaging/SMTP component”
myMessageObj.Text = strMsgTxt ‘=== Assigns
➥body text
myMessageObj.Send showDialog = false ‘=== Send the
➥message
set myDestinationObj = Nothing ‘=== Clean up
➥objects
set myMessageObj = Nothing
set myMAPISession = Nothing
%>

Once the previous code is run, Figure 17.5 verifies that the message was sent and received by
checking the Inbox of the Microsoft Exchange HTML/ASP interface.

22.1389-6 CH17 349 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 350 Chapter 17 Implementing Existing Server-Side Components

FIG. 17.5
The Active Server Code
generated this mail
message as viewed by
the Exchange HTML/ASP
interface.

To check on the contents of the message, select the hyperlink under the From categories and
view the detail of the message, as shown in Figure 17.6.

FIG. 17.6
The Exchange HTML
interface displays the
message generated
from the ASP code.

You can see where this automatic messaging system can be very useful and very straight-
forward to implement. For example, a remote sales-force automation application can benefit
from the automatic mailer system. The automatic e-mail notification can be used to trigger an
action item at headquarters when a remote user updates information.

The Active Messaging HTML (AMHTML) object model is a very powerful interface respon-
sible for exposing the native functionality found in Outlook into an HTML format.

22.1389-6 CH17 350 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Active Messaging/SMTP Component 351

In order to expose the AMHTML object model, the AMHTML Component must be instanti-
ated, as shown in the following line of code:
Set myExchObj = Server.CreateObject(“AMHTML.Application”)

where myExchObj is the name of the object instantiated by the AMHTML application. In order
for the different message items and views to be accessed, they must be translated or rendered
into HTML. This rendering process is activated by calling the CreateRenderer method, as seen
below:
Set myRenderingObj = myExchObj.CreateRenderer(arg)

where myRederingObj is the instantiated reference to the CreateRenderer tag and arg is the
CreateRenderer argument. The exposed items can then be obtained by extracting information
from the session Inbox or OutBox. When the specific view is called, the Datasource object is
used to act as an index to retrieve specific properties from the selected item. To use the Part
Datasource property, use the following code example: IV
myRenderingObj.DataSource = myMessageItem Ch
Keep in mind that Exchange requires proper authorization every time you call any Active Mes- 17
saging HTML (AMHTML) object. Therefore, session validation will be required when interfacing
the AMHTML object on different ASP pages. The authorization can be verified by using the
Response object, as covered in Chapter 14. To ensure that proper authorization is available,
use the following code to trap the authentication method supported by the server:
<%
bAuthType = Request.ServerVariables(“auth_type”)
If InStr(1, “_basicNTLM”, bAuthType, vbTextCompare) < 2 Then
Response.Buffer = TRUE
Response.Status = (“401 Authorized”)
Response.AddHeader “WWW.Authenticate”, “Basic”
Response.End
End If
%>

The previous code example extracts the authentication scheme from the Request object. Next,
the VBScript InStr function checks to see whether the requested authentication scheme sup-
ports NTML authentication. If so, the Response object is used to prompt the user for a
username and password by forcing a basic authentication dialog with the server. Basic authen-
tication transmits the username and password between the browser and the Web server to
provide login permissions to the Web server. NTML authentication also verifies user authenti-
cation but does not send the actual username and password between the browser and the Web
server to prevent possible security violations. As a result of this ASP code, the user is pre-
sented with a authentication dialog box to log in to the system, as shown in Figure 17.7.

Table 17.8 gives a description of the available AMHTML objects that are available once the user is
authenticated. These AMHTML object can be accessed via ASP create, read, send, and delete
messages.

22.1389-6 CH17 351 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 352 Chapter 17 Implementing Existing Server-Side Components

FIG. 17.7
The user is prompted
for proper authorization
to access the Active
Messaging/SMTP
objects.

Table 17.8 Active Message Component Objects

Object Purpose
AddressEntry Specifies address information for an individual messaging
user.
Application Acts as a top-level object for a messaging application.
Attachment Associates an additional object with a mail item.
Attachment collection Accesses all attachments on a mail item; creates new
attachments.
Folder Opens the default Inbox or Outbox folder in a message store.
Items collection Accesses all mail items in a folder; creates new mail items.
MailItem Composes, populates, sends, and receives e-mail documents.
Recipient Specifies information for a messaging user intended to
receive a mail item.
Recipients collection Accesses all recipients of a mail item; creates new recipients.
Session Establishes a connection between an application and a
messaging system.
Sessions collection Accesses all sessions for an application; creates new sessions.

The power of the AMHTML object can be easily seen in a simple ASP example that opens the
Inbox view of an Exchange Server user profile view and reads the first message.
<%
Set myMapiObj = Server.CreateObject( “MAPI.Session” ) ‘=== Creates MAPI
➥Session
strProfile = “myExchangeServer” + vbLF + “myUserName” ‘=== Creates Login
➥String
myMapiObj.Logon “”, “”, False, True, 0, True, strProfile ‘=== Login the
➥MAPI Session
Set myInboxObj = myMapiObj.Inbox ‘=== Opens the
➥InBox
Set myMessageObj = myInboxObj.Messages.GetFirst ‘=== Get the first
➥message
Set myRenderApp = Server.CreateObject(“AMHTML.Application”) ‘=== Creates
➥AMHTML Object

22.1389-6 CH17 352 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Active Messaging/SMTP Component 353

Set myRendererObj = myRenderApp.CreateRenderer(2) ‘=== Begin

➥translation
myRendererObj.DataSource = myMessageObj ‘=== Set the
➥datasource
%>

The previous code logs you into the Exchange mail session and retrieves the first mail mes-
sage in the Inbox folder. When the code has finished processing the datasource, the current
message position is located on the first message. Now ASP code be integrated with HTML
code to extract the specific fields of the Exchange item, as shown in Listing 17.5.

Listing 17.5 Extracting information from an Exchange e-mail message.

<HTML>
<HEAD> Part
<TITLE>Read The First Exchange Inbox Message</TITLE>
</HEAD> IV
<BODY TEXT=000000 BGCOLOR=FFFFFF>
Message from:<% AddrType = Ch
myRendererObj.RenderProperty(ActMsgPR_SENT_REPRESENTING_ADDRTYPE)
Address =
17
myRendererObj.RenderProperty(ActMsgPR_SENT_REPRESENTING_EMAIL_ADDRESS)
myRendererObj.RenderProperty ActMsgPR_SENT_REPRESENTING_NAME, 0, Response %><BR>

To: <% myRendererObj.RenderProperty ActMsgPR_DISPLAY_TO, 0, Response %><BR>

Cc: <% myRendererObj.RenderProperty ActMsgPR_DISPLAY_CC, 0, Response %><BR>
Subject: <% myRendererObj.RenderProperty ActMsgPR_SUBJECT, 0, Response %>
Time: <% myRendererObj.RenderProperty ActMsgPR_CLIENT_SUBMIT_TIME, 0, Response
%><BR>
Message: <% myRendererObj.RenderProperty ActMsgPR_RTF_COMPRESSED, 0, Response %>
</Body>
</HTML>

Listing 17.5 assumes that the user has sufficient permissions to open the view and that a first
record exists in the Exchange Inbox. This listing will generate the HTML interface, shown in
Figure 17.8, which reads the message originally generated from the ASP code.

FIG. 17.8
The Active Messaging
HTML Component
exposes an SMTP
interface that enables
you to read, write, and
manage your mail
messages via ASP.

22.1389-6 CH17 353 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 354 Chapter 17 Implementing Existing Server-Side Components

The Active Messaging/SMTP Components are robust object models that are not completely
covered within this chapter. For complete information on the Active Messaging/SMTP object
models, see your online documentation or download the Active Messaging SDK from
www.microsoft.com.

The Tools Component

The Tools Component provides additional server-side processing mechanisms that add func-
tionality to server-side processing. The Tools Component uses a total of five different methods,
as shown in Table 17.9. Two of the five Tool Component methods, Owner and PluginExists,
are only applicable to the Personal Web Server for the Macintosh.

Table 17.9 The Tools Component Methods

Method Name Description
FileExists Checks for the existence of a file.
Owner Checks to see whether the current user is the site administrator
(Macintosh only).
PluginExists Checks the existence of a server plug-in (Macintosh only).
ProcessForm Processes an HTML form.
Random Generates a random integer.

The Tools Component can be created with either global scope or page-level scope. To create a
Tools Component in the GLOBAL.ASA file, use the following syntax:
<OBJECT RUNAT = Server SCOPE = Session ID = Tools PROGID = “MSWC.Tools”>
</OBJECT>

This OBJECT reference tag creates the Tools Component with session-level scope that can be
accessed from any ASP page for a particular user. The session-level object will exist until the
user Session object is destroyed. To create a page-level instance of the Tools Component in a
ASP file, use the CreateObject tag as shown below:
Set myToolObj = Server.CreateObject(“MSWC.Tools”)

The myToolObj object is instantiated from an instance of the TOOLS.DLL, which is typically in-
stalled in the \WINNT\System\InetSrv directory. The page-level instance of the myToolObj
object will be destroyed when the ASP is finished processing.

The FileExists Method

The FileExists method checks for the existence of a file within a virtual directory and returns
whether the file exists. This is a particularly useful feature because it not only helps prevent
broken HREF links to other pages, but it can also verify the existence of other files before
these files are called or used. For example, you might want to use the FileExists method to

22.1389-6 CH17 354 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Tools Component 355

verify that an image file, content scheduling file, or include file exists before trying to reference
the desired file. This method is particularly useful when sites are moved from Web server to
Web server to keep up with scaling and growth needs on your sites.

To use the FileExists method, the Tools Component requires the following syntax:
myToolObj.FileExists(URL)

where myToolObj represents the instantiated Tools Component and URL is the path to the re-
quest file.

The FileExists method returns either one of two values. The return value of –1 is used to
symbolize that the file is found as designated within the URL. If the file cannot be found, the
FileExists method returns a 0.

The FileExists method is also helpful when you want to ensure that a file exists before you Part
reference the file. You can use the FileExists method to prevent runtime errors or broken
link items. For example, to prevent a broken hyperlink, the FileExists method first checks for
IV
the existence of the calling page in the potential hyperlink, as shown below: Ch

<% Set Tools = Server.CreateObject(“MSWC.Tools”) 17

If Tools.FileExists(“=”/OnLineLearningCenter/chapter2.asp”) Then %>
<p> <a href=”/OnLineLearningCenter/chapter2.asp”>chapter2.asp</a></p>
<% Else %>
<p>File Not Found</p>
<% End If %>

In this code example, the FileExists method is used to determine whether the
/OnLineLearningCenter/chapter2.asp page exists. If the file does exist, the code generates a
valid link to that page. If the file does not exist, it prints a File Not Found message. Similarly,
the same process can be used to validate whether image files exist, as shown below:
<% Set Tools = Server.CreateObject(“MSWC.Tools”)
If Tools.FileExists(“corplogo.gif”) Then %>
<p> <a href=”http://www.i3solutions.com/”><img src=”corplogo.gif”
border=0></a><p>
<% Else %>
<p><a href=”http://www.i3solutions.com/”>Use Link Instead</a></p>
<% End If %>

In this code example, if the requested image file corplogo.gif does not exist, a text hyperlink
is generated to replace the missing image file.

In most situations, you will use the FileExists method in conjunction with other components
that do not have built-in error trapping for files that cannot be located. For example, the follow-
ing code checks to see if the acolades.txt exists before calling the Tools Component.
<%
Dim objTool
Set objTool = Server.CreateObject(“MSWC.Tools”)
If objTool.FileExists(“acolades.txt”) Then %> ‘=== If the file
➥exists, use the component
<TR><TD>
<%

22.1389-6 CH17 355 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 356 Chapter 17 Implementing Existing Server-Side Components

Dim NextTip
Set NextTip = Server.CreateObject(“MSWC.ContentRotator”)
Response.Write NextTip.ChooseContent(“acolades.txt”)
%>
</TD></TR>
<TR><TD>
<% Response.Write NextTip.ChooseContent(“acolades.txt”)%>
</TD></TR>
<%
Else ‘=== File does not exist, write a text version
%>
<TR><TD>
<FONT FACE=”Arial”>Make the world spin a little faster -
<A HREF = “userprofile1.asp”>Reserve</A> your
copy of the MIND-Bender ActiveX control today.</FONT>
</TD></TR>
<% End If %>

In this code example, the Tools Component is used to check that the acolades.txt file does
exist before creating an instance of the Content Rotator control. If the Content Rotator control
called the missing acolades.txt file, the control would generate a fatal runtime error. To pre-
vent this situation from occurring, the Tools Component verifies whether the file exists. If the
file does exist, the Content Rotator control generates content from the acolades.txt file. If the
Tools Component does not find the acolades.txt file, a static message is displayed.

The Owner Method

The Owner method is specific to the Personal Web Server on the Macintosh computer. The
Owner method checks to see whether the current user of the site has administrator privileges
for the Web server. To use the Owner method, use the following syntax:
Tools.Owner

The Owners method returns–1 if the current user has administration privileges, and 0 if the
current user does not have administration privileges. For example, the following code displays
the administration menu for an application to the site administrator only.
<%
Set Tools = Server.CreateObject(“MSWC.Tools”) ‘=== Creates Tools Object
If Tools.Owner Then ‘=== Checks Owner Method
Response.Redirect “/menu/AdminMenu.asp” ‘=== Redirect to Administration
➥Menu
Else
Response.Redirect “/menu/RegularMenu.asp” ‘=== Redirect to Normal Menu
End if
%>

The PluginExists Method

The PluginExists method is also a Macintosh Personal Web Server-specific method that veri-
fies if server-side plug-ins have been installed. To use the PluginExists method, use the fol-
lowing syntax:
Tools.PluginExists(PluginName)

22.1389-6 CH17 356 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Permission Checker Component 357

where PluginName is the name of the Web server plug-in. The PluginExists method also re-
turns True or False to indicate whether the plug-in exists. If the specific plug-in does exist, the
method returns –1. Likewise, if the server-side plug-in is not found, the PluginExists method
returns 0. The following example performs a test to a plug-in named myPlugin on the
Macintosh PWS.
<%
If Tools.PluginExists(myPlugin) Then
‘=== Perform Plug-in methods
End If
%>

The Random Method

The Random method is used to generate a random integer between -32768 and 32767. To use
Part
the Random method, use the following syntax:
Tools.Random
IV
Ch
To display a random number, use the following code example:
<% Set Tools = Server.CreateObject(“MSWC.Tools”) %>
17
<% Response.Write Tools.Random %><BR> <!--Writes a Random number to the
page -->

Keep in mind that you can use other arithmetic functions in conjunction with the random
method. For example, to limit the scope or range of the randomly generated numbers, use the
MOD and ABS statements:

<% = ( Abs( Tools.Random ) ) Mod 10 %> <!--Generates a random number from 0

to 10 -->

This example generates random numbers between 0 and 10.

The Permission Checker Component

The Permission Checker Component is used to identify whether the requesting user has per-
mission to access a file, which enables you to maintain strict security levels for files that are
tailored for both private and public viewing. The Permission Checker Component not only
gives you the ability to check access permissions after a request for a page has been made, but
also to control navigation flow. For example, you can first check permissions and, depending on
whether the user is authorized, display or hide navigation links. This capability to monitor
access is particularly important for permission-dependant applications and pages with intranet-,
extranet-, and Internet-based applications. For example, you can use the Permission Checker
Component to only give managers access to monthly payroll accounts that regular employees
should not see.

To use the Permission Checker Component, use the following syntax:

Set myPermissionObj = Server.CreateObject(“MSWC.PermissionChecker”)

22.1389-6 CH17 357 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 358 Chapter 17 Implementing Existing Server-Side Components

where myPermissionObj is the name of the instantiated Permission Checker object. When
using the Permission Checker Component, keep in mind that Internet Information Server
supports three different authentication schemes. Anonymous access uses the NT Internet ac-
cess account to control access to files within the Web Server domain. If anonymous access is
selected for files, the Permission Checker Component has nothing to authenticate because the
requesting browser already has access to the files within NT.

N O T E For permission-dependent applications and pages, disable the anonymous access

authentication to prevent it from overriding the Permission Checker Component. ■

If anonymous access has been disabled, NT uses the basic or NTLM authentication method to
monitor permissions. You can set individual or group permissions by using the Access Control
List. Chapter 14 demonstrated how to force authentication by using the following script:
<%
If Request(“LOGON_USER”) = “” Then
Response.Status = “401 Unauthorized”
End If
%>

However, the Permission Checker Component accomplishes the same task by using the
HasAccess method, as shown in Table 17.12.

Table 17.12 The Permission Checker Component’s Method

Method Name Description
HasAccess Verifies whether the requesting browser has access to a file.

The HasAccess Method

The HasAccess method is used to determine whether the requesting user has permission to
read a file. To use the Permission Checker Component, the HasAccess method requires the
following syntax:
HasAccess(FilePath)

where FilePath is the physical or virtual path to the requested file. The HasAccess method
returns a True value if the user is successfully authorized and False if the authentication fails.

You can use the Boolean result from the HasAccess method to check access permissions. The
following code displays the access permission for both a virtual and a physical file path.
<% Set myKeyObj = Server.CreateObject(“MSWC.PermissionChecker”) %>
Permissions: <BR>
Virtual Directory: <%= myKeyObj.HasAccess(“c:\inetpub\wwwroot\default.asp”)
%><BR> <!-- Virtual -->
Physical Directory: <%= myKeyObj.HasAccess(“/default.asp”) %><BR> <!--
Physical -->
Mis-spelling: <%= myKeyObj.HasAccess(“/default.asp”) %><BR> <!-- Mis-spelled
-->

22.1389-6 CH17 358 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Permission Checker Component 359

This code examines the default.asp file permissions and returns the results. Notice that the
HasAccess method does not return an error if the file does not exist. If the file cannot be found,
the HasAccess method returns a False value, as demonstrated below:
Permissions:
Virtual Directory: True
Physical Directory: True
Mis-spelling: False

The most common means of implementing the HasAccess method is to check the access per-
missions for a Web page or file and, if the user is authorized, allow access to a requested site or
page. Typically, this is implemented by presenting a hyperlink pointing to the permission-
dependent site. For example, Listing 17.6 demonstrates using the Permission Checker Compo-
nent to check access rights to the report1.asp.
Part
Listing 17.6 Using the Permission Checker to authorize access to information. IV
<% Ch
dim vUserHasAccess
Set myKey = Server.CreateObject(“MSWC.PermissionChecker”) ‘=== Creates 17
Permission Object
UserHasAccess = myKey.HasAccess(“/private/report1.asp”) ‘=== Checks
permission of report1.asp
%>
<% If vUserHasAccess Then %>
<A HREF = “/private/default.htm”>Administrative Reporting!</A> ‘=== If
Authorized, display this link
<% Else %> ‘=== Else warn off
the user
Sorry, you are not authorized to access this report. Your IP Address has
been recorded.
<% End If %>

Keep in mind that you do not have to display a results page with a hyperlink to let the user
know if he has the appropriate permissions. You can provide a seamless navigational flow di-
rectly to the requested page using the Response.Redirect method. The direct routing requires
fewer steps for the user and prevents unnecessary requests from being processed by the
server. The following example uses the Response.Redirect method to eliminate displaying the
Web page that displays the Administrative Reporting hyperlink and to automatically route the
browser to the /private/default.htm page:
<%
dim vUserHasAccess
Set myKey = Server.CreateObject(“MSWC.PermissionChecker”) ‘=== Creates
➥Permission Object
vUserHasAccess = myKey.HasAccess(“/private/report1.asp”) ‘=== Checks
➥permission of report1.asp
If vUserHasAccess Then
Response.Redirect “/private/report1.asp” ‘=== If Authorized,
➥route user to this page
End If
%>

22.1389-6 CH17 359 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 360 Chapter 17 Implementing Existing Server-Side Components

The Content Linking Component

The Content Linking Component is used to help manage and control links within a set of docu-
ments. The set of documents is usually clearly defined, such as a table of contents for a docu-
ment, an online manual, or the required steps for a wizard. The Content Linking Component
provides value by generating and maintaining the order of the files and tracking which item is
active in this list.

To use the Content Linking Component, use the following syntax:

Set myNextLink = Server.CreateObject(“MSWC.Nextlink”)

where myNextLink is the reference to the instantiated MSWC.Nextlink object. The Content
Linking Component is an instantiated instance of the Nextlink.DLL file, which is usually in-
stalled in the \WINNT\System\InetSrv directory.

The NextLink Files

The Content Linking object requires two files to properly function on the Web server, as shown
in Table 17.13. The first is the NextLink.DLL filter, which is required for the creation and inter-
pretation of the Content Linking object. The second file is the Content Linking list. The Con-
tent Linking list is responsible for managing the sequential list of files within a single text file,
and managing the current location of the user in the schedule file list. Now groups of pages can
be easily managed and maintained without having to update links on each page. This automatic
update of the navigation links can be seen in Figures 17.9 and 17.10. Figure 17.9 illustrates
displaying content with a next page link that points to the next URL.

FIG. 17.9
The Content Linking
Component stores your
relative position in a list
of sequential URLs and
automatically updates
the navigation links with
the appropriate links.

After clicking the Next Page link, the second page of content is displayed and the Content
Linking Component creates a Previous Page link and updates another Next Page link, as
shown in Figure 17.10. In this situation, the Previous Page link points to the first page of con-
tent and the Next Page link points to the third URL. If the Next Page link were activated, the
same process of automatically updating the URL navigation link would be repeated.

22.1389-6 CH17 360 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Linking Component 361

FIG. 17.10
After moving to the
second page, the
Content Linking
Component updates
the URL locations for
the navigation links with
new URLs embedded in
the Previous Page and
Next Page hypertext.

Part
IV
Table 17.13 The Content Linking Component Required Files Ch

Name Description 17
NextLink.DLL Required to create the Content Linking object.
Content Linking list A text file used to sequentially list the files to be viewed.

The Content Linking List

The Content Linking list serves as the master index for managing workflow and scheduling
content on the Web. The Content Linking file stores the Web page location, textual information
describing the location, and comments for the specified item. The Content Linking list uses the
tab character to separate the URL field, the description field, and the comments field, as shown
in the following syntax:
Page-URL[➡][ text-description[➡][ comment]]

where Page-URL is the destination of the text description field, text-description is the de-
scription of the URL, and comment is comments to help identify and track the pages.

N O T E The Content Linking file searches for the tab character to separate the fields and a hard
return to create a new row. ■

The following code is an example of a Content Linking list that is used to consecutively manage
and navigate three pages and their descriptions:
---------- MylinkText.txt ----------
Chapter1.asp Chapter 1 First Chapter
Chapter2.asp Chapter 2 Second Chapter
Chapter3.asp Chapter 3 Third Chapter

22.1389-6 CH17 361 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 362 Chapter 17 Implementing Existing Server-Side Components

In the previous example, the Content Linking list is named MylinkText. The Content Linking
list contains three Web pages that are to be displayed in sequential order. The first page to be
displayed is the Chapter1.asp file, with a description of Chapter 1. The comment First chap-
ter is added just as a reference for the authors or administrators of the list file.

The Content Linking Methods

After the series of files has been designated in the Content Linking list, the Content Linking
object uses methods to retrieve and navigate through the Content Linking list. The Content
Linking Component uses eight methods to help manage your Content Linking list, as shown in
Table 17.14. These methods are generally separated into the forward and reverse navigation
(or up and down) through the list or are used to retrieve information about the link.

Table 17.14 The Content Linking Methods

Name Description
GetListCount Counts the number of items linked in the Content Linking
list file.
GetNextURL Gets the URL of the next page listed in the Content
Linking list file.
GetPreviousDescription Gets the description line of the previous page listed in the
Content Linking list.
GetListIndex Returns the index of the current page in the Content
Linking list.
GetNthDescription Gets the description of the Nth page listed in the Content
Linking list.
GetPreviousURL Gets the URL of the previous page listed in the Content
Linking list.
GetNextDescription Gets the description of the next page listed in the Content
Linking list.
GetNthURL Gets the URL of the Nth page listed in the Content Linking
list.

The GetListCount and GetListIndex methods are used to refer to the number of items in the
Content list and the current page. The GetNextURL, GetPreviousURL, and GetNthURL methods
use the Content Linking Component to retrieve the URL locations from the Content Linking
list. The GetNextDescription, GetPreviousDescription, and GetNthDescription methods
complement the previous methods to extract description values from the Content Linking list.

22.1389-6 CH17 362 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Linking Component 363

GetListCount The Content Linking Component uses the GetListCount method to retrieve
the total number of pages stored in the Content Linking list. The total number of pages stored
in the list is important to help loop through the Content list. To use the GetListCount method,
use the following syntax:
GetListCount(listURL)

where listURL is the virtual path to the Content Linking list. To determine the number of
pages in the Content Linking list, use the following example:
<%
Set myLinkObj = Server.CreateObject(“MSWC.NextLink”)
iCount = myLinkObj.GetListCount(“MylinkText.txt”)
%>

In this situation, the iCount returns a value of 3. The GetListCount index is 1-based, meaning
Part
that it starts at 1 and contains integers greater than 1. However, if the Content Linking list
MylinkText.txt is not found, the GetListCount method returns a zero value. If you try to IV
perform subsequent methods on a Content Linking list that cannot be found, a runtime error Ch
will be generated. To prevent these runtime errors from occurring, trap the value of the count
property, as shown below:
17
<%
Set myLinkObj = Server.CreateObject(“MSWC.NextLink”)
iCount = myLinkObj.GetListCount(“MylinkText.txt”)
Do While (iCount > 0) %>
--- Do Something ---
Loop
%>

The GetListCount method is also useful in navigation management. For example, you may
want to add special hyperlinks or closing messages when the last page is reached. You can
accomplish this by using the GetListCount and GetListIndex methods. The GetListCount
method returns the number of pages in the list and the GetListIndex determines the current
page, as demonstrated in the following code:
<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”) ‘=== Create
➥Content Linker
iCurrent = myNextLinkObj.GetListIndex (“/data/mylink.txt”) ‘=== Retrieves
➥current position
iMax = myNextLinkObj.GetListCount(“/data/mylink.txt”) ‘=== Determines
➥maximum amount of links
If iCurrent = iMax Then ‘=== Checks for
➥max condition
Response.Write (“Thank you for registering for “&vClassTitle & “<BR>”)
Response.Write(“<A HREF = ’top.menu.Reload(main)’>Main</A>)
End If
%>

In this example, if the user has successfully navigated to the last page of the registration wiz-
ard, that is, iCurrent = iMax, a thank you message is written to the page with a hyperlink that
calls a function named Reload, which is located in the frame named menu.

22.1389-6 CH17 363 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 364 Chapter 17 Implementing Existing Server-Side Components

GetListIndex The GetListIndex method is used to return the list index of the current page in
the Content Linking list. To use the GetListIndex method, use the following syntax:
GetListIndex(listURL)

where listURL is the virtual path to the Content Linking list. The GetListIndex method re-
turns an integer result identifying the current position within the Context Linking list. To re-
trieve the current position within the content list, use the GetListIndex method, as shown in
the following example:
<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”)
IcurrentLink = myNextLinkObj.GetListIndex (“/data/mylink.txt”)
%>

In most situations, you will use the GetListIndex method to customize the display for the first
page within the list. For example, if you only want have a single hyperlink to the second page,
but have forward and reverse links on the remaining pages, you would trap the GetListIndex
value for a return result of 1 and tailor the results accordingly.
<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”)
If (myNextLinkObj.GetListIndex (“/data/mylink.txt”) > 1) Then
‘=== Display only the forward navigation toolbar on first page
Else
‘=== Display standard forward and reverse toolbar on remaining pages
End If
%>

GetNextURL The next three methods, GetNextURL, GetPreviousURL, and GetNthURL, are used
to retrieve the URL locations from the Content Linking list. The GetNextURL method is used to
take the current position indicator and retrieve the next URL in the Content Linking list. To
implement the GetNextURL method, use the following syntax:
GetNextURL(listURL)

where listURL is the virtual path to the Content Linking list. For example, to extract the next
URL location, use the GetNextURL method as shown in the following example:
<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”) %>
<a href = “<%= myNextLinkObj.GetNextURL (“/data/mylink.txt”) %>”>Next Page</a>

This example retrieves the next URL in the Content Linking list and inserts the URL into the
hyperlink ANCHOR tag, as you saw in Figure 17.9. When the user clicks the Next Page link, the
next URL in the list will be viewed. For example, if the user is located on the second content
page, chapter2.asp, the Next Page hyperlink will point to chapter3.asp.

N O T E The GetNextURL will point to the first URL in the Content Linking list if the user is
positioned on the last page in the list. ■

You can combine the GetListIndex method sample code with the previous GetNextURL sample
code to only generate a Next Link hyperlink when the user is on the first page in the schedul-
ing file.

22.1389-6 CH17 364 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Linking Component 365

<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”)
If (myNextLinkObj.GetListIndex (“/data/mylink.txt”) > 1) Then %>
<a href = “ <%= myNextLinkObj.GetNextURL (“/data/mylink.txt”) %>”>Next
➥Page</a>
<% Else %>
‘=== Display standard forward and reverse toolbar on remaining pages
<% End If %>

GetPreviousURL The GetPreviousURL tag complements the GetNextURL method by enabling

reverse or descending navigation through the items in the content list. The GetPreviousURL
method can be executed by using the following syntax:
GetPreviousURL(listURL)

where listURL is the virtual path to the Content Linking list. To navigate to the previous URL
in the list, use the GetPreviousURL method, as shown in the following example: Part

<%
IV
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”) Ch
<a href = “<%= myNextLinkObj.GetPreviousURL (“/data/mylink.txt”) %>”>Previous
Page</a>
%>
17
Now you can combine the GetPreviousURL and the GetNextURL methods to provide page-by-
page navigation through your content.
<% ‘============ Next Page Navigation HyperLink ===============================
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”) %>
<a href = “<%= myNextLinkObj.GetNextURL (“/data/mylink.txt”) %>”>Next Page</a>
<% ‘========================== Previous Page Navigation Hyper Link =========== %>
<a href = “<%= myNextLinkObj.GetPreviousURL (“/data/mylink.txt”) %>”>Previous
Page</a>

This code sample is what generates the browser navigation you saw in Figures 17.9 and 17.10.

GetNthURL The GetNthURL method rounds out the URL navigation methods by enabling the
user to skip beyond the sequential navigation techniques demonstrated with the
GetPreviousURL and GetNextURL navigation methods. Typically this specific index navigation is
used when specific items are picked from a list, such as a table of contents or picking the start-
ing alphabetical letter form an index ranging from A to Z. The GetNthURL provides the capability
to navigate to any URL within the Content Linking list based on the index number. To imple-
ment the GetNthURL method, use the following syntax:
GetNthURL(listURL, iContent)

where listURL is the virtual path to the Content Linking list and iContent is the URL location
specified in the Content Linking list. For example, to jump from any page within the content list
and navigate to the chapter2.asp listing, set the iContent variable to 2, as demonstrated in the
following example:
<% Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”) %>
<a href = “<%= NextLink.GetNthURL(“/data/mylink.txt”, 2)%>”> “>Chapter 2</a>

With ASP filename information, you can display the entire navigational list, as shown in the left
frame in Figure 17.11.

22.1389-6 CH17 365 10/26/98, 9:30 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 366 Chapter 17 Implementing Existing Server-Side Components

FIG. 17.11
Use the GetNthUrl
method to provide list or
table of contents
navigation.

Now, with this ability to manipulate URL information, you can create a complete navigational
toolbar with move-first, move-last, move-next, and move-previous navigation links. This func-
tionality is implemented using the combination of the GetNthURL, GetNextURL, and
GetPreviousURL methods (see Figure 17.12).

FIG. 17.12
Use the combination of
the GetNthUrl,
GetPreviousUrl, and
GetNextURL methods
to construct a complete
navigational toolbar.

22.1389-6 CH17 366 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Linking Component 367

The navigation links displayed in Figure 17.12 were created by using the code shown in List-
ing 17.7.

Listing 17.7 Creating a content navigation toolbar using the Content Linking
Component.
<%
Set myNextLinkObj = Server.CreateObject (“MSWC.NextLink”)
iMaxLinks = myNextLinkObj.GetListCount(“/data/mylink.txt”)
iMinLink = 1 ‘=== Linking file is 1-based
%>
<Center>
<Table Cellpadding = 4 CellSpacing = 4 border = “0”>
<TR align = center><TD>
<a href = “<%= myNextLinkObj.GetNthURL(“/data/mylink.txt”, iMinLink)%>”> Part
<<< S T A R T</a>
</TD><TD> IV
<a href = “<%= myNextLinkObj.GetPreviousURL (“/data/mylink.txt”) %>”>
Ch
< P R E V </a>
</TD><TD>
<a href = “<%= myNextLinkObj.GetNextURL(“/data/mylink.txt”)%>”>
17
N E X T > </a>
</TD><TD>
<a href = “<%= myNextLinkObj.GetNthURL(“/data/mylink.txt”, iMaxLinks)%>”>
E N D >>> </a>
</TD></TR>
</Table>
</Center>

In Listing 17.7, the hyperlinks to the move-first, move-next, move-previous, and move-last links
are embedded within an HTML table to ensure page layout. The iMaxLinks variable is used to
set the total number of links contained within the Content Linking list.

Now you can combine the table of content view with the sequential page navigation methodol-
ogy to enable complete navigation through your content (refer to Figure 17.11).

The table of contents view, shown in the left frame, is generated by looping through all the URL
locations in the Content Linking list. In this situation, the direct ASP filename is being dis-
played as the hyperlink.
<% Set myNextLinkObj = Server.CreateObject(“MSWC.NextLink”) %>
<% count = myNextLinkObj.GetListCount(“mylink1.txt”) %>
<% I = 1 %>
<% Do While (I <= count) %>
<A Href = “<%= UCASE(myNextLinkObj.GetNthURL(“mylink1.txt”, I))%>”>
<% Response.Write UCASE(myNextLinkObj.GetNthURL(“mylink1.txt”, I))%></A><BR>
<% I = (I + 1) %>
<% Loop %>

GetNextDescription The remaining three methods, GetNextDescription,

GetPreviousDescription, and GetNthDescription, mimic the functionality of the previously
discussed URL navigation features, but instead retrieve the descriptions from the Content
Linking file.

22.1389-6 CH17 367 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 368 Chapter 17 Implementing Existing Server-Side Components

The GetNextDescription method is used to retrieve the description of the next content item in
the Content Linking list. To use this method, the following syntax is required:
GetNextDescription(listURL)

where listURL is the virtual path to the Content Linking list. The GetNextDescription method
provides a better description of the content file than just displaying the filename itself.
<% Set myNextLinkObj = Server.CreateObject(“MSWC.NextLink”) %>
<a href = “<%= myNextLinkObj.GetNextURL(“/data/mylink1.txt”)%>”>
<%= myNextLinkObj.GetNextDescription (“/data/mylink1.txt”)%></a>

In the previous code example, the next content page URL address is placed inside the anchor
list using the GetNextURL method. But now the description of the page can also be extracted
from the Content Linking list and used as the description within the hyperlink. For example, if
the user is currently on the chapter1.asp page, which is the first item in the Content Linking
list, the hyperlink would read Chapter 2 and the HTML anchor tag would point to
chapter2.asp.

GetPreviousDescription The GetPreviousDescription method complements the

GetNextDescription method by supplying the capability to move to the previous description
within the Content Linking list. To implement this feature, use the following syntax:
GetPreviousDescription(listURL)

where listURL is the virtual path to the Content Linking list. To provide a user-friendly mes-
sage description to return to a previous file, use the GetPreviousDescription method, as
shown in the following example:
<% Set myNextLinkObj = Server.CreateObject(“MSWC.NextLink”) %>
<a href = “<%= myNextLinkObj.GetPreviousURL(“/data/mylink1.txt”)%>”>
<%= myNextLinkObj. GetPreviousDescription (“/data/mylink1.txt”)%></a>

If the user has navigated to the second page within the Content Linking list, this sample code
would create a hyperlink with the text of Chapter 1 and the HTML anchor tag of
chapter1.asp.

GetNthDescription The GetNthDescription method enables you to extract the URL descrip-
tion from any valid content index without using relative positions, as is the case with the
GetNextDescripton and GetPreviousDescriptions methods. To implement this non-sequential
description retrieval method, the Content Linking Component requires the following syntax:
GetNthDescription(ListUR, iURL)

where ListURL is the virtual path to the Content Linking list and iURL is the URL location speci-
fied in the Content Linking list. The GetNthDescription method complements the GetNthURL
method to provide a user-friendly description of the URL. For example, an easier to read table
of contents can be created by looping through the Content Looping file and using the
GetNthDescription method to display the descriptions, Chapter 1, Chapter 2, and Chapter 3, as
shown in Figure 17.13.

22.1389-6 CH17 368 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Linking Component 369

FIG. 17.13
The
GetNthDescription
method provides a
useful description of
the content pages.

Part
IV
Ch

17

The code behind Figure 17.13 illustrates the various aspects of using the GetListCount
method, looping through the Content file, and using the GetNthDescription and GetNthURL
methods to provide navigation and description information.
<B>Online Study Guide:</B><BR><BR>
<% Set myNextLinkObj = Server.CreateObject(“MSWC.NextLink”) %>
<% count = myNextLinkObj.GetListCount(“mylink.txt”) %>
<% I = 1 %>
<% Do While (I <= count) %>
<A HREF = “<%= myNextLinkObj.GetNthURL(“mylink.txt”, I)%>”>
<%= myNextLinkObj.GetNthDescription (“mylink.txt”, I)%></A><BR>
<% I = (I + 1) %>
<% Loop %>

Keep in mind that you are not just limited to using hyperlinks for your page navigation. Figure
17.14 demonstrates using HTML buttons to provide a navigation methodology.

Furthermore, remember to exploit the description field within the Content Linking list. This
description field is treated as a text field, and any values within this description field are trans-
lated into HTML. This means that you can embed HTML within the content location file, such
as image files. Figure 17.15 illustrates the use of adding HTML tags with the DESCRIPTION tag
of the Content Linking list.

22.1389-6 CH17 369 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 370 Chapter 17 Implementing Existing Server-Side Components

FIG. 17.14
Use any workflow or
navigation motif, such
as images, buttons, or
links, with the Content
Linking Component.

FIG. 17.15
HTML tags, such as the
IMAGE tag, can be
embedded within the
description field of the
Content Linking file.

Figure 17.15 demonstrates the capability to embed various HTML tags within the description
field of the Context Linking list. This entire HTML page, including the images, hyperlinks,
anchors, and bulleted lists, exists within the description field. To ensure that the following code
examples are easy to read, only the image attributes have been included in the following Con-
tent Linking list.

22.1389-6 CH17 370 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Rotator Component 371

---------- myLink.txt with embedded HTML tags----------

chapter1.asp <img src = “bplaney.gif” border=0 width=30 height=20><Font
Size=+1><B>Chapter 1</B></font></a>: This chapter ...
chapter2.asp <img src = “bswoop.gif” border=0 width=30 height=20><Font
Size=+1><B>Chapter 2</B></font></a>: The second chapter will discuss the
residual effects of the revolution on the emerging <a href =
“peasants.htm”>peasant class</a>.<BR><BR>
chapter3.asp <img src = “bglobey.gif” border=0 width=30 height=20><Font
Size=+1><B>Chapter 3</B></font></a>: The third chapter identifies the key ...

Remember that the Context Linking Component only looks for tabs to separate the URL, de-
scription, and comment fields when parsing through the Content Linking list. Therefore, you
can embed HTML characters directly in the Context Linking list.

The Content Rotator Component Part

The Content Rotator Component is responsible for displaying and managing a schedule of how IV
often content is displayed on a page. The role of the Content Rotator is similar to the Ad Rota- Ch
tor Component; however, instead of displaying an image banner, it delivers HTML content to
the page. The Content Rotator expands this role by not only providing the capability to transfer
17
content, but also providing the capability to embed HTML tags within the content string itself.
This embedding feature enables you to provide HTML formatting, hyperlinks, and image tags.

To use the Content Rotator Component, the following syntax is required:

<% Set myObjVar = Server.CreateObject(“MSWC.ContentRotator”) %>

where myObjVar is the reference to the instantiated Content Rotator Component.

The Content Rotator requires two files, the ContRot.dll and the Content Schedule file, as
shown in Table 17.15. The ContRot.dll file is the Dynamic Link Library needed to create the
Content Rotator Component on the server.

Table 17.15 The Content Rotator Required Files

Name Descriptions
ContRot.dll Required DLL for the Content Rotator.
Content Schedule file Text file that manages the content and the frequency with
which it is displayed.

The Content Schedule File

The Content Schedule file has two responsibilities. The first responsibility is simply to house
the different content items. The second responsibility is to set the frequency or weight with
which the content is displayed. The Content Schedule file uses the following syntax:
%% [#Weight] [//Comments] ContentString

22.1389-6 CH17 371 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 372 Chapter 17 Implementing Existing Server-Side Components

where the double percent symbols (%%) designate a new content item; #Weight is an optional
value that specifies the relative weight, which determines how often the specific content item is
displayed; Comments enables comments to be attached to the content item; and ContentString
is the content string.

Listing 17.8 is an example of a Content Linking file.

Listing 17.8 A Sample Content Rotator file.

-------------content.txt--------------------
%% #1 // The DataSource Solutions Item
<body><FONT FACE = “ARIAL” SIZE = “2”>
<IMG SRC = “barrow.gif”>World-wide Intranet rollout <a href = “/DS/
Intro.ASP”>saves $300k</a> per month
</FONT></BODY>

%% #1 // NetHouse AD
Download the newest <a href = “/NH/default.asp”>IP packet tools</a> from
NetHouse Inc

%% #0 // Place ad once bill is paid

<Body BGCOLOR = “#CC9900”>
New design tools from <H2>World Environmental Technologies Inc</H2> <A HREF =
“/wet.asp”>(WETInc)</A> help simulation programs reduce vehicle emissions
</Body>

The weight of each content item acts as the relative weight compared to the other content
items. The greater the relative weight of the content item, the greater the frequency with which
content item will be displayed. The relative weight is determined by dividing the weight of the
content item by the sum of the other items in the Content Schedule file, as viewed in the follow-
ing formula:
Probability = Weight of Content Item / (Sum of all weights in Content File)

For example, in the listed Content Schedule, the first item is assigned a weight value of 1 and
the second item is assigned the weight of 3. The probability that the first content item will be
displayed is 25 percent. The second content item has 75 percent probability that it will be dis-
played. For example, the Amazing DataWidget WareHouse online catalog, as demonstrated in
Chapter 19, “Putting It All Together: Creating an Online Registration System,” uses this prob-
ability to determine the frequency with which product and related news information will be
displayed on the site. To gain attention to a particular sale item or news event, increase the
weight in the Content Schedule file.

By default, if no weight is assigned, the content is assigned a value of 1. If a content item is

assigned a weight of 0, the content item will be ignored.

22.1389-6 CH17 372 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Content Rotator Component 373

Notice that the content items can house HTML tags. For example, the third content link
changes the background color of the page to #CC9900.

Remember that this content information is not being physically rotated in a circle on the HTML
page. The Content Rotator simply determines what information will be loaded the next time the
page is requested.

The Content Rotator Methods

The Content Rotator uses two methods to post and rotate content to a page (see Table 17.16).
The ChooseContent method is used to retrieve a specific content item. The GetAllContent
method retrieves all items from the Content Schedule file.

Table 17.16 The Content Rotator Methods Part

Name Description IV
ChooseContent Extracts a specific content item from the Content Schedule file. Ch

GetAllContent Extracts all the content items from the Content Schedule file. 17

The ChooseContent Method

The ChooseContent method extracts individual content items from the Content Linking list
based on the item’s relative weight each time the page is called. To retrieve content informa-
tion, use the following syntax:
ChooseContent(ContentSchedule)

where ContentSchedule is the virtual path pointing to the Content Schedule file. For example,
to extract the content items from the previous content file, use the following code:
<BODY bgcolor = “#FFFFFFF”><center>
<%
Set myContentItem = Server.CreateObject(“MSWC.ContentRotator”) %>
<% Response.Write myContentItem.ChooseContent(“/data/content.txt”) %>
</Center></Body>

This example produces the rotating content in the bottom frame shown in Figures 17.16 and 17.17.

Once the page is revisited or refreshed, the Content Rotating Component automatically up-
dates the page with new content from the Content Schedule file (see Figure 17.17).

Notice that the HTML BODY background tags in the calling ASP of white (#FFFFFF) are over-
written by the HTML contained within the Content Linking file.

22.1389-6 CH17 373 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 374 Chapter 17 Implementing Existing Server-Side Components

FIG. 17.16
The Content Rotator
pulls a content string
from the scheduling file
based on its weight or
frequency setting.

FIG. 17.17
The next time the page
is refreshed, new
content is retrieved from
the schedule file.

The GetAllContent Method

The GetAllContent method retrieves all the content in the Content Schedule file and displays
the content items on the calling page separated by horizontal rule (<HR>) tags. To use the
GetAllContent method, use the following syntax:

GetAllContent(ContentSchedule)

22.1389-6 CH17 374 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Page Counter Component 375

where ContentSchedule is the virtual path to the Content Schedule file. To display all the items
within the Content Schedule file, use the GetAllContent method, as shown in the following
example:
<HTML><BODY bgcolor = “#FFFFFFF”><center>
<% Set Tip = Server.CreateObject(“MSWC.ContentRotator”) %>
<% = Tip.GetAllContent(“/data/content.txt”) %>

This code produces the output displayed in Figure 17.18.

FIG. 17.18
The GetAllContent
method extracts all the
content information
from the Content Part
Schedule file.
IV
Ch

17

Notice that this example performs in a manner similar to the ChooseContent method in terms
of overwriting the calling HTML tags. The HTML tags assigning white (#FFFFFF) as the back-
ground color are overwritten by the last calling instance of the HTML BODY tag in the Context
Schedule file. In this situation, the HTML BGCOLOR tag is changed from #FFFFFF to #669900 in
the first content item and to #CC9900 in the last content item.

The Page Counter Component

The Page Counter Component is used to keep track of how many times a specific page has
been opened. The Page Counter Component temporarily stores in memory the number of
times a page has been opened and then stores the counter information in a text file at regular
intervals. This memory caching prevents excessive resources from being consumed with a
large number of input/output and read/write requests to the hard drive. To use the Page
Counter Component, the following syntax is required:
<% Set myObjVar = Server.CreateObject(“MSWC.PageCounter”) %>

22.1389-6 CH17 375 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 376 Chapter 17 Implementing Existing Server-Side Components

The Page Counter Component requires the combination of two files, the PageCnt.DLL and the
Hit Count Data file, as shown in Table 17.17.

Table 17.17 Required Page Counter Component Files

Name Description
PageCnt.Dll Dynamic Link Library needed to create the Page Count
component
Hit Count Data file The text file to which the Page Count Component writes the
current hit count.

The PageCnt.DLL is the Dynamic Link Library needed to create the Page Count Component on
the server. The Hit Count Data file is the text file that the Page Count Component writes the
current page count to. The Page Count Data file acts as a storage area to store the number of
times a page was opened. After a given number of hits, the page count that is stored in memory
is written to a text file. The Registry settings are used to specify how often this storing to a text
file occurs and stores the path location to the Hit Counter file. Table 17.18 summarizes the
Registry file settings.

Table 17.18 The Page Counter Component Registry Settings

Registry Name Description
File_Location Specifies the location and filename of the counter file.
Save_Count Specifies the increment at which the current page count is
written to the Hit Count Data file (default is 25).

By default, the Hit Count Data file name is HitCnt.cnt in your Windows directory. If you follow
the standard NT installation, the HitCnt.cnt file will be located in the \WINNT directory. The
HitCnt.cnt file stores two pieces of information: the name of the page and the number of times
that page was requested. An example of the HitCnt.cnt file is displayed in the following:
-------------- HitCnt.cnt --------------
11025 /data/download/default.asp
43775 /data/default.asp

The Page Counter Component Methods

The Page Counter Component uses two methods, as shown in Table 17.19, to manage page
count information.

22.1389-6 CH17 376 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Page Counter Component 377

Table 17.19 The Page Counter Component Methods

Method Description
Hits Returns the number of times a page has been opened.
Reset Initializes the value of a page count to 0.

The Hits Method

The Hits method of the Page Counter Component returns the number of times an Active
Server Page has been requested. The Hits method uses the following syntax:
Hits([pathInfo])
Part
where pathInfo is the optional location for the requested page. If no path information is given,
the current page is assumed. For example, to use the Page Counter Component to track and IV
display the page counter information for the current page, use the following code example: Ch
<% Set MyPageCounter = Server.CreateObject(“MSWC.PageCounter”) %>
This Web page has been viewed <%= MyPageCounter.Hits %> times.
17
Alternately, if you want to display information about the number of times that another page was
requested, use the Hits method and specify the path to the desired page, as shown below:
<%
Set MyPageCounter = Server.CreateObject(“MSWC.PageCounter”)
vPageStr = “/data/download.asp”
%>
The Web page, ‘<% =vPageStr %>’, has been viewed <%= MyPageCounter.Hits(vPageStr)
%> times.

This example looks at the page stored in the virtual path of /data/download.asp and extracts
the page count for that page, as shown in the following results:
The Web page, ‘/data/download.asp’, has been viewed 24750 times.

The Reset Method

The Reset method clears the current page count value and reinitializes the page count for the
page to zero. To utilize this method, use the following syntax:
Reset([pathInfo])

where pathInfo is the virtual path to the requested ASP. The Reset method shares many simi-
lar characteristics with the Hits method. If no path information is specified, the Reset method
assumes that it should be activated on the current page. The path information can also be
applied to various other ASP files by specifying the path information to the desired file. For
example, to reset the current ASP count to 0, use the following syntax:
<%
Set MyPageCounter = Server.CreateObject(“MSWC.PageCounter”) ‘=== Create
➥Component Instance

22.1389-6 CH17 377 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 378 Chapter 17 Implementing Existing Server-Side Components

MyPageCounter.Reset ‘=== Resets the

➥current page
%>

Furthermore, use the Path statement to reset counter information in multiple files located in
different directories. This use of the Path statement can be seen in the following example:
<%
Set MyPageCounter = Server.CreateObject(“MSWC.PageCounter”) ‘=== Create
➥Counter Object
vPageStr = “/data/download.asp” ‘=== Set File/Path Info
MyPageCounter.Reset(vPageStr) ‘=== Reset counter for ‘/data/
➥download.asp’
MyPageCounter.Reset ‘=== Rest Counter for current page
%>

The Counter Component

The Counter Component plays a role similar to the Page Counter Component, except that the
Counter Component operates not at the individual page level, but at the variable level. In addi-
tion, the Counter Component is used to manage one or multiple counter variables within ASPs.
These variables can symbolize the frequency of any action you may want to keep track of, such
as the number of times a user runs a specific script or has failed to correctly enter information
on a page.
The Counter Component counter variables are different from the Page Counter Component
objects because they do not automatically increment when a page is accessed. The Counter
Component requires explicit actions to increment the individual or multiple counter variables.

You can see that this same functionality could be achieved by using application-level variables
and writing an incremental variable to a text file for storage. However, the Counter Component
has done all that work for you by grouping that functionality into one method.

The Counter Component variables can be updated from any ASP page in the application. For
example, you can use the Counter Component to record the total number of errors that occur
through an application by incrementing a variable, named iErrorCount, whenever an error
occurs. If an error occurs on the login.asp page, the Counter Component can be used to in-
crement the iErrorCount variable. Likewise, if the user fails to complete a specified application
task, the Counter Component can be used to increment another variable, named
iErrorLoanProcess.

To use the Counter Component, two files are needed, as illustrated in Table 17.20. The first file
is the Dynamic Link Library itself that is needed to create the Counter Component functional-
ity. The second file, the Counters.Txt file, is a simple text file that is used as the storage loca-
tion for one or multiple counter variables. The following is an extract of the Counters.txt file:
-------------- Counters.txt --------------
iErrorCount:57
iTimeOut:3
iBadRequest:7
iDownLoads: 382

22.1389-6 CH17 378 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 The Counter Component 379

Table 17.20 The Counter Component Files

File Name Description
Counters.Dll Dynamic Link Library needed to create the counter object.
Counters.Txt The text file used to store the integer values of the counter object.

To initialize the Counter Component, you have to create the Counter Component in the
GLOBAL.ASA as shown below:

<OBJECT RUNAT = Server SCOPE = Application ID = Counter progID = “MSWC.Counters”>

</OBJECT>

After the Counter Component has been created, its four methods, the Get, Increment, Set, and
Remove methods, are used to manage the various counter values (see Table 17.21). Part
IV
Table 17.21 The Counter Component Methods Ch

Name Description 17
Get Returns the count property of the given counter name.
Increment Increments the counter name by one.
Set Assigns the counter name counter to a specific value.
Remove Removes the counter name from the counter object.

The Get Method

The Counter Get method is used to retrieve the value of the counter variable. To extract this
value, use the following syntax:
Counter.Get(CounterName)

where CounterName is the name of the variable that is to be retrieved. To retrieve the current
value of a counter variable, use the Get method as shown in the following example:
<% Response.Write (“<Center>” & “Admin Summary:” & “</Center><BR>”)%>
Timeout Errors: <% = myCounterObj.Get(“iTimeOutCount”) %>
Bad HTTP Request Errors: <% = myCounterObj.Get(“iTimeOutCount”) %>

Over the course of the ASP application, if five timeout errors and two bad HTTP request errors
are produced, the previous code would produce the following results:
Admin Summary:
Timeout Errors: 5
Bad HTTP Request Errors: 2

If the counter variable does not exist, the variable is automatically created and assigned a
value of 0.

22.1389-6 CH17 379 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 380 Chapter 17 Implementing Existing Server-Side Components

Of course, the counter variables do not just have to deal with ASP applications. For example,
you can return the results of a recent Web-based poll that asked a question about co-ed ice
hockey:
Should Co-ed Ice Hockey continue?
94% No
2% Yes
4% Undecided

The Increment Method

The Increment method is used to automatically increment a counter variable. If the variable
called by the Increment method does not exist, the variable is created, its value is set to 0, and
then it is incremented. To implement this method, use the following syntax:
Counter.Increment(CounterName)

where CounterName is the variable to be incremented. This method is used to create a running
total for a variable. For example, the following example instructs the user to seek professional
guidance after failing to complete a task 10 times.
<%
Dim iMaxTries
iMaxTries = 10
Counter.Increment(“UserError”) %>
If Counter.Get(“UserError”) > iMaxTries Then
Response.Write “Stop wasting bandwidth and call the help desk”
End If
%>

The Set Method

The Set method is used assign a specific integer value to a counter variable name. The Set
method requires the following syntax:
Counter.Set(vCounterName, iCountValue)

where vCounterName is the name of the variable to be assigned a value, and iCountValue is the
integer value that is assigned to the variable name. For example, if you want to send an e-mail
to the webmaster after every tenth application error that occurs, the Set method can be used to
reset the counter variable name.
<%
If (myCounterObj(“iErrorCount”) > 10) Then ‘=== Checks how the value of
➥errors
iReturn = SendMail(“ASP Manager”) ‘=== Executes a SendMail
➥Function
myCounterObj.Set(iErrorCount,0) ‘=== Resets error count to 0
End If
%>

22.1389-6 CH17 380 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 From Here… 381

The Remove Method

The Remove Method is used to remove a counter variable from the counter object and the
Counters.txt files. The Remove method requires the following syntax:

Counter.Remove(CounterName)

where CounterName is the counter variable name. For example, to remove the iErrorCount
variable from the Counter object and disregard its value, use the Remove method as shown in
the following example.
<% Counter.Remove(“iErrorCount”) %>

From Here… Part

This chapter has demonstrated the installable ActiveX components that can be used by your
ASP applications. These components represent significant server-side functionality that ex- IV
tends the reach and connectivity of the Internet Information Server that is not available in Ch
native ASP scripts. In particular, this chapter has covered how to use:
17
■ The Advertisement Rotator Component to create dynamic image banners.
■ The Browser Capabilities Component to determine browser characteristics.
■ The Database Access Component to connect to any ODBC-compliant database.
■ The Counter Component to create incremental counter variables.
■ The Page Counter Component to track how many times a page is requested.
■ The Content Rotator Component to automatically rotate or change content on a page.
■ The Tools Component to access additional server-side administration features.
■ The Content Linking Component to provide automatic page linking capabilities.
■ The Permission Checker Component to manage user access permission for your site.

These components can be used individually or collected to add exciting functionality to your
site. The next chapter will cover how to create your own ActiveX components. After you learn
how to create your own components, Chapter 19 will demonstrate how to use the components
covered in this chapter to create the fictitious DataWidget Warehouse Web site. For additional
related information, see the following chapters:

■ Chapter 18, “Creating Your Own Active Server Components,” walks you through the
tools and examples of how to create tailored functionality via custom-created compo-
nents.
■ Chapter 19, “Putting It All Together: Creating an Online Registration System,” demon-
strates how to use the components that ship with IIS 4.0 to build a sample Web site.

22.1389-6 CH17 381 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 382 Chapter 17 Implementing Existing Server-Side Components

■ Chapter 20, “Database Access with ADO,” teaches how to use the Data Access Compo-
nent to connect to any ODBC-compliant database.

■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.

22.1389-6 CH17 382 10/26/98, 9:31 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH17 lp#3
 18
383

C H A P T E R

Create ActiveX Components

Creating Your Own ActiveX components are the build-
ing blocks of server programs. They

Active Server enable you to create functionality

using Visual Basic and call it from
your Active Server Pages.

Components Deal with Datatype Differences

The fact that ASP supports only the
Variant datatype causes some
problems that you must deal with.
Part
Access the ASP Intrinsic
Objects from within an
IV
In Chapter 17, “Implementing Existing Server-Side Compo- ActiveX Component Ch
nents,” you learned how to declare and use the Dynamic You can access the ASP intrinsic 18
Link Libraries that ship with the Internet Information objects from within Visual Basic.
Server. While these components provide great functional-
ity, there are times when no component that you can Wrap Built-in Visual Basic
download or purchase will fit your needs exactly. In these Functions
cases, you will have to create your own component. Luck- If you need a VB function in your
ily, they are not difficult to create if you understand how to program, just wrap it in an ActiveX
program in Visual Basic. Server.

Creating your own components can greatly enhance your Encapsulate Complex
value as an Internet programmer. Components enable you Calculations
to create classes that can do nearly anything that can be If your logic is complex, you may
done on a computer, except create a user interface. Nor- find it easier to program in VB than
mally, programmers create objects in order to accomplish in ASP.
one of the following:
Speed Up Your ASP Program
Learn how to use ActiveX compo-
nents to speed up your applications.

Port Existing Applications

If you have complex calculations
already written in VB, don’t port
them, wrap them.

23.1389-6 CH18 383 10/26/98, 9:35 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 384 Chapter 18 Creating Your Own Active Server Components

■ Gain access to functions that are not supported by the dialect of VBScript that runs in the
Active Server Pages.
■ Create complex calculations that could potentially be reused in many Visual Basic, C++,
and ASP applications. This removes the need to code a special version of the program for
ASP.
■ Speed up applications. If your application contains complex logic, you can move program-
ming logic from the Active Server Pages to the component where it can be compiled.
■ Port code to ASP. Many of the existing Visual Basic programs will be ported to Active
Server Pages applications. Moving the code into components instead of into ASP files
can alleviate much of the pain associated with this.
■ Debug complex algorithms. Microsoft has promised improved debugging facilities for
ASP in future releases of the Internet Information Server, but it will be some time before
it can compete with the Visual Basic’s heavy-weight champion debugger. ■

Creating ActiveX Components

ActiveX components are the next generation of what was previously known as OLE Automation
Servers. They are created in the same fashion as the OLE Servers were in previous releases of
Visual Basic and Visual C++.

In this section, you will create a simple ActiveX component using Visual Basic 5.0, although
ActiveX components can be created using any COM-compliant language such as C++.

This example creates a component that multiplies two numbers and returns the result. The
simplicity of this example will allow us to concentrate on the process of component creation.
The examples that follow will be of increasing complexity.

First start Visual Basic 5.0. The first screen to appear is the New Project dialog box shown in
Figure 18.1.

FIG. 18.1
The New Project dialog
box gives you a choice
of project types to
create.

23.1389-6 CH18 384 10/26/98, 9:35 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Creating ActiveX Components 385

Choose the ActiveX DLL icon. (Visual Basic calls ActiveX components ActiveX DLLs because
the file suffix is .DLL, which stands for Dynamic Link Library.) This action creates a new
project called Project1, which contains a single class called Class1. Change these names to
the following:

■ Project—SimpleProj
■ Class—SimpleClass

Next, save the project by choosing Save from the File menu. At this point, you are ready to add
a function to the class, and thereby to your component. To do this, choose Code from the View
menu. This will display the code window in the center region of your screen.

Listing 18.1 SIMPLECLASS.CLS —A simple function illustrating the use of

ActiveX components.
Function Mult1(X As Variant, Y As Variant) As Variant
Mult1 = X * Y
End Function

This function was chosen for its simplicity, and to maximize the chances that you will be able to Part
get it to work on the first try. It is a good idea to test all ActiveX components in the Visual Basic IV
debugger before using them in Active Server Pages. To perform this test, you need to add
another project. To do this, choose Add Project from the File menu. When the New Project Ch
dialog box appears, choose the Standard EXE icon. This will add a second project to your IDE 18
session. There is no need to rename this project or the form that it creates, as we are only
using it to test our component.

Create a form that looks like the one shown in Figure 18.2.

FIG. 18.2
A test form is needed
to verify that the
component works
properly.

The following names are used in this example:

■ txtX—Text box
■ txtY—Text box
■ txtAnswer—Text box
■ cmdMultiply—Command button

23.1389-6 CH18 385 10/26/98, 9:35 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 386 Chapter 18 Creating Your Own Active Server Components

In order to use the debugger, you must remove the component project by highlighting
SimpleProj in the Project Window and choosing Remove Project from the File menu. You can
immediately add it back by choosing Add Project from the file menu. Finally, choose the Refer-
ences entry from the Project menu and choose the SimpleProj entry. This makes the
SimpleClass available to the test program. This odd procedure makes the project visible to the
debugger.

At this point, you are ready to use the SimpleClass and its method. To do this, enter the code
shown in Listing 18.2.

Listing 18.2 TESTMULT.BAS —The object must be created before its meth-
ods can be used.
Private Sub cmdMultiply_Click()
‘Create the Object
Dim objSimple As New SimpleClass
Dim Num1 As Variant
Dim Num2 As Variant

Num1 = txtX
Num2 = txtY

‘Use the Object to call the Mult1 method

txtAnswer = objSimple.Mult1(Num1, Num2)

End Sub

If you look at the code, you can see that I had to create an instance of the SimpleClass and
assign the name objSimple to the new object. I then used the object’s name to make a call to
the method named Multi. I passed it two Variant variables and it returned the answer as ad-
vertised.

If you look at the SimpleClass in the Object Browser, you can see that it has the same kind of
attributes as any other object, just not as many of them. This is due to the simplicity of the
object rather than to its mode of creation. Choose Object Browser from the View menu to bring
up the browser, as shown in Figure 18.3.

Now that you are armed with a working ActiveX component, it is time to use it in an Active
Server Page. To do so, open WordPad or any other text editor and enter the code shown in
Listing 18.3.

23.1389-6 CH18 386 10/26/98, 9:35 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Creating ActiveX Components 387

FIG. 18.3
The Object Browser can
be used to examine the
interface of an ActiveX
component.

Part
IV
Listing 18.3 TESTMULT.ASP —ActiveX components can be called from
Ch
Active Server Pages.
<HTML>
18
<HEAD>
<TITLE>
ASP calling an ActiveX Component
</TITLE>
</HEAD>
<H3>
<BODY>
<%
X = 10
Y = 15 %>

The value of X: <% = X %> <BR>

The value of Y: <% = Y %> <P>

<%
Set tstObj = Server.CreateObject(“SimpleProj.SimpleClass”) %>

<%
Amt = tstOBJ.Mult1(X,Y)
Response.Write(“The Answer is :” )
Response.Write(Amt)
%>
</BODY>
</HTML>

23.1389-6 CH18 387 10/26/98, 9:35 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 388 Chapter 18 Creating Your Own Active Server Components

Looking at the code, you can see that there are parallels between this code and that in the
Visual Basic example above. In both cases, the object was created and given a name. In the
ASP example, this was done using a method in the Server called CreateObject. This name was
used to provide access to the Mult1() method. Instead of writing the output to a form, the ASP
example used the Response object’s Write() method to send the output to a browser.

FIG. 18.4
The result of the call to
the Mult1() method is
displayed in the
browser.

One of the complications of creating ActiveX components is changing them once they have
been built. If you try to make SimpleProj.DLL again because you found and fixed a bug in it,
you will likely receive an error message that states “Permission Denied”. This is happening
because your Internet server has used this class and Visual Basic thinks that IIS has it
“checked out.” The way around that is to shut down the Internet Information Server. Visual
Basic can then make the DLL.

Take a moment to let the significance of what you have seen demonstrated sink in. This ex-
ample shows that you can create an arbitrary ActiveX component and call it from any Active
Server Page. You can call an Active Server Page from any browser that has permission to ac-
cess your server. You can create an ActiveX component that can do anything that a Visual Basic
application can do, therefore you now have the power to create an application that can do any-
thing that you want it to, and make it available to anyone in the world. This means that you can
program for the Internet using native Visual Basic.

23.1389-6 CH18 388 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Dealing with Datatype Differences 389

Dealing with Datatype Differences

One limiting factor that you saw in the preceding example was that the method received pa-
rameters that were of the Variant datatype. This is the easiest way to create an ActiveX compo-
nent, but not the only way. Suppose that you have an existing application that accepts two pa-
rameters, one a Single and the other an Integer that returns a Single. In this case, you have
two choices:

■ Rewrite the ActiveX components to accept and return Variants.

■ Find a way to call the ActiveX component as it is.

The first option is only available if the component is not being used by other non-Internet appli-
cations. Even then, the work required to recode and debug the server can be considerable.
Clearly the second way is preferred.

It is possible to change the datatype of a parameter by casting its type using one of the func-
tions shown in Table 18.1.

Table 18.1 The Data Conversion Functions

Function Name Returns Variant Subtype Part

CBool Boolean
IV
Ch
CByte Byte

CDate Date 18
CDbl Double

CInt Integer

CLng Long

CSng Single

CStr String

The return type of the function is handled automatically because the assignment is from its
type to a variant in the ASP file. The following example shows how this is done.

Create an ActiveX component. Call the project TypeCast and the class TypeCastClass. Next,
create a method called AddThem(). Add the code shown in Listing 18.4 to this method.

Listing 18.4 TYPECASTCLASS.CLS —This class is expecting to be passed a

Single, an Integer, and returns a Single.
Function Mult2(X As Single, Y As Integer) As Integer
Mult2 = X * Y
End Function

23.1389-6 CH18 389 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 390 Chapter 18 Creating Your Own Active Server Components

Next, you can run the ASP file in Listing 18.5 and see what happens.

Listing 18.5 TESTTYPEMULT.ASP—Trying to call a method with non-variant

datatypes without type casting.
<HTML>
<HEAD>
<TITLE>
ASP calling an ActiveX Component
</TITLE>
<H1>The Data Type Example
</HEAD>
<H3>
<BODY>
<%
X = 10
Y = 15 %>
The value of X: <% = X %> <BR>
The value of Y: <% = Y %> <P>
<%
Set tstObj = Server.CreateObject(“TypeCast.TypeCastClass”) %>
<%
Amt = tstOBJ.Mult2(X,Y)
Response.Write(“The Answer is :” )
Response.Write(Amt)
%>
</BODY>
</HTML>

Because the method Mult2() was written to accept a Single and an Integer parameter, I
would not expect it to work in this example, and it doesn’t, as shown in Figure 18.5.

There is, however, a simple fix to this problem. By type casting the ASP variables to be the
right subtype, the ActiveX component will be able to accept them as shown in the corrected
version of this example, shown in Listing 18.6.

Listing 18.6 TESTTYPEMULT2.ASP—Type casting is the solution to the Type

Mismatch problem.
<HTML>
<HEAD>
<TITLE>
ASP calling an ActiveX Component
</TITLE>
<H1>The Data Type Example
</HEAD>
<H3>
<BODY>
<%
X = 10

23.1389-6 CH18 390 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Dealing with Datatype Differences 391

Y = 15 %>
The value of X: <% = X %> <BR>
The value of Y: <% = Y %> <P>
<%
Set tstObj = Server.CreateObject(“TypeCast.TypeCastClass”) %>
<%
Amt = tstOBJ.Mult2(CSng(X),CInt(Y))
Response.Write(“The Answer is :” )
Response.Write(Amt)
%>
</BODY>
</HTML>

FIG. 18.5
Mixing up datatypes is
a sure way to get this
error message.

Part
IV
Ch

18

Notice in the 18th line that the parameters are being converted to the correct subtype before
being sent to the Mult2() method for processing. This is sufficient to solve the problem, and it
causes the correct answer to appear as was shown in Figure 18.4.

Now we have the ability to call ActiveX component methods regardless of the parameter types
that they expect. This opens up all the existing classes that you have written for use in Visual
Basic applications and makes them callable from within your ASP applications. From this point
forward, you can use ActiveX components to perform a number of useful functions in your
programs. I will not explicitly show the testing of these components in the Visual Basic IDE,
but I encourage you to apply the techniques demonstrated earlier in this chapter to verify the
correctness of your classes before deploying them in an ASP application.

23.1389-6 CH18 391 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 392 Chapter 18 Creating Your Own Active Server Components

N O T E Any ActiveX components, including those created with Visual C++, can be used in Active
Server Pages. ■

Wrapping Built-In Visual Basic Functions

At this point, you know how to create ActiveX components from scratch, and how to overcome
datatype conflicts between ASP applications and components that expect parameters of types
other than Variant. Now you are in a position to do some really useful work with these compo-
nents.

One of the addictions that all Visual Basic programmers share is to the built-in utility functions
that ship with the product. Some of the most common of these functions are made available to
ASP programmers by the Internet Information Server, but many useful ones are not. In order
to provide access to these functions, you must write an ActiveX component that provides an
interface to the ASP application. Because the ActiveX component is really a Visual Basic pro-
gram, all built-in functions are available to it. The process of encapsulating functionality in this
manner is called wrapping, and the components that are produced are called wrappers. This
name is derived from the analogy of a piece of candy being wrapped in a protective covering
made of paper or cellophane.

Let’s work an example to illustrate this approach. The Format() function in Visual Basic pro-
vides a number of useful conversions concerning time, dates, and format of data. If I want to
provide this capability to my programs, then I must wrap this function in an ActiveX compo-
nent, as it is not available to ASP applications normally.

To create this wrapper, create an ActiveX DLL project exactly as you did above. Name the
project ConvertIt and the class Formats. Next, write code to add several methods to the class,
as shown in Listing 18.7.

Listing 18.7 FORMATS.CLS—The methods in this class wrap the Format()

function to provide its functionality to ASP applications.
‘This method accepts time in AM and PM format then converts it to Military Time.
Public Function Militarytime(Intime As Variant) As Variant
Militarytime = Format(Intime, “hh:mm:ss”)
End Function

‘This method accepts time in military format and converts it to AM and PM.
Public Function AMPM(Intime As Variant) As Variant
AMPM = Format(Intime, “h:mm:ss AMPM”)
End Function
‘This method converts a terse date into a day of the week, month, day, year
➥format.
Public Function FullDate(InDate As Variant) As Variant
FullDate = Format(InDate, “dddd,mmm d yyyy”)
End Function

23.1389-6 CH18 392 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Wrapping Built-In Visual Basic Functions 393

‘This method capitalizes all letters in a string that it is passed

Public Function AllCaps(InString As Variant) As Variant
AllCaps = Format(InString, “>”)
End Function

The following methods make up this class:

■ MilitaryTime()—Accepts time in AM and PM format, and then converts it to military

time.
■ AMPM()—Accepts time in military format and converts it to AM and PM.
■ FullDate()—Converts a terse date into a day of the week, month, day, year format.
■ AllCaps()—Capitalizes all letters in a string that it is passed.
These functions are contained in the same object because all of them are based on the
Format() built-in function. Packaging of methods in ActiveX components is a trade-off between
simplicity in coding and program size. If you put every function that you would ever need in
one class, then every ASP application could just perform one Server.CreateObject() method
and access everything that they would ever need. On the other hand, that would create a mas-
sive DLL that would have to be uploaded to every Web server that wanted to run the ASP
application. In addition, the load size of the DLL could make the working set of the program Part
very large and take resources away from other programs running on the same Web server. IV
The DLL created here is tiny, so it is not a problem. Normally, this size problem will only have
Ch
an impact when dozens of fairly large functions are created.

Using these methods is simply a matter of declaring the class and calling them from your ASP
18
file, as shown in Listing 18.8.

Listing 18.8 CONVERTITTEST1.ASP—You can call the new utility functions

by creating an object of that class type.
<HTML>
<H1> Testing the ConvertIT ActiveX Component
<H3>
<%
Set MyFormats = Server.CreateObject(“ConvertIt.Formats”)
OutTime1 = MyFormats.MilitaryTime(“5:13:14 PM”)
OutTime2 = MyFormats.AMPM(“17:15”)
OutDate = MyFormats.FullDate(“7/3/97”)
OutString = MyFormats.AllCaps(“Hello,Planet”)
%>
Input = 5:13:14 PM
Output =
<% = OutTime1 %><BR><BR>
Input = 17:15
Output =
<% = OutTime2 %><BR><BR>
Input = 7/3/97

continues

23.1389-6 CH18 393 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 394 Chapter 18 Creating Your Own Active Server Components

Listing 18.8 Continued

Output =
<% = OutDate %><BR><BR>
Input = Hello, Planet
Output =
<% = OutString %>
</HTML>

Running this file gives the results shown in Figure 18.6.

FIG. 18.6
All of the component’s
methods are now
available to you inside
the ASP file.

Proving that this technique works will open up more programming doors for you. Not only do
you now have the entire library of built-in Visual Basic functions available to your program, you
also have any other module available by this same technique.

Encapsulating Complex Calculations

There are times when you are performing complex calculations that require considerable
debugging. In addition, these calculations are needed by non-ASP applications as well. In these
cases, ActiveX components provide the dual-purpose capability that you need.

In addition, any calculation that has been placed in a DLL is compiled and might result in a
performance advantage over calling the same function from within an ASP file. You cannot
assume that this is so in all cases because of the overhead associated with calling a DLL versus
calling a function in the same ASP file.

23.1389-6 CH18 394 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Encapsulating Complex Calculations 395

An example of this kind of calculation is the net present value calculation that is used to calcu-
late the cost of breaking a lease early. This calculation takes into account the fact that many
leases have a built-in formula for rent increases in future years. Additionally, it is customary for
the lease to provide a discount allowing for the time value of money. This means that the com-
pany renting the building will be given a discount over the full value of all future rent to reflect
the fact that they are paying off the lease all at once, instead of slowly over time.

To perform this calculation, create an ActiveX component called Financials and a class called
Formulas. This class contains only one method, called CalcNPV(). This method will accept a
maximum of three rent increases over term of the lease. Table 18.2 shows the input parameters
for this method.

Table 18.2 The Inputs to the CalcNPV Function

Parameter Datatype Meaning
NowPayment Variant The current rent amount
LeaseExp Variant The date that the lease expires
YrPctInterest Variant The yearly interest rate for calculating the discount
Part
ReDate1 Variant The first date that the rent will go up
IV
ReAmount1 Variant The amount of the rent after the first date
Ch
ReDate2 Variant The second date that the rent will go up
18
ReAmount2 Variant The amount of the rent after the second date
ReDate3 Variant The third date that the rent will go up
ReAmount3 Variant The amount of the rent after the third date

Notice that all the parameters are of the Variant datatype. This allows us to avoid type conver-
sions. The code for this calculation is shown in Listing 18.9.

Listing 18.9 FORMULAS.CLS—The net present value calculation method.

Public Function CalcNPV(NowPayment As Variant, LeaseExp As Variant,
➥YrPctInterest As Variant, ReDate1 As Variant, ReAmount1 As Variant, ReDate2 As
➥Variant, ReAmount2 As Variant, ReDate3 As Variant, ReAmount3 As Variant) As
➥Variant
Dim Today
Dim PmtArray() ‘ Declare dynamic array.
Dim Months
Dim I
Dim ThatMonth
Dim pctInterest
Dim TempValue

continues

23.1389-6 CH18 395 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 396 Chapter 18 Creating Your Own Active Server Components

Listing 18.9 Continued

Today = Now()

‘Find out how many months until the lease expires

Months = DateDiff(“m”, Today, LeaseExp)
If Months < 1 Then
CalcNPV = 0
Exit Function
End If

‘Deal with the fact that some leases have less than 3 rent increases
If IsNull(ReDate1) Then
ReDate1 = LeaseExp
End If

If IsNull(ReDate2) Then
ReDate2 = LeaseExp
End If

If IsNull(ReDate3) Then
ReDate3 = LeaseExp
End If

ReDim PmtArray(Months) ‘ Allocate elements.

ThatMonth = Today

‘fill in the pmtarray with the right rent for each month through lease exp.
For I = 1 To Months
Select Case ThatMonth

Case Is < ReDate1 ‘first escalation date hasn’t been reached

PmtArray(I) = NowPayment ‘today’s fixed rent
Case Is < ReDate2
PmtArray(I) = ReAmount1
Case Is < ReDate3
PmtArray(I) = ReAmount2
Case Else
PmtArray(I) = ReAmount3
End Select
‘ThatMonth is the month that is being represented in the array
ThatMonth = DateAdd(“m”, 1, ThatMonth)
Next I

TempValue = 0
pctInterest = YrPctInterest / 100 / 12

‘Apply the discount

For I = 1 To Months
TempValue = TempValue + (PmtArray(I) / (1 + pctInterest) ^ I)
Next I
CalcNPV = TempValue
End Function

23.1389-6 CH18 396 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Encapsulating Complex Calculations 397

The calculation is somewhat complex due mainly to the rent increases. Once the rent that
would normally be due is loaded into the PmtArray, the discount is applied to each payment.
The payments that are far in the future are discounted more than the payments that are due
sooner.

You can use this class and method from either a Visual Basic or an ASP application. Listing
18.10 shows this object being used to perform 1,000 calculations.

Listing 18.10 TESTNPV.ASP—Testing the execution speed of a calculation

in a DLL.
<%@ LANGUAGE = VBScript %>
<HTML>
<HEAD>
<TITLE>PropertyWorks</TITLE>
</HEAD>
<H1> Testing the speed of Complex Calculations
<H3>
<%
‘Create the Object
Set tstNPV = Server.CreateObject(“Financials.Formulas”)
Part
‘Assign some dummy values
NowPayment = 1000
IV
LeaseExp = “12/31/2005” Ch
YrPctInterest = 12
18
Redate1 = “1/1/2000”
ReAmount1 = 1100
Redate2 = “1/1/2002”
ReAmount2 = 1200
Redate3 = “1/1/2004”
ReAmount3 = 1300

‘Perform the call 1000 times to measure the speed

For i=1 to 1000
Response.Write(i& “,”)
NPVAmt = tstNPV.CalcNPV(NowPayment, LeaseExp, YrPctInterest, ReDate1,
➥ReAmount1, ReDate2, ReAmount2, ReDate3, ReAmount3 )
Next

%>
<BR>
<%= FormatCurrency(NPVAmt) %>
</BODY>
</HTML>

Running this code shows the output in Figure 18.7 in approximately 38 seconds on my
Pentium laptop.

23.1389-6 CH18 397 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 398 Chapter 18 Creating Your Own Active Server Components

FIG. 18.7
The ActiveX component
version of this
application ran in
approximately 38
seconds.

The question naturally arises as to whether this is slower or faster than running the same
calculation from within the ASP file. In order to test this, the code was ported to the ASP file. In
the following section I will port the code to ASP and test the execution speed.

Porting Existing Applications

It would be wonderful if the creators of VBScript and those of Visual Basic Professional Edition
communicated so that any differences in the language would be intentional and based on the
differing missions of the two products. Unfortunately, this is not the case. When porting from
Visual Basic to VBScript, you will find more tiny differences in the capabilities and syntax of
the languages than can be explained by common sense.

There are two kinds of changes that you have to make to port your code to VBScript and run it
in an ASP file. The first kind makes sense and is based on the fact that all variables are Variant
in ASP files. This means that all parameters accept no explicit typing, even as Variant, and the
function return type is missing. It would be better if VBScript checked to see if the return type
and parameter types were Variant before objecting to them, but life is not fair, so it doesn’t.
This means that you will have to remove them by hand.

The second type of change will have to be made for no reason other than the fact that your
program won’t run if you don’t. These changes are the most annoying of all because they cause
you to debug code that is already working. Listing 18.11 shows a version of the TestNPV.asp
program called TestASPNPV.asp. It contains the CalcNPV routine that has been ported to be run
as 100 percent VBScript.

23.1389-6 CH18 398 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Porting Existing Applications 399

Listing 18.11 TESTASPNPV.ASP—The CalcNPV calculation can be coded

entirely in the ASP file.
<%@ LANGUAGE = VBScript %>
<HTML>
<HEAD>
<TITLE>PropertyWorks</TITLE>
</HEAD>
<H1> Testing the speed of ASP Calculations
<H3>
<%
Function CalcNPV(NowPayment, LeaseExp,YrPctInterest, ReDate1 , ReAmount1 ,
➥ReDate2, ReAmount2, ReDate3, ReAmount3)
Dim Today
Dim PmtArray() ‘ Declare dynamic array.
Dim Months
Dim I
Dim ThatMonth
Dim pctInterest
Dim TempValue

Today = Now()

‘Find out how many months until the lease expires Part
Months = DateDiff(“m”, Today, LeaseExp)
If Months < 1 Then
IV
CalcNPV = 0 Ch
Exit Function
End If 18
‘Deal with the fact that some leases have less than 3 rent increases
If IsNull(ReDate1) Then
ReDate1 = LeaseExp
End If

If IsNull(ReDate2) Then
ReDate2 = LeaseExp
End If

If IsNull(ReDate3) Then
ReDate3 = LeaseExp
End If

ReDim PmtArray(Months) ‘ Allocate elements.

ThatMonth = Today

‘fill in the pmtarray with the right rent for each month through lease exp.
For I = 1 To Months

PmtArray(I) = ReAmount3
If ThatMonth < ReDate3 Then
PmtArray(I) = ReAmount2
End If

continues

23.1389-6 CH18 399 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 400 Chapter 18 Creating Your Own Active Server Components

Listing 18.11 Continued

If ThatMonth < ReDate2 Then
PmtArray(I) = ReAmount1
End If
If ThatMonth < ReDate1 Then ‘first escalation date hasn’t been reached
PmtArray(I) = NowPayment ‘today’s fixed rent
End If

ThatMonth = DateAdd(“m”, 1, ThatMonth)

Next

TempValue = 0
pctInterest = YrPctInterest / 100 / 12

‘Apply the discount

For I = 1 To Months
TempValue = TempValue + (PmtArray(I) / (1 + pctInterest) ^ I)
Next
CalcNPV = TempValue
End Function
NowPayment = 1000
LeaseExp = “12/31/2005”
YrPctInterest = 12
Redate1 = “1/1/2000”
ReAmount1 = 1100
Redate2 = “1/1/2002”
ReAmount2 = 1200
Redate3 = “1/1/2004”
ReAmount3 = 1300
For i=1 to 1000
Response.Write(i& “,”)
NPVAmt = CalcNPV(NowPayment, LeaseExp, YrPctInterest, ReDate1, ReAmount1,
➥ReDate2, ReAmount2, ReDate3, ReAmount3 )
Next
%>
<BR>
<%= FormatCurrency(NPVAmt) %>
</BODY>
</HTML>

As you can see, the type declarations have been removed from the parameters and from the
function itself. Next, the Select Case statement (around the 48th line) has been removed and
replaced with a very strange set of If...Then...Else logic. This is because the ASP processor
objected to both the Case statement and the equivalent If...Else If logic. In order to get this
to work, that part of the algorithm had to be replaced. Finally, when it runs, it produces the
output shown in Figure 18.8 below.

Notice that there is a $.03 difference between these two calculations. This can be attributed to
the fact that the code in the ASP case was being processed by ASP.DLL, which is being run by
IIS. The Visual Basic version was compiled by the VB compiler, which is not only a different
piece of software, but also a different kind of software—a compiler. Clearly, the rounding algo-
rithms are slightly different between the DLL and the compiler.

23.1389-6 CH18 400 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Accessing the ASP Intrinsic Objects from Within an ActiveX Component 401

FIG. 18.8
The ASP version of the
CalcNPV processing
produced a slightly
different answer.

Part

The speed of execution issue produced a mild surprise. The ASP version of the application only
IV
took 22 seconds to execute. This is probably because the overhead associated with making the Ch
call to the class was greater than the speed advantage gained by performing the calculation 18
when compiled. The moral of the story has two parts:

■ Don’t rush off to create all of your calculations in ActiveX components. They may be
slower than the ASP version for processing up to a certain level of complexity.
■ Avoid porting existing Visual Basic code to ASP if possible. Make an ActiveX component
out of it instead. The difficulty encountered in performing the port may outweigh any
advantage gained.
For example, a debugged piece of code may not port as written. In this case, create an ActiveX
component out of it. On the other hand, as shown above, performance improvement is not
guaranteed.

Accessing the ASP Intrinsic Objects from Within an

ActiveX Component
Another very intelligent feature of the Active Server Pages design is the fact that you can ac-
cess the five intrinsic objects (Server, Request, Response, Application, and Session) from
within an ActiveX component. This has the potential to improve the performance of your appli-
cation by allowing you to perform almost any task for either the ASP file or the ActiveX compo-
nent that it calls. The following process is used to accomplish this:

23.1389-6 CH18 401 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 402 Chapter 18 Creating Your Own Active Server Components

1. Create a new ActiveX DLL in Visual Basic 5.0.

2. Add a reference to the Microsoft Active Server Pages 1.0 Object Library.
3. Create two page-level event procedures within your component: OnStartPage and
OnEndPage.

4. Add the following code to your component’s general declaration section:

Dim rp as Response
Dim rq as Request
Dim ap as Application
Dim sr as Server
Dim sn as Session

This will give your program access to the objects through these variables.
5. Add the following code to the OnStartPage() event procedure:
Public Sub OnStartPage(mySC as ScriptingContext)
Set rp as mySc.Response
Set rq as mySc.Request
Set ap as mySc.Application
Set sr as mySc.Server
Set sn as mySc.Session
End Sub

This will connect the variables to the objects.

6. Create your application and use the objects declared in the OnStartPage() event
procedure.

In order to understand how this works, here’s a simple example. Create a new ActiveX compo-
nent; call the project SCProj and the class SCClass. Add the code shown in Listing 18.12 to the
class.

Listing 18.12 SCCLASS.CLS—The SC class uses the scripting context to

access the intrinsic object.
Dim rp as Response
Dim rq as Request
Dim ap as Application
Dim sr as Server
Dim sn as Session

‘This routine is called when the component is used

Public Sub OnStartPage(mySC As ScriptingContext)
Set rp as mySc.Response
Set rq as mySc.Request
Set as as mySc.Application
Set sr as mySc.Server
Set sn as mySc.Session
rp.Write “<I> Been to StartPage</I><P>”
End Sub

23.1389-6 CH18 402 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Accessing the ASP Intrinsic Objects from Within an ActiveX Component 403

‘This routine runs when the application is done.

Public Sub OnEndPage()
rp.Write “<P><I>Been to EndPage</I>”
Set rp = Nothing
Set rq = Nothing
End Sub

‘ This subroutine echoes back the user’s input

Public Sub Echo()
UserInput = rq.Form(“AnyNumber”)
rp.Write “<H3>Visual Basic Echo Method Called<BR></H3>”
rp.Write “<H3>You entered “ & UserInput & “<BR></H3>”
End Sub

This component uses the Request object to find out what the user entered in the Input box on
the form. It then uses the Response object to echo that number back to the browser along with
several lines of code that trace the execution. The ASP file shown in Listing 18.13 is called from
an HTML file.

Listing 18.13 ECHO.ASP—This ASP file calls the Echo() method.

<HTML>
Part
<HEAD><TITLE>Echo Tester</TITLE></HEAD>
<BODY>
IV
<% Ch
Set MyCompute = Server.CreateObject(“SCProj.SCClass”)
MyCompute.Echo 18
%>
</BODY>
</HTML>

Notice that this code doesn’t touch either the input or output at any time. It merely serves as a
mechanism to call the Echo() method in the SCClass class.

The final piece of the puzzle is the HTML file (see Listing 18.14).

Listing 18.14 TESTECHO.HTM—This HTML gathers user input to be passed

to the Echo ASP application
<HTML>
<HEAD><TITLE>Echo Tester</TITLE></HEAD>
<BODY>
Please provide a number to echo
<FORM METHOD=POST ACTION = http://default/echo.asp>
Your Number: <INPUT NAME=AnyNumber SIZE=25><P>
<HR><INPUT TYPE=SUBMIT><INPUT TYPE=RESET><P>
</FORM>
</BODY>
</HTML>

23.1389-6 CH18 403 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 404 Chapter 18 Creating Your Own Active Server Components

Running the TestEcho.htm file in the browser brings up the screen shown in Figure 18.9.

FIG. 18.9
The input from the
screen will be passed to
the ActiveX component
by an ASP file.

The result of running this program is shown in Figure 18.10.

FIG. 18.10
The Echo program
displays the input back
to the user in the
browser.

23.1389-6 CH18 404 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Speeding Up Your ASP Program 405

This technology allows you to use the Active Server Page in a sort of “pass thru,” which allows
an HTML file to call a Visual Basic program indirectly.

Speeding Up Your ASP Program

Let’s see if you can use this access to intrinsic objects from within Visual Basic to speed up a
program. To do this, you will measure the program running in its normal mode as an ASP
application, and then convert it to an ActiveX component that uses the intrinsic objects. You will
then measure the difference in wall clock time to execute.

The application that we will use retrieves records from the BIBLIO.MDB database (which ships
with VB 5) using the ActiveX Data Object (ADO). This database is normally found in the c:\VB
directory. It does some non-trivial formatting and writes its results to the browser. Listing 18.15
shows the ASP version of this application.

Listing 18.15 DBASP.ASP—Accessing a database from an ASP application.

<%@ LANGUAGE = VBScript %>
<HTML>
<HEAD> Part
<TITLE>Biblio Author Listing</TITLE>
</HEAD>
IV
Author Listing Ch
<HR>
<%
Set OBJdbConnection = Server.CreateObject(“ADODB.Connection”)
18
OBJdbConnection.Open “Biblio”
SQLQuery = “Select * From Authors Where Au_ID < 500”
Set RSAuthorList = OBJdbConnection.Execute(SQLQuery)
%>
<TABLE COLSPAN=8 CELLPADDING=5 BORDER=0>
<!-- BEGIN column header row -->
<TR>
<TD>
Author ID
</TD>
<TD>
Author Name
</TD>
<TD>
Year Born
</TD>
</TR>
<!-- Put the data in the table -->
<% Do While Not RSAuthorList.EOF %>
<TR>
<TD>
<%= RSAuthorList(“Au_ID”)%>
</TD>

continues

23.1389-6 CH18 405 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 406 Chapter 18 Creating Your Own Active Server Components

Listing 18.15 Continued

<TD>
<%= RSAuthorList(“Author”)%>
</TD>
<TD>
<%= RSAuthorList(“Year Born”)%>
</TD>
</TR>
<!-- Next Row -->
<%
RSAuthorList.MoveNext
Loop
%>
</TABLE>
</BODY>
</HTML>

This application takes approximately 11 seconds to complete. It looks like Figure 18.11 below.

FIG. 18.11
The no-frills database
example takes 11
seconds to complete.

Now, let’s look at an ActiveX component version of this same application. In order do this,
create an ActiveX DLL in Visual Basic. Call the project ADOProj and the class ADOLookup. Next,
create a method called LookUp and add the code shown in Listing 18.16 to it.

23.1389-6 CH18 406 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 Speeding Up Your ASP Program 407

Listing 18.16 ADOLOOKUP.CLS—This is an ActiveX component version of

the DBASP.
Dim rp As Response
Dim rq As Request

Public Sub OnStartPage(mySC As ScriptingContext)

Set rp = mySC.Response
Set rq = mySC.Request
End Sub

Public Sub OnEndPage()

Set rp = Nothing
Set rq = Nothing
End Sub

Public Sub Lookup()

rp.Write “ “
rp.Write “ “
rp.Write “<HTML>”
rp.Write “<HEAD>”
rp.Write “<TITLE>Biblio Author Listing</TITLE>”
rp.Write “</HEAD>”
rp.Write “Author Listing” Part
rp.Write “<HR>” IV
Dim RSAuthorList As New ADODB.Recordset Ch
RSAuthorList.ActiveConnection = “DSN=Biblio”
RSAuthorList.Open “Select * From Authors Where AU_ID < 500” 18
rp.Write “<TABLE COLSPAN=8 CELLPADDING=5 BORDER=0>”
‘ BEGIN column header row
rp.Write “<TR>”
rp.Write “<TD>”
rp.Write “Author ID”
rp.Write “</TD>”
rp.Write “<TD>”
rp.Write “Author Name”
rp.Write “</TD>”
rp.Write “<TD>”
rp.Write “Year Born”
rp.Write “</TD>”
rp.Write “</TR>”
‘ Put the data in the table
Do While Not RSAuthorList.EOF
rp.Write “<TR>”
rp.Write “<TD>”
rp.Write RSAuthorList(“Au_ID”)
rp.Write “</TD>”
rp.Write “<TD>”
rp.Write RSAuthorList(“Author”)
rp.Write “</TD>”
rp.Write “<TD>”

continues

23.1389-6 CH18 407 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 408 Chapter 18 Creating Your Own Active Server Components

Listing 18.16 Continued

rp.Write RSAuthorList(“Year Born”)
rp.Write “</TD>”
rp.Write “</TR>”
‘ Next Row
RSAuthorList.MoveNext
Loop
rp.Write “</TABLE>”
rp.Write “</BODY>”
rp.Write “</HTML>”
RSAuthorList.Close
End Sub

Running the application this way requires that you create a simple ASP file like that shown in
Listing 18.17, even though all the work is done inside the ActiveX component.

Listing 18.17 ADODBTEST.ASP—An ASP file is required even if all of the

work is done inside an ActiveX component.
<HTML>
<HEAD><TITLE>ADODB Tester</TITLE></HEAD>
<BODY>
<%
Set MyCompute = Server.CreateObject(“ADOProj.ADOLookup”)
MyCompute.Lookup
%>
</BODY>
</HTML>

This version of the application ran in nine seconds, which is slightly faster than the same appli-
cation written entirely in VBScript. This will not be universally true, especially for small applica-
tions.

From Here…
In this chapter, you learned how to use ActiveX components to improve your programs. You
saw how these servers can help in porting existing applications to the Internet. You also saw
how ActiveX components can improve the speed of execution of your applications. The follow-
ing chapters contain related topics:

■ Chapter 19, “Putting It All Together: Creating an Online Registration System,” demon-
strates how to use the components that ship with IIS 4.0 in building a sample Web site.
■ Chapter 20, “ Database Access with ADO,” teaches how to use the Data Access Compo-
nent to connect to any ODBC-compliant database.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.

23.1389-6 CH18 408 10/26/98, 9:36 AM

Brand03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH18 lp#3
 19
409

C H A P T E R

Using the Content Rotator and

Putting It All Together: the Ad Rotator
Implement the Content Rotator to

Creating an Online vary information displayed at your

site and discover how to use the Ad
Rotator to include a rotating image

Registration System banner in your site.

Determining the Browser Type

Learn how to use the Browser
Capability Component to identify
and utilize different browser capa-
bilities.

This chapter will use some of the installable components Tools Component and the
Page Component
covered in Chapter 17, “Implementing Existing Server-
Side Components,” to build an online product registration Prevent broken links and errors in
wizard for a company Web site. In this sample application, application logic by using the Tools Part
a fictitious company called Amazing DataWidgets Component, and use the Page
WareHouse Inc. has created a server-side ActiveX design- Component to display the number
IV
of times a page has been accessed. Ch
time control for Visual InterDev that is capable of generat-
ing Active Server Page code by reading your mind! To Implementing the Content 19
better market its product, the company is giving away free Linker Component
beta copies of the control from their site. This company
Use the Content Linker to create a
wants to continuously change information that is delivered
programmatic application workflow
to the browser. To achieve this effect, the application process.
automatically varies the product’s accolades and the mar-
keting information in different banners at the bottom of
the page, as seen in Figure 19.1. ■

24.1389-6 CH19 409 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 410 Chapter 19 Putting It All Together: Creating an Online Registration System

FIG. 19.1
This Amazing
DataWidgets WareHouse
Web site offers changing
content as well as
changing product
banners.

The site also uses scrolling effects to display a disclaimer across the bottom of the page that
varies depending on the browser type. If the browser is the Internet Explorer, the Marquee
control is used. If the browser is Netscape Navigator, a scrolling text box is used, as shown in
Figure 19.2.

FIG. 19.2
This site also formats
output for different
browser types.

24.1389-6 CH19 410 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Rotating the Content with the Content Rotator Component 411

The site uses various components to make these features possible: The Browser Component is
used to determine the type of browser that is accessing the site; the Ad Rotator Component is
used to display product information at the bottom of the page; the Content Rotator Component
is used to vary information that is displayed to the user; the Content Linker Component is used
to create an online registration wizard; the Page Counter Component tracks the page access for
the site; and the Tools Component is used to verify that files exist on the server before they are
referenced. In addition, client-side JavaScript is implemented throughout the site to enhance
the functionality delivered by the Active Server Page components.

Rotating the Content with the Content Rotator

Component
One of the easiest ways to create the appearance of a dynamic site is to periodically change the
information on the Web pages. However, this task of updating the Web site is no small job. One
of the easiest ways to minimize the maintenance effort needed to maintain a dynamic site like
this is to rotate the content. There are definite advantages and disadvantages to using the rotat-
ing approach when it comes to content. Although changing the content of your site makes it
appear that you are always updating the information, the users must be able to navigate back to
the site to find the information that they originally found at your site. If they can’t, your site will
lose it effectiveness and frustrate the users.

The Amazing DataWidgets WareHouse Web site features many different praises for their
MIND-bender design-time control. To create a dynamic effect, the site will rotate these differ-
ent accolades. In this situation, the first two items describing the product change every time
Part
you load the page, as shown in Figures 19.1 and 19.2. This capability to rotate the content is
made possible by the Content Rotator Component. The Content Rotator Component reads IV
content from a scheduling file, and the scheduling file determines the content that is displayed Ch
and the frequency in which it is displayed.
19
To begin this project, start your Active Server Page by creating the template for the first page.
As discussed in Chapter 8, “Putting It All Together with VBScript,” you can create and maintain
your Active Server Pages by hand by using an ASCII editor, or you can use a development
environment such as Visual InterDev. The easiest way to minimize the amount of time writing
and testing HTML code is to use a GUI HTML editor, such as FrontPage or HotMetaL. Using
these tools enables you to quickly generate visual templates for placement of information. This
method also enables the new members of the Web development team, such as the graphic
artists and content creators, to have direct input in sharing the workload. Use an HTML editor
to create the title, opening paragraph, and a two-column, three-row border-less table, as shown
in Figure 19.3.

24.1389-6 CH19 411 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 412 Chapter 19 Putting It All Together: Creating an Online Registration System

FIG. 19.3
Use a GUI HTML editor
to generate a visual
template to place your
Active Server code.

To help determine where to put your code modules when you are in an editor coding Active
Server Pages, place comments to yourself to help identify where the Active scripts should be
used. This example will use the second column of each row to display the rotating content. To
place a mental bookmarker where to add the code, type Rotating Content Line 1. Now, save
the ASP file, and re-open it in your editor of choice. Find the location of the text Rotating Con-
tent Line 1 to determine where to write the first instance of the dynamic content.

The Content Rotator, as discussed in Chapter 17, “Implementing Existing Server-Side Compo-
nents,” extracts content and scheduling information from a text file. This text file can contain
HTML tags that are transported by the Content Rotator to the Web page. To create an instance
of the Content Rotator Component, use the Server.CreateObject statement and reference the
appropriate content scheduling file, as shown in Listing 19.1.

You can also separate the CreateObject tag so it is located outside of the table to make things
look neater, as shown in Listing 19.1. Remember, you want to batch process multiple lines of
Active Server scripts to improve performance.

Listing 19.1 Embedding the Content Rotator in an HTML table.

<%
Dim NextTip
Set NextTip = Server.CreateObject(“MSWC.ContentRotator”)
%>
<TABLE border=”0" cellpadding=”2" width=”60%”>
<TR>
<%

24.1389-6 CH19 412 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Rotating the Content with the Content Rotator Component 413

Response.Write NextTip.ChooseContent(“acolades.txt”)
%>
</TR>
<TR>
<%
Response.Write NextTip.ChooseContent(“acolades.txt”)
%>
</TR>
<TD align=”center”>
<img src = “images/TN00605A.gif” width=”77" height=”69"></TD>
<TD><FONT FACE=”Arial”>Make the world spin a little
faster - </FONT><A Href = “userprofile1.asp”><FONT
FACE = “Arial”>Reserve</font></A><FONT FACE = “Arial”> your
copy of the MIND-Bender ActiveX control today.</FONT></TD>
</TR>
</TABLE>

In this example, the Content Rotating object uses information from the content scheduling file
named acolades.txt by using the NextTip method. The Content scheduling file enables you to
specify the relative weight or frequency that the content item will be displayed and the content
itself, as shown in Listing 19.2.

Listing 19.2 The Content Rotator scheduling file.

%% #1 // Money Bags CEO
<TD align=”center”><font face=”Arial”>
<img src=”/ASPstore/images/bs00044a.gif” width=”67"
height=”67"></font></TD><TD><font
face=”Arial”>&quot;We saved 2.4 million dollars by firing all our Part
application developers and replacing then with the sales
staff!&quot; -- Another Wealthy CEO</font></TD> IV
Ch
%% #1 // Java Quote
<TD align=”center”><font face=”Arial”><img
src=”/ASPstore/images/PE03328A.gif” width=”76"
19
height=”73"> </font></TD>
<TD><font face=”Arial”>&quot;Simply Amazing&quot; -- The
Caffeine Free Press</font></TD>

%% #1 // Developer
<TD align=”center”><font face=”Arial”>
<img src=”/ASPstore/images/bs00044a.gif” width=”67"
height=”67"></font></TD><TD><font
face=”Arial”>&quot;Now I don’t have to worry about practicing
any social skills!&quot; -- Another Heads-Down Coder</font></TD>

In this situation, all the weights are equal to provide an even level of content distribution to
requesting browsers. When a browser requests the calling page, the text is translated directly
to the server and is displayed to the client as if the page were generated entirely by hand. Also,
notice that the scheduling file contains the table definition cells (<TD> and </TD>). As a result,
the call to the NextTip.ChooseContent method is called from the table row (<TR>) definition.

24.1389-6 CH19 413 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 414 Chapter 19 Putting It All Together: Creating an Online Registration System

Rotating the Banner with the Ad Rotator Component

The same concept used in the Content Rotator can be applied to rotating images on a Web page
with the Ad Rotator Component. The Ad Rotator Component can be used to rotate images to
promote your own products and services, sell advertising space, or simply to rotate and display
images. The Ad Rotator Component also uses a scheduling file to determine the image that is
going to be displayed and the frequency with which it will be displayed. The Ad Rotator Com-
ponent uses the GetAdvertisement method to read the specified scheduling file, as shown in
Listing 19.3.

Listing 19.3 Creating a rotating image banner.

<%
Dim ScheduleAD, adrot
ScheduleAD = “datawidgetswarehouse.txt”
Set adrot = Server.CreateObject(“MSWC.AdRotator”)
Response.Write adrot.GetAdvertisement(ScheduleAD)
%>

The Ad Rotator scheduling file enables you to set the properties of the displayed image (see
Listing 19.4).

Listing 19.4 The Ad Rotator scheduling file.

REDIRECT GoTo.ASP
WIDTH 468
HEIGHT 60
BORDER 1
*
/aspstore/images/banner1.gif
/aspstore/userprofile1.asp
The Active Server Component the developers didn’t tell you about!
50
/aspstore/images/banner2.gif
/aspstore/userprofile1.asp
Sign up for the Ultimate Design-Time Control
50

The REDIRECT options located on the first line in the Listing 19.4 specify a redirection file that is
used to route the user to the proper destination. The redirection file is used as a counting
mechanism to track the effectiveness of the banner. This file provides you the flexibility to keep
track of your banners and the number of times that a banner is clicked. The redirection file, in
this situation named GoTo.ASP, is an ASP file that you create to route the user to the appropriate
URL parameter, as shown in the following code.
<%@ LANGUAGE=”VBSCRIPT” %>
<% Response.Redirect (Request.QueryString(“URL”)) %>

24.1389-6 CH19 414 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Verifying that a File Exists with the Tools Component 415

This file uses the Request.Redirect method to redirect the browser to the designated URL.
The round-about method of directing the user to the final destination is used to measure the
effectiveness of an advertisement banner. Now you can build extensive logging capabilities into
this ASP file that can be used to measure and record the amount of hits on the REDIRECT file.

Verifying that a File Exists with the Tools Component

In simple terms, the Web server is just a file server that sends files to the client’s browser. The
browser receives the files and translates the ASCII text into graphical output. When images
cannot be found or links are broken, the Web server skips over the missing files and delivers
the remaining information until it stops looking for the missing file. However, if required files
are missing from server-side components, the Active Server scripting engine will produce a
runtime error and terminate the processing of any more information on the page. For example,
if you misspell the name of the Content Rotator file, the following runtime error is produced:
MSWC.ContentRotator.1 error ‘80070002’
The system cannot find the file specified.
/aspstore/storefrong.asp, line 86

The Tools Component can be used to help minimize these errors by using the FileExists
method. The FileExists method provides server-side validation of any file within a virtual
directory on the server. For example, the previous error could have been prevented if
FileExists was applied to the Content Rotator file. In the Amazing DataWidgets WareHouse
site, the Tools Component is used explicitly for this purpose. For instance, Listing 19.5 demon-
strates not calling the Content Rotator Component if the Content Rotator scheduling file is not
found.
Part
IV
Listing 19.5 Using the Tools Component to check on the existence of files
Ch
on the Web server.
<TABLE border = “0” cellpadding = “2” width = “60%”> 19
<%
Dim oDoesThe ‘=== Variable Declaration
Set oDoesThe = Server.CreateObject(“MSWC.Tools”) ‘=== Creates Tool Component
➥instance
If oDoesThe.FileExists(“acolades.txt”) Then ‘=== If the file exist then
➥write table
%>
<TR>
<%
Dim NextTip
Set NextTip = Server.CreateObject(“MSWC.ContentRotator”)
Response.Write NextTip.ChooseContent(“acolades.txt”) ‘=== Write content
%>
</TR>
<TR>
<%

continues

24.1389-6 CH19 415 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 416 Chapter 19 Putting It All Together: Creating an Online Registration System

Listing 19.5 Continued

Response.Write NextTip.ChooseContent(“acolades.txt”)
%>
</TR>
<%
End If ‘=== File not found, start constructing default table
%>
<TR><TD>
<TD align=”center”><font face=”Arial”><img
src=”/ASPstore/images/TN00605A.gif” width=”77" height=”69"> </font></TD>
<TD><font face=”Arial”>Make the world spin a little
faster - </font><a href=”../ASPstore/UserProfile1.asp”><font
face=”Arial”>Reserve</font></a><font face=”Arial”> your
copy of the MIND-Bender ActiveX control today.</font></TD>
</TD></TR>
</TABLE>

If the Content schedule file acolades.txt is not found, only the last row of the table is gener-
ated (see Figure 19.4).

FIG. 19.4
The FileExists
method of the Tools
Component can validate
whether the file is found
on the Web server.

If the scheduling file is found, the Content Rotator is called to display the full table displayed in
Figures 19.1 and 19.2. Keep in mind that this same principle can be applied not only to compo-
nent-dependent files, but also to any file within the server’s virtual root, such as GIF, JPG, text,
and even ASP files themselves.

24.1389-6 CH19 416 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Detecting the Browser with the Browser Capabilities Component 417

Detecting the Browser with the Browser Capabilities

Component
The next step in building the Amazing Web site is to determine what type of browser is looking
at the site. This is useful if you build a site that is designed to meet a specific HTML standard,
or support advanced functionality of a specific browser without generating errors. Because this
site deals specifically with product sales, you did not want to limit the user base, but at the
same time you do want to give advanced browser functionality to newer browsers. For the
Amazing DataWidgets WareHouse site, the CEO insists on not using any client-side ActiveX
controls or Java applets, but at the same time wants to use a scrolling effect to display the dis-
claimer across the bottom of the page. To meet these needs, a marquee is displayed if the
browser is Internet Explorer, as you saw in Figure 19.1. If Netscape Navigator is used, as pic-
tured in Figure 19.2, an HTML text box uses JavaScript to scroll the text. If another browser is
used, the disclaimer is simply displayed on the page.

To accomplish this, a server-side ASP script will initially detect and deliver HTML to the appro-
priate browser. If a Netscape browser is used, the server-side ASP will generate client-side
JavaScript that scrolls the text in a text box and displays the time in the status bar of the
browser. If Internet Explorer is used, the server-side VBScript code will generate the HTML
tags for the marquee and generate the JavaScript for the status bar clock. If neither of these
browsers is used, it won’t generate any JavaScript or marquee tags.

To determine the requesting browser’s display functionality, the combination of VBScript,

the Active Server Request object, and the Browser Capabilities Component will be used. The
Request object, as covered in Chapter 15, is used to extract information about the requesting
Part
browser. The Request object will retrieve information from the client’s HTTP header request.
The Request.ServerVariables property, as shown in Listing 19.6, is used to retrieve the IV
HTTP_USER_AGENT server variable.
Ch

19
Listing 19.6 Determining the browser type using the Request object.
<%
If Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”) Then
‘=== This is an IE browser
Elseif Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”Mozilla”) Then
‘=== This is a Netscape Browser
End If
%>

This server-side script uses the VBScript InStr function to search for a sub-string within a
parent string. In this situation, VBScript is looking for the string MS IE within the
HTTP_USER_AGENT server variable. If the string is found, the requesting browser is an Internet
Explorer browser. If the browser is not Internet Explorer, then the same test can be applied to
determine if the user is using a Netscape browser.

24.1389-6 CH19 417 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 418 Chapter 19 Putting It All Together: Creating an Online Registration System

Now you need to extract more information about the specific capabilities of the requesting
browser. The Browser Capabilities Component enables you to find out detailed information
about the requesting browser by extracting the HTTP_USER_AGENT header and performing look-
ups against the BROWSAP.INI file. The BROWSCAP.INI file is a text file that contains browser
definitions that define the capabilities of each browser. For the latest BROWSCAP.INI file, visit
http://www.microsoft.com/iis/usingiis/developing/updates.htm. You can use the combi-
nation of the server variables and the Browser Capabilities Component to extract browser
capabilities, as shown in Listing 19.7.

Listing 19.7 Extracting version information with the Browser Capabilities

Component.
<%
Dim BrowserType, MinBrowser, isitIE
isitIE = Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”MSIE”)
Set BrowserType = Server.CreateObject(“MSWC.BrowserType”)
If isitIE Then ‘=== Drill down on
➥specifics
If BrowserType.MajorVer = “3” Then ‘=== Test version
➥using component
MinBrowser = True
Else
MinBrowser = False
End If
Else ‘=== Its not IE
If Instr(Request.ServerVariables(“HTTP_USER_AGENT”),”Mozilla”) Then
MinBrowser = True
End If
End If
%>

This code example is used to set to variables that will be used throughout the remaining code:
the isitIE and the MinBrowser. For example, when the page is loaded into the browser, the
ONLOAD event of the HTML <BODY> tag is used to trigger different JavaScript functions (see
Listing 19.8).

Listing 19.8 Triggering different JavaScript functions on the ONLOAD event

of an HTML page.
<% IF isitIE Then %>
<body bgcolor=”#FFFFCC” onload=”loadclock()”>
<% Elseif MinBrowser Then%>
<body bgcolor=”#FFFFCC” onload=”loadinit()”>
<% Else %>
<body bgcolor=”#FFFFCC”>
<% End If %>

24.1389-6 CH19 418 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Determining Page Hits with the Page Counter Component 419

Likewise, you can use this same approach to determine whether you are going to display the
marquee, the JavaScript scrolling text box, or the standard text disclaimer, as shown in List-
ing 19.9.

Listing 19.9 Determining display options with the Browser Capabilities

Component.
<%
Dim strDisclaimer
strDisclaimer = ”*Note, the Amazing DataWidgets WareHouse in not responsible for
other information extracted from the mind and is not responsible for memory
lapses relating to appointments, birthdays or anniversaries.”
%>
<%If isitIE Then %>
<p><font size=”1" face=”Arial”><marquee scrollamount=”2"
scrolldelay=”50"><% Response.Write strDisclaimer %></marquee></font></p>
<% ElseIf MinBrowser Then %>
<form name=”frmmsg”>
<p align=”center”><Font face=”Arial” color=”#FFFFFF” size=”2"><input
type=”text” size=”50" name=”txtmsg”></font></p>
</form>
<% Else %>
<% Response.Write strDisclaimer %>
<% End If %>

Determining Page Hits with the Page Counter Part

Component IV
The Page Counter Component is used to keep track of how many times a page has been re- Ch
quested. The Page Counter Component stores the count in memory until it writes the amount
to a text file. The value is written to the file at a regular intervals to prevent grossly inaccurate 19
numbers as a result of failure. To change this value from the default value of 25, change the
Save_Count Registry setting. The Amazing DataWidgets WareHouse site uses the Page
Counter to display how many times the site has been requested (see Listing 19.10).

Listing 19.10 Creating a page counter with the Page Counter Component.
<%
Dim oPageCount
Set oPageCount = Server.CreateObject(“MSWC.PageCounter”)
%>
<font face=”Arial” size=”1">This Web page has been viewed <%= oPageCount.Hits %>
➥times.</font>

24.1389-6 CH19 419 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 420 Chapter 19 Putting It All Together: Creating an Online Registration System

The Page Counter Component tracks only the amount of requests for a page, not the total
amount of hits per page. The distinction between the two is important to understand. Because
HTTP must make individual connections and requests for every item contained on a Web page,
a Web page that has five images would generate at least six hits, one for the initial page request
for the page and five separate requests for each image on the page.

Build a Web Registration with the Content Linker

At this point, the main page for the Amazing DataWidgets WareHouse is finished. However,
you need to build a registration wizard to help download the new design-time control. This
registration wizard enables you to sign up for the component, designate the shipping informa-
tion, and then assign billing information. The Content Linker Component is used to provide an
automatic page linking and navigation builder that enables the links to pages to be automati-
cally updated. Consider the situation if you build an application that requires four screens and
you have to add a new form into page position two. As a result of the insert, the remaining links
have to be updated by hand to reflect the new changes. However, the Content Linker Compo-
nent eliminates hand-coding navigation issues by providing the automatic page linking and
relative positioning between pages.

The Content Linker uses a file to create and maintain the navigation list. The navigation list
acts as a table of contents specifying what files to access. This table-of-contents method enables
you to access files from anywhere in the navigation list at any time. The Amazing DataWidgets
WareHouse uses the combination of the Content Linker Component and client-side JavaScript
to help provide navigation through the beta wizard. By having the Content Linker automatically
update links to determine what page is next and previous, you are effectively letting the Con-
tent Linker control the order of the pages in the registration wizard. The Content Linker con-
trols page navigation by reading values from a Content Linking List. This text file controls the
order and descriptions of the Web pages. For example, the Content Linking List, named
userwizard.txt for the Amazing DataWidgets WareHouse, is shown in Listing 19.11.

Listing 19.11 The Content Linking List File for the DataWidgets site.
storefront.asp Main Page
userprofile1.asp Setup user
userprofile2.asp Billing Info
userprofile3.asp Shipping Info
summary.asp Summary Page

In this situation, the Content Linking List File provides sequential navigation and is used to
determine relative positioning within this list. The Content Linker uses this list to keep track of
what page the user is currently viewing to help determine what page is next and what page is
previous. In this situation, if the user is currently browsing the userprofile2.asp page, the
Content Linker provides a pointer to the previous page, userprofile1.asp, and to the page
after the current page, userprofile3.asp. For example, Listing 19.12 determines the total
amount of files in the Content Linking File by using the GetListCount method.

24.1389-6 CH19 420 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Build a Web Registration with the Content Linker 421

Listing 19.12 Using the GetListCount method to determine the number of

files in the Content Linking File.
<%
dim NextLink, MaxPage, CurrentPage, LinkingFile
LinkingFile = “userwizard.txt”
Set NextLink = Server.CreateObject (“MSWC.NextLink”)
MaxPage = NextLink.getlistcount(LinkingFile)
CurrentPage = NextLink.GetListIndex (LinkingFile)
%>

This example also extracts the index of the current page the user is browsing with by using the
GetListIndex method.

In Chapter 17, “Implementing Existing Server-Side Components,” you saw examples of using
the Content Linking Component to create hypertext links to navigate among the files in the
content list. This is fairly straightforward when you just need to provide a navigation technique
though textual documents, such as an online reference guide or newspaper. However, passing
information from page to page becomes a little more difficult. For the Amazing DataWidgets
WareHouse site, buttons were chosen as the navigation motif, as shown in Figures 19.5, 19.6,
and 19.7.

FIG. 19.5
The registration wizard
guides the user through
requested information.
Part
IV
Ch

19

The purpose of the registration wizard is to gather information from the user and submit the
collected information. You only have two ways to pass information from page to page in HTML.
The first method passes information in the URL string by concatenating variable names and

24.1389-6 CH19 421 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 422 Chapter 19 Putting It All Together: Creating an Online Registration System

values onto the posted URL string. This method is typically used when you use hyperlinks as
the navigation method. The requested page can then extract this information from the URL
and use it as needed.

The second method passes the HTML form information by using the POST method. The POST
method transfers HTML form values to the calling page specified in the ACTION method of the
Form tag. Because the Form tag can only have one direction for controlling programming flow,
the difficulty arises when you want to give the user the ability to navigate in different direc-
tions. There are two means to work around the limitation of only one Action item per form.
The first method uses the button’s OnClick event to trigger the page to browse to a new loca-
tion by setting the location.href property. To the user, using the OnClick event of a button
causes the page to react like a hyperlink and sends the browser to the designated page. How-
ever, the location.href property does not automatically submit or pass the form information
to the Web server. If you want to pass information onto the next page, the variable and values
must be posted to the server.

The second away around these issues is to use a non-visual Active Server Page that processes
different application logic depending on the value of the button submitted to the Active Server
Page. Once the value of the button is sent to an ASP on the Web server, the Web server can
direct the browser to the appropriate page. An example of posting values to different locations
using a single Form is demonstrated in Chapter 16, “Putting It All Together: Building an Online
Catalog.”

However, in this example, the approach demonstrated in Chapter 16 is not used because you
are relying on the Content Linker to manage consecutive visual Active Server Pages. For the
Amazing DataWidgets WareHouse site, combinations of the HREF and POST methods are used
to control navigation. For forward navigation, the Form POST method is used to transfer infor-
mation to the next page designated by the Content Linking File. For reverse navigation, the
locations.href event is executed to return to the previous page designated in the Content
Schedule File. Both of these events are activated using two client-side JavaScript functions: the
LoadNextPage and LoadPrevPage functions. Listing 19.13 displays the Active Server VBScript
that generates the client-side JavaScript. JavaScript is used on the client to ensure the largest
potential browser community.

Listing 19.13 Creating client-side JavaScript navigation techniques using

the Content Linker Component, Submit, and Href HTML attribute.
<%
Response.Write “<Script language=’JavaScript’>”
Response.Write “function loadnextpage(form) {“
Response.Write “Form.Submit()”
Response.Write “}”

Response.Write “function LoadPrevPage() {“

Response.Write “location.href= ‘“ &NextLink.GetPreviousURL(“userwizard.txt”)
&”’”
Response.Write “}”
Response.Write “</Script>”
%>

24.1389-6 CH19 422 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 Build a Web Registration with the Content Linker 423

The LoadNextPage function triggers the Form.Submit event for the specific form. When this
event is executed, the form information is passed to the page specified in the Action event of
the form. To transfer information to the next page in the Content Linking File, the GetNextURL
method is used, as shown in the following example:
<form method = “POST” name = “frmuser” Action = “<%
➥=NextLink.GetNextURL(“userwizard.txt”) %>”>

The LoadPrevPage JavaScript uses the location.href event to navigate to the previous page in
the Content Linking File. These two generic functions can be used to uniformly control naviga-
tion throughout all pages on the registration wizard.

The next phase needed to finish the registration wizard is to create the individual buttons on
the appropriate pages. The first page of the registration wizard, as shown in Figure 19.5, will
only have one button that enables the user to only navigate to the second page. The remaining
pages in the wizard will enable forward and reverse navigation. For example, Figure 19.6 dem-
onstrates the billing information page that uses the Content Linking File to create the Previous
and Continue buttons.

FIG. 19.6
The registration wizard
uses the Content
Linking Component to
aid in navigation.

Part
IV
Ch

19

Of course, you want to create a generic function to automatically create the buttons when new
pages are added to the wizard. To accomplish this feat, an Active Server Page VBScript func-
tion is written to automatically create the correct navigational buttons throughout the online
wizard. The self-sustaining code, as shown in Listing 19.14, will retain complete control over
the links within the online wizard and does not require significant coding effort when new
pages are added to the wizard.

24.1389-6 CH19 423 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 424 Chapter 19 Putting It All Together: Creating an Online Registration System

Listing 19.14 Creating forward and reverse navigation buttons.

<%
dim nextlink, maxpage, nextpage, currentpage, str, LinkingFile
LinkingFile = “userwizard.txt”
Set NextLink = Server.CreateObject (“MSWC.NextLink”)
Maxpage = Nextlink.getlistcount(LinkingFile)
Currentpage = NextLink.GetListIndex (LinkingFile)
Sub CreateNavButtons()
If (currentpage > 2) and (currentpage < maxpage) Then
‘============================= Display forward and next buttons============
‘=== Prev Page ====
str =”<TR><TD><input type=’button’ name=’btnprev’ value=’ << Previous’ “
str = str & “language=’JavaScript’ onclick=’loadprevpage()’></TD>”
Response.Write str
‘============================== Next Page button ==========================
str =”<TD><input type=’button’ name=’btnnext’ value=’Continue >>’ “
str = str & “language=’JavaScript’ onclick=’loadnextpage(this.form)’></
TD></TR>”
Response.Write str
Elseif currentpage = 2 Then
‘============================== only display next page ====================
str =”<TR><TD><input type=’button’ name=’btnnext’ value=’Continue >>’ “
str = str & “language=’JavaScript’ onclick=’loadnextpage(this.form)’></
TD><TR>”
Response.Write str
Elseif currentpage=maxpage-1 Then
‘============================== display previous and finish buttons =======
‘=== display previous button ===
str =”<TR><TD><input type=’button’ name=’btnprev’ value=’<< Previous’ “
str = str & “language=’JavaScript’ onclick=’loadprevpage()’></TD>”
Response.Write str
‘=== display finish button
str =”<TR><TD><input type=’button’ name=’btnnext’ value=’Order Now!’ “
str = str & “language=’JavaScript’ onclick=’loadnextpage(this.form)’></
TD><TR>”
Response.Write str
End If
End Sub
%>

This function tests for three conditions to determine what buttons are to be created on the
page. If the current page is the last page of the wizard, that is, page 3 in the Content Linking
File, two buttons will be created that are labeled << Previous and Order Now!, as shown in
Figure 19.7.

To implement this function, call the CreateNavButtons function where you need the button to
be added.

24.1389-6 CH19 424 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 From Here… 425

FIG. 19.7
Form navigation
information is made
possible by using the
Content Linker,
JavaScript, and HTML
POST and HREF
commands.

<TABLE border=”0" cellpadding=”2" cellspacing=”0">

<% CreateNavButtons %>
</TABLE>

The final result of the wizard displays the information that was collected on various pages on
the summary page, as shown in Figure 19.8.

FIG. 19.8
The registration wizard
displays the summary
information.

Part
IV
Ch

19

From Here…
In this chapter, you have seen the combination of several installable components that ship with
Active Server Pages. Specifically, the Browser Capability Component, in conjunction with
server variables, was used to demonstrate how to identify different browsers to tailor output to
their specific viewing capabilities. The Ad Rotator and Content Rotator Components were used
to illustrate how to rotate images and content on your site. The Content Linker was used to
create an automatic navigational method between pages based on files listed in the Content

24.1389-6 CH19 425 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 426 Chapter 19 Putting It All Together: Creating an Online Registration System

Linking file. The Tools Component was used to demonstrate extra server-side administration
tools to help prevent broken links or errors from missing required files. Finally, the Page Com-
ponent was used to display the number of times a page has been accessed.

For related information, please refer to the following chapters:

■ Chapter 18, “Creating Your Own Active Server Components,” walks you through the
tools and examples on how to create tailored functionality via custom created compo-
nents.
■ Chapter 20, “Database Access with ADO,” teaches how to use the Data Access Compo-
nent to connect to any ODBC-compliant database.

■ Chapter 26, “Putting It All Together: Creating a Dynamic Database Application,”

illustrates how to use the Application object to store information extracted from a
database to provide database interaction for the Amazing DataWidget WareHouse.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various feature of the Internet Information Server to help
manage your sites and applications.

24.1389-6 CH19 426 10/26/98, 9:37 AM

Brands3/nhq4 SEUsingActiveServerPgs 1389-x 11.13.97 polly CH19 LP#3

 P A R T

V
Database Management with Active
Server Pages
20 Database Access with ADO 429

21 Working with ADO’s Connection Object 451

22 Using Command Objects 469

23 Working with ADO’s Recordset Object 485

24 Managing Your Database 505

25 Accessing Databases with Remote Data Services 521

26 Putting It All Together: Creating a Dynamic

Database Application 535

25.1389-6 Part V 427 10/26/98, 9:38 AM

Brands SEU Active Server Pages #1389-6 Part V SRing 10/28/97 lp#1
25.1389-6 Part V 428 10/26/98, 9:38 AM

Brands SEU Active Server Pages #1389-6 Part V SRing 10/28/97 lp#1
 20
429

C H A P T E R

Learn the History of ADO

Database Access Why is ADO so significant? How
does it relate to other data access

with ADO methods like DAO, RDO, and

ODBC?

Examine the ADO Object Model

Learn how and when you should
exploit ADO for maximum
efficiency.

So far, you have seen how ASP can help you create a rich Discover ADO’s Transaction
and Rollback Processing
interactive environment for your Web site. The Session
Capabilities
and Application objects enable you to create Web sites
that “remember” users. The powerful scripting allowed by ADO enables you to maintain your
VBScript further extends what you can put on your Web applications data integrity with easy-
to-use transaction processing.
page.

Now, you will examine ActiveX Database Objects (ADO). See Why the Advanced Data
ADO is a collection of objects that enable ASP developers Connector Can Save You Hours
to connect to databases. Database access can be the differ- of Development Time
ence between a Web application that is a toy and a full- A firm understanding of the funda-
powered, enterprise-level application. mentals of the Advanced Data
Connector can save you hours of
Many of your ASP applications will rely on some sort of Part
development time.
database connectivity. While you can build ASP pages that V
aren’t data driven, there are many more useful (and fun!) Build a Working Example of a
Ch
applications that you can build by using a database behind Database-Driven Application
the scenes. For instance, you can create a simple Web- Build a drill-down application that 20
based telephone directory for your company’s employees finds a fellow employee’s phone
very quickly and easily with ADO. number in a database using ASP
with ADO.
This chapter will introduce you to ADO and its objects,
collections, and methods. Later chapters examine each
object and its methods in more detail. Feel free to play
with the example code in the text as you work through
each chapter. ■

26.1389-6 CH20 429 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 430 Chapter 20 Database Access with ADO

Understanding ADO
If you have ever used data objects before, ADO will seem very familiar. And while there are
some differences you will need to be aware of, you should be able to start coding with it very
quickly.

Unlike DAO and RDO, there is no complicated hierarchy to be navigated in order to create a
recordset. Sometimes you just want to get to your data. ADO enables developers to create a
recordset quickly.

Accessing versus Creating

This would be a good time to explain the difference between accessing and creating a recordset. The
basic difference is one of usage. You create a recordset. That recordset is then used to access the
data.
An example in the real world would be telephoning a friend. By dialing (creating the recordset) you
can talk to (access the underlying data) your friend.

At first I thought that Microsoft was just renaming DAO to make it fit the ActiveX marketing
plan. But I soon discovered I was wrong: ADO is built on an entirely different foundation than
DAO, or RDO for that matter. ADO is part of OLE DB (Object Linking and Embedding for
Databases), an exciting new way to access and manipulate data that Microsoft has developed.

OLE DB
OLE DB is an object-oriented specification that enables you to access data. It is an Application
Programming Interface (API) based on the Component Object Model (COM). ODBC, for
instance, is an API based on C. However, don’t think that ODBC has been replaced by OLE
DB; ODBC is a subset of OLE DB and it is still a great way to interact with SQL-based data-
bases. Figure 20.1 shows how OLE DB fits into the access paradigm.

OLE DB versus ODBC

So, when do you use OLE DB? When should you think about using ODBC? In reality, it isn’t really a
question of one versus the other, but more of an understanding which technology is more appropri-
ate to apply to the problem at hand.
OEL DB is a specification that is a superset of ODBC. However, OLE DB enables developers to access
data from more than just the row-based databases that ODBC was designed for.
ODBC is a specific API that gives developers access to SQL-compliant databases. Both OLE DB and
ODBC give developers low-level access to databases.

But OLE DB is better because it is based on component architecture. Data is in a tabular for-
mat, but is not reliant on SQL. This means that regardless of what type of datasource you are
accessing—for instance, an e-mail store—you can still use the same components you would
with a SQL server database. With OLE DB, you have data providers and data consumers,
which are components that enable this encapsulation.

26.1389-6 CH20 430 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 Understanding ADO 431

FIG. 20.1 Your Application

OLE DB offers a better
approach to accessing
your data.

Data Consumer

Data Provider

OLE DB is going to be a big deal in the near future, and ADO is a great way to start under-
standing and using OLE DB. Understanding OLE DB will put you ahead of the pack. But that’s
the future. Now, in the present, by using ADO, you can simplify and speed up your develop-
ment. You can use its simplified, data-centric architecture to create robust Web-based applica-
tions now.

N O T E You can learn more about OLE DB, along with other information about Microsoft’s data
access strategy, at www.microsoft.com/Data. ■

Part
Web Application Development
Developing Web-based applications can be very different from other types of programming. V
Unlike normal client/server applications, Web pages are stateless, which means that you call Ch
the page, it grabs the data it needs, and then it sends your document to you. Once a Web
server sends you a page, it forgets about it. In other words, the server does not know if this is
20
the fortieth time you have accessed a page or the first.

In traditional client/server development, a client application maintains a connection with its

server component. Besides requesting and sending data, the client queries the server periodi-
cally to make sure the connection is still alive. If the server goes down, it will sense this and
take the appropriate measures such as sending an error to the user and exiting.

As a Web application developer, you not only have to deal with the short-lived memory of Web
servers, you also have to deal with the Internet being a connectionless network. When a

26.1389-6 CH20 431 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 432 Chapter 20 Database Access with ADO

browser requests a page, it is served up and sent out. That’s it. The server is not going to
check and see whether the page and all of its components (graphics, sounds, and so on) made
it to the browser.

But don’t despair. ASP and ADO are significant steps in dealing with these issues. An Active
Server doesn’t forget the way an old-fashioned server does; it has a memory. Think of an ASP
server as a loyal machine that can remember its friends and past dealings. When you are deal-
ing with millions of hits a month, this feature will become a lifesaver.

Of course, this new functionality comes at a cost. An Active Server needs more RAM than a
server that doesn’t offer the capability to “remember” users. And if you use databases to store
information about users that span across multiple visits (as opposed to using the Session ob-
ject to remember information per visit), you will need more hard disk space to hold the infor-
mation your database contains.

Products for Web Development

Some of you might be familiar with IDC/HTX, and you may wonder whether Microsoft has
abandoned IDC. Not exactly, it still works with IIS 3.0. But don’t expect to see any improve-
ments in version 4. However, Microsoft is offering a tool to help port your existing IDC
projects into ADO.

You might also be a Web developer who has used a really great product from Allaire called
Cold Fusion. Cold Fusion works much like IDC/HTX. It enables you to connect to an ODBC
datasource, and embed its data in your HTML pages. Basically, it’s a better version of IDC/
HTX (with a larger price tag as well!).

Cold Fusion is very easy to learn and it still has a definite place in your toolkit. ASP works best
when it is used for larger, more complex programs that need the robustness a scripting lan-
guage can offer. Cold Fusion is a great choice for simple data-driven programs.

N O T E Allaire’s Web site, www.allaire.com, is always a great place to check out the latest
information on Cold Fusion. Updates, new releases, and great examples of Cold Fusion
products can be found there. ■

The ADO Object Model

The ADO object model provides developers a quick, yet powerful, method of accessing a
datasource. Like DAO, you have Recordset objects. The Connection object is new and pro-
vides a similar functionality that DAO’s Workspace object gives.

■ Connection
■ Error
■ Command
■ Parameter

26.1389-6 CH20 432 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 433

■ Recordset
■ Field

The list above comprises all the objects that make an ADO hierarchy. There are three major
objects in ADO that provide an interface to your data: the Connection, Command, and Recordset
objects. Think of the Connection interface as the road to the datasource. Without it, you can’t
get there. It also contains the Error object, where errors from the datasource are logged. The
Command interface provides a way to send parameters to stored procedures when opening a
recordset. The Command interface is optional, but still very important. Finally, the Recordset
interface is the workhorse of ADO. It is the object that you will use to interact with your data.
Recordset uses the Field object.

Objects, Collections, and Interfaces

In many texts on ADO and other object models, authors use the words collection, object, and
interface. But they rarely explain what they mean by these words (if anything!).
An object has both properties and methods. For instance, a phone has a dialer (a property) and
speed dial (a method). Other properties of a phone object might include its color, location, and
whether it is a pulse or rotary dialer. However, some of these properties, such as color, might not be
publicly accessible. To change the phone’s color, you might need to invoke a method called
ChangeColor, and send that method the color you want to change the phone to. Other properties,
however, might be publicly accessible, allowing you to change them without having to call a certain
method.
The set of properties and methods in an object that are publicly accessible are collectively referred
to as the interface of the object. In practice, however, interface and object hold the same meaning.
So, when I say “the Collection object” or “the Collection interface,” it will be easiest to equate
these as speaking of the same thing.
(Most authors, including myself, will switch between the two quite frequently. Why? To alleviate the
boredom of writing—and reading—the same word 20 times on a page.)
However, collection is different. A collection is just a collection of objects. Sometimes, you don’t have
just one object; you have several of them grouped together. This group of objects is referred to as a Part
collection.
V
For instance, in ADO, there is the Error object. But, generally, I will refer to the Errors collection—
meaning the set of Error objects. Collections in ADO often have their own properties and methods Ch
that are separate form the properties and methods of the individual object. 20

Properties and Methods of ADO Objects

ADO object’s properties and methods are fairly easy to understand and use. They offer devel-
opers a non-complicated way to access and manipulate data. If you are at all familiar with the
hierarchical model imposed on developers from DAO, ADO will seem like a breath of fresh air
with its capability to quickly access data without having to navigate through an extensive hier-
archy of objects first.

26.1389-6 CH20 433 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 434 Chapter 20 Database Access with ADO

Like all good objects, ADOs have both methods and properties that are publicly accessible.
Many of these properties and objects offer very similar functionality. For instance, the
Connection, Command, and Recordset objects all offer a way to specify a datasource that will
be the active connection.

The Connection Object The Connection object provides a connection to the datasource.
Because it is the connection to the datasource, it also contains the Error collection for any
errors that the datasource may return. You can learn more about the Connection object in
Chapter 21, “Working with ADO’s Connection Object.”

Table 20.1 lists the various properties of the Connection object. These properties cover a wide
variety of functionality, and give developers quite a bit of control over the connection to the
datasource.

Listing 20.1 shows an example of how easy it can be to create and configure a Connection
object, and then use it to connect to a datasource.

Listing 20.1 20ASP1.ASP—Creating a Connection object can be easy!

Dim dataConn
dataConn = Server.CreateObject(“ADODB.Connection”)
dataConn.ConnectionString = “SampleDSN”
dataConn.open

Table 20.1 The Connection Object’s Properties

Property Description
Attribute Specifies whether a new transaction will be started after a
transaction rollback or commit.
CommandTimeout How long to wait for a command to process. Default is 30
seconds. Setting this value to 0 will make the wait indefinite.
ConnectionString The DSN, provider, username and password, and other
information necessary to connect to the database.
ConnectionTimeout How long to wait to connect to the datasource. Default is 15
seconds. Set to 0 to wait indefinitely.
DefaultDatabase If the datasource you are connecting to enables multiple
databases, this property enables you to specify which one is the
default.
Isolation The isolation level of the connection.
Mode Used to set the permissions on a connection, such as read-only
or read-write.

26.1389-6 CH20 434 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 435

Property Description
Provider The provider of a connection.
Version The version of ADO being used.

Table 20.2 lists the methods of the Connection object. There are relatively few methods used
with the Connection object. Generally, the properties of the Connection object are used to
configure the connection, and then either the Execute or Open method is called.

Table 20.2 The Connection Object’s Methods

Method Description
BeginTrans Begins a transaction.
ComitTrans Commits the transaction to disk.
RollBackTrans Rolls back the transaction.
Close Closes the connection.
Execute Executes a command on the datasource, returning a recordset.
Open Opens the connection.

The Error Collection and Error Object The Error collection is part of the Connection object.
When the datasource generates and returns an error (or errors), this is where they will be
found. Sometimes, these errors are actually just warnings. More about the Error object and
Errors collection can be found in Chapter 21.

Listing 20.2 shows an example of working with the Error collection. In this example, after
opening a connection, the Errors collection is checked to see if any errors were generated by
this opening.

Part
Listing 20.2 20ASP1.ASP—Checking the Error collection.
V
Dim dataConn
dataConn = Server.CreateObject(“ADODB.Connection”) Ch
dataConn.ConnectionString = “SampleDSN”
dataConn.open
20
If dataConn.Errors.Count > 0 Then
‘We know errors occurred, so we can
‘investigate those errors, or (as in this case)
‘we can print them out
for each error in dataConn.Errors
Response.Write(Error.Number & “:” & Error.Description)
Next
End if

26.1389-6 CH20 435 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 436 Chapter 20 Database Access with ADO

Table 20.3 lists the properties of the Error object. These properties can be used to uncover
such information as the type of error that occurred, what its source was, and a description of
the error.

Table 20.3 The Error Object’s Properties

Property Description
Description This is the description of the error.
HelpContext The help topic in the help file that contains more information on
the error.
HelpFile The filename of the help file that has more information on the
error.
NativeError This is a provider-specific error code.
Number A unique ID to distinguish this Error object form others in the
Error collection.

Source The name of the application (or object) that returned the error.
SQLState The five-digit SQL error code.

Table 20.4 shows the properties of the Errors collection. These properties are helpful in under-
standing the number of errors that have been raised, and in accessing these errors individually.

Table 20.4 The Error Collection’s Properties

Property Description
Count The number of Error objects in the collection.
Item Returns a specific Error object by name or number.

Table 20.5 lists the methods found in the Errors collection. Many of the methods of one collec-
tion will work with another, but understanding what is different about each collection can be
very helpful.

Table 20.5 The Error Collection’s Methods

Method Description
Clear Removes all the objects in the collection.

26.1389-6 CH20 436 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 437

The Command Object The Command object is a specific command executed against the
datasource, and it returns a recordset. The Command object is optional, but is particularly valu-
able when dealing with stored procedures. You can learn more about the Command object in
Chapter 22, “Using Command Objects.”

N O T E What is a command? A command is a series of instructions executed by a database that

return a recordset.
Commands are most often represented as SQL, which stands for Structured Query Language. You can
learn more about SQL in Chapter 24, “Managing Your Database.” ■

Listing 20.3 shows an example of the Command object at work. Here, you are creating a Command
object, setting its ActiveConnection property to the name of your DSN, and setting its
CommandText to a SQL command. You then invoke the Execute method of the Command object.

Listing 20.3 20ASP1.ASP—Creating a Command object.

Dim dataCmd
Set dataCmd = Server.CreateObject(“ADODB.Command”)

dataCmd.ActiveConnection = “SampleDSN”
dataCmd.CommandText = “SELECT * FROM tblStuff”

set rs = dataCmd.execute (, , adCmdTxt)

In Table 20.6, the properties of the Command object are listed. These properties can be used to
configure the Command object, including setting its data connection, and the command that it
contains, along with many other valuable settings.

Table 20.6 The Command Object’s Properties

Property Description
ActiveConnection Specifies the Connection object the Command object belongs to. Part

CommandText A text string that indicates the command to run against the
V
datasource. Ch
CommandTimeout Specifies how long to wait for a command to execute. Default 20
is 30.
CommandType The type of command.
Prepared Tells the provider to prepare, or compile, the command before
execution.

26.1389-6 CH20 437 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 438 Chapter 20 Database Access with ADO

Table 20.7 shows the methods of the Command object. The methods of the Command object are
used to create new parameter objects and execute the command.

Table 20.7 The Command Object’s Methods

Method Description
CreateObject Creates a new Parameter object.
Execute Executes the command stored in the CommandText property.

The Parameter Object and Collection The Parameter object is used to store parameters
that will be passed to a stored procedure. Parameters are variables that you can send to the
database that will be needed for the execution of certain queries and stored procedures. You
can learn more about the Parameter object in Chapter 22.

Some commands are already stored in the database in a special form called a stored procedure
(or query, depending on the database). Some of the queries need a parameter (or parameters)
sent to it to execute.

For instance, there might be a stored procedure in your database for returning the row that
contains a customer’s phone number. This procedure will need the customer’s name to look up
the phone number. In this case, the customer’s name is the parameter.

In Listing 20.4, a parameter is created and appended to the Parameter collection.

Listing 20.4 20ASP1.ASP—Creating a Parameter object.

StrName = “customerName”
StrValue = “Jones“
intType = adChar
intSize = 50
intDirection = adParamInput
dataParam = DataConn.CreateParameter(strName, intType, intDirection, intSize,
➥strValue)
DataConn.Append DataParameter

Table 20.8 lists and explains the properties of the Parameter object. These properties are used
to configure the Parameter before it is appended to the Parameter collection and sent to the
database. For instance, in the case of sending the customer name to find the phone number,
you need to configure the parameter as an input parameter, meaning that it will be used as
input to the query.

26.1389-6 CH20 438 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 439

Table 20.8 The Parameter Object’s Properties

Property Description
Attributes Indicates what type of data the parameter accepts.
Direction Indicates whether the parameter is an input or output param-
eter, or both.
Name The parameter’s name.
NumericScale For numeric parameter values, indicates the number of digits to
the right of the decimal point that it will be resolved to.
Precision For numeric parameter values, indicates the number of digits
that will be used.
Size The maximum size, in bytes, of the Parameter object.
Type The data type of the parameter.
Value The value of the parameter.

Table 20.9 lists the methods of the Parameter object. Basically, the only method you will ever
use with the Parameter object is that of appending a chunk of data to it.

Table 20.9 The Parameter Object’s Methods

Method Description
AppendChunk Appends a large amount of data to a Parameter object.

Table 20.10 lists the properties of the Parameter collection. Like the properties of most collec-
tions, these deal with accessing a particular parameter by number and counting the number of
parameters.
Part
Table 20.10 The Parameter Collection’s Properties V
Property Description Ch
Count The number of Parameter objects in the collection. 20
Item A specific parameter in the collection.

Once you have created a Parameter object, it needs to be appended to the Parameter collection.
Why have a collection of parameters? Some stored procedures take more than one parameter.
For instance, in the procedure you looked at earlier that returns a customer’s phone number,
you would probably send the first and last names as separate parameters—thus, the need for a
collection. Table 20.11 lists the methods of the Parameter collection.

26.1389-6 CH20 439 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 440 Chapter 20 Database Access with ADO

Table 20.11 The Parameter Collection’s Methods

Method Description
Append Adds a new parameter to the collection.
Delete Deletes a parameter from the collection.
Refresh Retrieves information from the provider on the parameters in the
stored procedure specified in the CommandText property.

The Recordset Object The Recordset object is the workhorse of ADO. It is the object you
most often use to manipulate the data in your database. Using the Field collection of the
Recordset interface, you can access individual fields of the recordset and change, delete, and
update them. You can find out more about the Recordset object in Chapter 23, “Working with
ADO’s Recordset Object.”

In Table 20.12, the properties of the Recordset object are listed and described. These proper-
ties are most often used to find out various status-type information about the recordset, such as
the number of records or the current position of the record pointer.

Table 20.12 The Recordset Object’s Properties

Property Description
AbsolutePage Which page the current record is on.
AbsolutePosition The place in the recordset the current record is on.
ActiveConection The current connection the recordset is using.
BOF, EOF Beginning of file, end of file.
Bookmark A unique reference in a recordset set.
CacheSize The number of records stored in memory of the client machine.
CursorType The type of cursor to open the recordset with.
EditMode Indicates whether the current record is being edited or added.
Filter Filters a recordset.
LockType The type of locking the recordset will enforce.
MaxRecords The maximum number of records the recordset will return.
PageCount The number of pages in the recordset.
PageSize The number of records a page will have.
RecordCount The number of records in the recordset.
Source The source of the recordset (that is, a table or query).
Status The status of a record.

26.1389-6 CH20 440 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 441

Table 20.13 shows the methods of the Recordset object. Use these methods to manipulate the
set of data as a whole, and to navigate through the recordset.

Table 20.13 The Recordset Object’s Methods

Method Description
AddNew Creates a new record.
CancelBatch Cancels a pending batch update.
CancelUpdate Cancels the changes made to a recordset.
Clone Creates a new recordset with the records in the current
recordset.
Close Closes the recordset.
Delete Deletes the current record.
GetRows Copies the specified records into an array.
Move Moves to a specific record.
MoveFirst, MoveLast, Moves to the record specified in the name.
MoveNext, MovePrevious

NextRecordset Closes the current recordset and opens the next one.
Open Opens a recordset.
Requery Reruns the query and updates the recordset.
Resync Syncs up with the database. Updates any locally cached
records.
Supports Indicates whether certain functions will be present in the
datasource.
Update Saves the changes made to the current record.
Part
UpdateBatch Saves the changes from the batch update. V
Ch
The Field Object and Collection The Field object is the interface to a specific field in the
current record. The collection is the group of fields that make up the current record. 20
Table 20.14 shows the various properties of the Field object. Use these properties to discover
information about the underlying nature of that particular field in the database. For instance, if
you need to know the maximum size in bytes the field can be, look at its DefinedSize property.

26.1389-6 CH20 441 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 442 Chapter 20 Database Access with ADO

Table 20.14 The Field Object’s Properties

Property Description
ActualSize The actual length of the data.
Attribute This is data for the field.
DefinedSize The maximum size of the field.
Name The name of the field.
NumericScale For numeric field values, indicates the number of digits to the
right of the decimal point that it will be resolved to.
OriginalValue The value of the field before it was changed.
Precision For numeric field values, indicates the number of digits that the
field will be resolved to.
Type The type of data.
UnderlyingValue The current value of the field in the database.
Value The actual value of the field.

In Table 20.15, the methods of the Field object are listed. These methods are used to append
and retrieve chunks of information from the field when that information is very large.

Table 20.15 The Field Object’s Methods

Method Description
AppendChunk Appends data to a field.
GetChunk Gets data from a field.

Table 20.16 lists the properties of the Field collection. This collection is used to access a par-
ticular field by number or to find out how many fields are in the collection.

Table 20.16 The Field Collection’s Properties

Property Description
Count The number of fields in the collection.
Item The specific field in the collection.

Table 20.17 lists the lone method of the Field collection. This method doesn’t really do any-
thing, but is included for the sake of consistency.

26.1389-6 CH20 442 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 443

Table 20.17 The Field Collection’s Methods

Method Description
Refresh This really doesn’t do anything. You will need to use the
recordset’s Requery method to refresh a field collection.

ADO in Action
The best way to understand how these objects work together to provide access to your
datasource is to see them in action (see Listings 20.5 and 20.6).

Listing 20.5 CHAP20.ASP—Opening a recordset can take two lines...

Set rs_main = Server.CreateObject(“ADODB.Recordset”)
rs_main.Open “Select * from tbl_stuff”, “DSN=stuff”

In Listing 20.5, I created a recordset with just two lines of code. However, this recordset is
severely limited in the functionality it provides. In Listing 20.6, you can see an example of a
more functional recordset, one that took eight lines of code to create.

Listing 20.6 CHAPTER20.ASP—...or eight.

Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN=Stuff”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
DataCmd.CommandText = “SELECT * FROM tbl_stuff”
DataCmd.CommandType = adCmdTxt
Set DataCmd.ActiveConnection = DataConn
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
rs_main.Open DataCmd

The code above utilizes, in turn, a Connection object and then a Command object to get to a Part
Recordset object, which the developer can then use to access data. The Recordset is the inter- V
face that you will use to access and manipulate your data. The Connection and Command inter-
faces are just the roads that enable you to get to the Recordset. Ch

Although they have a more limited role than that of the Recordset object, the Command and 20
Connection objects still have very important roles to play. The Connection object enables the
developer to control many details of the actual connection to the datasource. The Command
object gives the developer the ability to manage the commands that will be executed by the
datasource (see Figure 20.2).

The Connection object is always present. If you don’t explicitly create one, ADO will implicitly
create one for you. The Connection object represents the connection to your datasource. It has
various properties that enable it to be configured before you actually open anything with it. In
Listing 20.5, I implicitly created a connection when I opened the recordset rs_main. In Listing
20.6, I explicitly created a Connection object, DataConn.

26.1389-6 CH20 443 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 444 Chapter 20 Database Access with ADO

FIG. 20.2
The Connection and
Recordset
Command interfaces are
for creation of a
recordset.

Command

Connection

Data

Naming Conventions
What should you name your ADODB variables? Naming conventions can become a heated debate.
Everyone has their preferred names for variables and methods. I suggest that when you are
developing with Visual Interdev—especially if you will be using any wizards—that you stick to
InterDev’s names.
For instance, DataConn is what the Data Range Wizard names the Connection object it creates. I
like that name, so I use it in my own code. I further decided to name my Command object DataCmd.
What really matters, however, is that you use names that will be descriptive. Choose a convention
that will allow you to automatically know what kind of variable you are looking at and you’ll be fine.
This means including in each variable name a couple letters that represent what type of variable it is.
For example: Conn for a Connection Object, Cmd for Command, and rs for Recordset.

After you create the Connection object, create a Command object, DataCmd. The Command object
contains a command (in this case, a SQL statement) that the datasource can execute. It also
allows for passing parameters, with the Parameters property.

In Listing 20.7, a recordset is created with the Command object. A Connection object is still being
created, but implicitly.

26.1389-6 CH20 444 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 The ADO Object Model 445

Listing 20.7 CHAP20.ASP—Opening a recordset with the Command object.

Set DataCmd = Server.CreateObject(“ADODB.Command”)
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
DataCmd.ActiveConnection = “phone_numbers”
DataCmd.CommandText=”qry_list_numbers”
Set rs_main = dataCmd.Execute

Benefits of Coding with the ADO Object Model

I know what you’re thinking: “If I can create a recordset with two lines of code, why take
eight?” Or, to put it another way: “Why explicitly use the Connection or Command objects?”
There are actually some very good reasons to explicitly create those objects whenever possible.

N O T E The Command object is definitely optional. As a matter of fact, not all of your datasources
will even support it. But it still has a definite use. ■

To see the benefits of coding with the object model in mind, let’s start by doing just the oppo-
site. The code in Listing 20.8 creates two recordsets, rs_people and rs_places. Both are cre-
ated with a SQL statement: one on a table, and one on a query.

Listing 20.8 CHAPTER20.ASP—Creating recordsets the wrong way.

Set rs_people = Server.CreateObject(“ADODB.Recordset”)
Set rs_places = Server.CreateObject(“ADODB.Recordset”)
rs_people.Open “Select * from tbl_people”, “DSN=PeopleAndPlaces”
rs_places.Open “Select * from qry_places”, “DSN=PeopleAndPlaces”

What is so bad about the above code? For starters, two Connection objects are being created.
There is only need for one. So resources are being wasted. Second, the Command object could be
used to manage the opening of these recordsets, which would result in improved performance.
Listing 20.9 is another code snippet (which should look pretty familiar by now) opening these Part
same recordsets in a more appropriate manner. V
Ch
Listing 20.9 CHAPTER20.ASP—Creating a recordset the better way.
20
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN= PeopleAndPlaces”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
DataCmd.CommandText = “tbl_people”
DataCmd.CommandType = adCmdTbl
Set DataCmd.ActiveConnection = DataConn
Set rs_people = Server.CreateObject(“ADODB.Recordset”)
rs_people.Open DataCmd
DataCmd.CommandText = “qry_places”
DataCmd.CommandType = adCmdStoredProc
Set rs_places = Server.CreateObject(“ADODB.Recordset”)
rs_places.Open DataCmd

26.1389-6 CH20 445 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 446 Chapter 20 Database Access with ADO

In the code above, there is only one Connection object. Because both recordsets are accessing
data from the same datasource, this makes sense. Just as you don’t build a separate road for
every car, you don’t need to create a new Connection object for every recordset. The same
goes for the Command object.

Also, notice how DataCmd’s CommandText is not a SQL statement. In both cases, instead of writ-
ing “Select *”, the CommandText property is the name of the table or query. This can give you
increased performance primarily because CommandType is also set to what you are asking for.
This primes the pump, so to speak, and readies the connection.

This doesn’t mean that every time you create a recordset you should have first created a con-
nection and command interface for it. Some data providers don’t allow a Command object. And if
all you are writing is a quick-and-dirty page, it might be overkill to create the Connection ob-
ject. That having been said, I would still always do it, at least at first. Better to overcode than
undercode, especially in the learning phase.

Now that you have looked at some benefits to coding with the ADO object model in mind, let’s
look at another great benefit of ADO: transaction processing.

Transaction and Rollback Processing

ADO makes transaction processing incredibly easy. You can increase the reliability of your data
and the speed of your application with only a few lines of code. It’s really that simple. But what
is transaction processing?

Imagine for a moment that you have created a program to manage bank accounts. Customers
have asked for a way to transfer funds from one account to another. You write the code, and
everything works great. It takes the money out of the first account, and puts it into the second
account. But one day there is a power outage and the money is taken out, but never put back in
anywhere.

How can you avoid this error in the future? Transaction processing. Instead of subtracting the
money first, and then adding it to the next account, you can do both inside a transaction. If
something goes wrong, the transaction won’t be written out to the database.

Listing 20.10 shows how easy it can be to include transaction processing in your code.

Listing 20.10 CHAPTER20.ASP—Transaction processing at work.

DataConn.BeginTrans

‘ Insert your code here

If DataConn.Errors.Count = 0 Then
DataConn.CommitTrans
Else
DataConn.RollbackTrans
End If

26.1389-6 CH20 446 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 Remote Data Services 447

Whatever data handling you insert into this code will now only execute and write to the data-
base if there are no errors. If the datasource returns an error, the transaction will be rolled
back. In other words, it will be as if it never happened.

As wonderful as data integrity is, there is an even better reason to get excited about transaction
processing: increased performance. By waiting before it writes out to disk, your application can
see a marked increase in speed. This is because everything is happening in RAM until
CommitTrans is called. Then it is written out to disk. The actual amount of difference you will
see will depends mostly on the size of your database and the tables involved.

TIP Take the above code and start wrapping it around any code you now have that may need it. You will be
amazed at the increased performance you will see.

Remote Data Services

So far, you have seen how to access data that can be put into a page any browser can under-
stand. That is the beauty of ASP—its browser independence. But this does present a drawback.
What if you want to create a fat client? The browser can be a rich environment, and there are
times (especially on an intranet) where you will want to enable the browser far more control
than you would in a more Internet-like environment.
This is where the Advanced Data Connector (ADC) and Remote Data Services (RDS) come in.
RDS enables developers to create rich, interactive experiences. However, it requires an
ActiveX-enabled browser, such as Microsoft’s Internet Explorer 4. You can learn more about
the Advanced Data Connector and Remote Data Services in Chapter 25, “Accessing Databases
with Remote Data Services.”

N O T E You can also check out Microsoft’s Web site on Remote Data Services and the Advanced
Data Connector at www.microsoft.com/data/adc. ■

In a page that utilizes the ADC, an invisible ActiveX control called AdvancedDataControl is Part
embedded. A visual control, like a data grid, is then added to the page and bound to the
AdvancedDataControl. Finally, client-side VBScript can be used to extend the functionality of
V
the page. Ch

20
Visual Controls and Data Grids
What are visual controls? If you are a developer who doesn’t have a VB background, this question
may be in your mind right now. That’s okay. Visual controls are easy to understand. And rest assured
that developing ASP does not require a VB background, though it can help.
Visual controls are visual components of a Windows program that have certain functionality and code
bound with them. For instance, a data grid control consists of several text fields that represent
different fields in a database, along with buttons for moving through the data.

26.1389-6 CH20 447 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 448 Chapter 20 Database Access with ADO

Creating a Simple Data-Driven ASP Program

Listing 20.11 is rather simple. It is a drill-down to find an employee’s phone number. On the
first page, search.asp, the user enters the last name of the employee. At Results.asp, a
recordset is created that displays all the employees with that last name. Also displayed are their
first names and phone numbers. From this list the user can choose the employee he wants and
dial their number.

Listing 20.11 SEARCH.ASP—A simple page that takes an employee’s last

name from the user.
<HTML>
<HEAD>
<TITLE>
Search for a phone number
</TITLE>
</HEAD>
<BODY>
<CENTER>
<H1>
Search for a phone number
</H1>
<P>
<FORM ACTION=”results.asp”method=”post”>
Enter Employees last name:
<INPUT TYPE=”text” NAME=”Last_Name”>
<BR>
<INPUT TYPE=”submit” Value=”Submit”>
</FORM>
</CENTER>
</BODY>
</HTML>

In Listing 20.12, you search the database for a matching employee. This is done by first creat-
ing a Connection object and setting its connection to the DSN that is your database. Next, you
create a Command object that contains a SQL statement that is executed against the datasource
to return the correct employee. Finally, you open a connection to the datasource and have a
recordset returned that contains the right employee.

Listing 20.12 RESULTS.ASP—The results page lists all employees with the
last name entered.
<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>
Document Title
</TITLE>
</HEAD>
<BODY>

26.1389-6 CH20 448 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 From Here… 449

<%
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN=Employees”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
DataCmd.CommandText = “SELECT * FROM tbl_employees WHERE LastName = ‘“ &
Request.Form(“Last_Name”) & “‘“
Set DataCmd.ActiveConnection = DataConn
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
set rs_main = DataCmd.execute
%>

<TABLE>
<TR>
<TD>First Name</TD>
<TD>Last Name</TD>
<TD>Phone</TD>
</TR>

<%
do While not rs_main.EOF
Response.write(“<TD>”)
Response.write(rs_main(“FirstName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“LastName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“Phone”))
Response.write(“</TD>”)
Rs_main.moveNext
loop
%>
</TABLE>
</BODY>
</HTML>

The above sample code might seem pretty simplistic. However, it does illustrate several points
from the chapter. For instance, notice how the recordset rs_main is created. It is created Part
through both a Connection and a Command object. In later chapters, when you extend the func-
tionality of this application, you will see further reasons why coding even simple apps to the
V
ADO object model is always a good idea. Ch

20
From Here…
In this chapter, you have been introduced to ActiveX Data Objects, also known as ADO. ADO is
an amazing new way to access a database inside of an Active Server Page. It is an implementa-
tion of OLE DB, which is a huge step forward in database connectivity. By learning about ADO
and OLE DB now, you are setting yourself up for huge returns in the future.

26.1389-6 CH20 449 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 450 Chapter 20 Database Access with ADO

This chapter has prepared you to begin leveraging ADO in your development efforts by intro-
ducing you to its fundamental building blocks. You’ve seen how each of the major objects inter-
relate, the benefits to coding with the ADO object model in mind, and how ADO’s transaction
processing capabilities can help you with your data integrity and application performance.
You’ve also seen how the Advanced Data Connector can simplify your development efforts.

There are three major objects you use with ADO: Connection, Command, and Recordset. Each
of these is discussed in the chapters ahead.

■ Chapter 21, “Working with ADO’s Connection Object,” will help you to further under-
stand how to use this interface in your code.
■ Chapter 22, “Using Command Objects,” will show you how and when to use this object for
better performance and ease of development.
■ Chapter 23, “Working with ADO’s Recordset Object,” will cover how to use the work-
horse of ADO, the recordset.

26.1389-6 CH20 450 10/26/98, 9:39 AM

Brands/Stone3 SEU Acitve Server Pages #1389-6 SRing 11/13/97 CH20 lp#3
 21
451

C H A P T E R

Linking to Your Database with

Working with ADO’s the Connection Object
Connecting to a database can be

Connection Object easy with the Connection object.

Creating it is the first step in data-
base access.

Configuring the Connection

Object and Its Properties
Configure the Connection object for
maximum efficiency. Learn how to
The previous chapter introduced you to ActiveX Database optimize for the type of application
you are creating.
Objects (ADO). With the exception of the Session object,
ADO is, perhaps, ASP’s greatest feature. Previously, data-
Controlling the Connection
base access within Web pages either relied on tools that
Object
were too simple to create very robust applications, or too
clumsy to expand. ASP enables thousands of VB program- See when to use the methods of
the Connection object in your
mers to start leveraging their knowledge to create work-
application.
ing applications immediately. Furthering the idea that
database access shouldn’t be difficult, ADO enables devel- The Error and Command
opers to create data-driven applications quickly. Objects
Now that you have had a chance to look at some of the Learn when to use the Error collec-
exciting features of ADO in Chapter 20, “Database Access tion for better error handling.
with ADO,” you are probably anxious to examine in more
detail the three objects that you will use with ADO: the
Connection, Command, and Recordset objects. The Command
and Recordset objects will be looked at in more detail in
the next two chapters. This chapter will examine the
Connection object. Part
The Connection object represents the connection to your V
data. It is the road to your data—in other words, it is the
Ch
way in which you will connect to your databases. By using
the Session and Application objects, you can implement 21
connection pooling with the Connection object. ■

27.1389-6 CH21 451 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 452 Chapter 21 Working with ADO’s Connection Object

In addition, you can execute SQL statements directly against your datasource. The Connection
object also contains the Error object, which you use for error handling. And best of all, the
Connection object allows for transaction processing.

Properties and Methods of the Connection Object

The Connection object contains a rich set of properties and methods for creating, maintaining,
and configuring itself. While some properties and methods (like Version) may not have many
uses, others will totally change how you create data-driven applications.

For instance, using the ConnectionString property, you can set up the string that contains the
connection information, and then send this information to the Session object for use through-
out the application. This allows you to develop your applications even more quickly.

Here are the properties and methods of the Connection object, along with a brief description.

Connection Object Properties

Attribute specifies whether a new transaction will be started after a transaction rollback or
commit. There are two values: adXactCommitRetaining, which specifies to automatically start a
new transaction after CommitTrans is called; and adXactAbortRetaining, which starts a new
transaction after RollbackTrans is called. Not all data providers will support this property,
though Microsoft’s Transaction Server will.
CommandTimeout determines how long to wait for a command to process (the default is 30
seconds). Setting this value to 0 makes the wait indefinite. This is a global value that applies to
all commands that are issued through the connection.

ConnectionString is a string that contains the information necessary to create the connection.
This string will contain information on the connection such as the DSN, provider, username,
password, and filename. When the connection is open, this is a read-only property.

ConnectionTimeout determines how long to wait to connect to the datasource (the default is 15
seconds). Set to 0 to wait indefinitely.

If the provider you are connecting to enables multiple databases, DefaultDatabase enables you
to specify which one is the default. This means that SQL commands issued to the provider will
have to specify which database to run the statement against if it is not the default.

IsolationLevel is the isolation level of the connection. This property applies after BeginTrans
is called. Use it to specify whether other transactions will be able to see your transactions ef-
fects and when (see Table 21.1).

27.1389-6 CH21 452 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Properties and Methods of the Connection Object 453

Table 21.1 Isolation Levels

Value Description
adXactUnspecified Isolation level of provider is unknown.
adXactChaos This value means that you cannot overwrite changes from
more highly isolated transactions.
adXactBrowse This means that you cannot view uncommited changes
from other transactions.
adXactReadUncommitted Same as above.
adXactCursorStability This value indicates that you can view changes from other
transactions after they have been committed.
adXactReadCommitted Same as above.
adXactRepeatableRead This value indicates that you can bring in a new recordset
by requerying.
adXactIsolated This value indicates that the transactions will be competely
isolated from one another.
adXactSerializable Same as above.

Mode is used to set the permissions on a connection, such as read-only or read-write. This prop-
erty can also be set to deny access to others.

Provider is the provider of a connection.

Version is the version of ADO being used.

Connection Object Methods

The methods of the Connection object give the developer quite a bit of power and freedom. For
instance, with the Execute method, you can execute a SQL statement against the datasource
without having to create a Command object.

BeginTrans begins a new transaction.

CommitTrans saves any changes and ends the current transaction. It may also start a new trans-
action if the Attributes property is set for this.

RollBackTrans rolls back the transaction; in other words, it cancels it. It, too, may start a new Part
transaction if the Attributes property is set for this. The following snippet of code shows an V
example of using all three of the preceding methods together.
Ch
DataConn.beginTrans ‘DataConn is our connection object
Rs.AddNew 21
Rs(“name”) = Keith
Rs(“phone”) = “555-1212”

27.1389-6 CH21 453 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 454 Chapter 21 Working with ADO’s Connection Object

Rs.Update
If DataConn.Errors = 0 then
DataConn.CommitTrans
Else
DataConn.RollbackTrans
End if

Close closes the connection. This method also closes any recordsets associated with the ob-
ject. Invoking this method looks like the following code.
DataConn.Close

Execute executes a command on the datasource and returns a recordset. This method can also
be used to find out how many records were effected by the command. This method can take
three optional parameters: CommandText, RecordsAffected, and Options. An example of this
method is seen in the following snippet.
myRecordset = dataConn.Execute(“SELECT * FROM tblMoney”, intNumRecs, adCmdTxt)

Now that you have called Execute, you can look at intNumRecs and see how many records were
returned. If you do pass a command text (such as the SELECT statement in this example), you
should also send the appropiate Options parameter telling the provider what type of command
this is.

Open opens the connection. This method is fairly simple to use. It can be sent three optional
parameters: ConnectionString, UserID, and Password. UserID and Password are just as they
say. ConnectionString is a more complicated parameter.

A ConnectionString is a string of name value pairs separated by semi-colons. The possible

names that can be set here are Provider, Datasource, User, Password, and FileName. The
following is an example of calling this method.
Set dataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN=myDataSource”

Linking to Your Data with the Connection Object

There are three basic uses for the Connection object: creating and maintaining a link to your
datasource, creating recordsets, and error handling. Later in this chapter, you will see how to
implement error handling and create recordsets. Now, let’s examine how to create a Connection
object.

In Listing 21.1, a Connection object is created with the CreateObject method of the Server
object. This object is then set to a variable named DataConn.

Listing 21.1 21code01.ASP—Creating a Connection object.

Set DataConn = Server.CreateObject(“ADODB.Connection”)

27.1389-6 CH21 454 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Linking to Your Data with the Connection Object 455

N O T E You will see the same line with slight modifications for the creation of the Command and
Recordset objects. If you’re like this author, you will tend to cut and paste similar lines
from one part of your code to another. Watch out that you don’t end up creating a second Connec-
tion object instead of a Recordset. ■

In Listing 21.1, a Connection object is created with the Server object’s CreateObject. Once
you have created a Connection object, you will probably want to assign it to the Session object.
Listing 21.2 does just that (again, in one line!).

Listing 21.2 21code02.ASP—Setting the Session property ConnObject.

Set Session(“ConnObj”) = DataConn

Later, in Chapter 24, “Managing Your Database,” you will see a better connection pooling than
what ODBC 3.0 gives. For now, this enables you to scope the Connection object across mul-
tiple pages. You could even set this as an Application object property and scope it across all
the sessions.

CAUTION
Sometimes you will create code that sets a Connection object to the Session. Don’t do this! In reality,
you will always want to use connection pooling as provided by ODBC 3.0 and above.

A Note from the Author

This author’s very first programming language was a spaghetti code version of BASIC found on the
Commodore Vic 20 computer. With this version of BASIC, all variables were global. This, of course,
would cause massive problems with debugging, and made creating interesting and useful application
almost impossible.
Next came C. With C, I learned to scope variables. (And learned the reason why my old BASIC
programs were so tough to decipher!) However, temptation would often lead to ignoring this feature
in C and still relying on variables of global scope. Of course, this soon lead to many of the same
problems I had before: programs too hard to maintain.
Around the time I learned C++, I also read Steve McConnell’s text Code Complete. Now, finally, I
understood why scoping was so important in the creation of programs that would stand the test of
time.
Part
How does this relate to ASP development? Well, it can be very easy to set an Application or
Session property, just as it can be very easy to set a global variable in a C++ program. And
V
sometimes, there are very good reasons to do so. But most of the time, it will pay off in the long run Ch
to scope your variables as small reasonable.
21

27.1389-6 CH21 455 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 456 Chapter 21 Working with ADO’s Connection Object

Configuring the Connection Object

Now that you have created a Connection object and set it to a Session property, you need to
configure it before you actually open the connection. There are several properties that you can
use to configure it. In this section, you will examine the ConnectionString, Mode, and
ConnectionTimeOut properties in terms of how they relate to configuring your connection
before opening it.

The ConnectionString Property From the list of Connection object properties, you may
remember the property ConnectionString. The ConnectionString is a string that contains
information on the data provider that is needed to connect to it. This string contains a set of
name value pairs delimited by semicolons. These name value pairs are optional.

A connection string generally will look something like Listing 21.3.

Listing 21.3 21code03.ASP—Setting the ConnectionString property.

Set DataConn.ConnectionString = “DSN=Customers;User=sa;Password=elmoisgod”

The above code tells the Connection object that the DSN it will connect to is called Customers.
The data provider will look for a user ID and password, so those are also set. As you can see, the
syntax when setting the ConnectionString property is argument = value. The following is a
list of the arguments you can set with the ConnectionString property:

■ Provider
■ User
■ Password
■ Filename

CAUTION
Set either the provider or datasource arguments, not both. They both provide similar functionality. If you
pass both to the ConnectionString property, your application may not work.

N O T E The ConnectionString property can be overridden with the Open method of the
Connection object. When you examine this method later, you will see this for yourself. For
now, just remember that you are not tied down to what is set with this property. ■

The Mode Property The Mode property is used to set the permissions for a connection. You
can also use it to find out what the permissions are. When the Connection object is closed, you
can set the permissions. When it is open, you can only read what the permissions are.

27.1389-6 CH21 456 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Linking to Your Data with the Connection Object 457

The most popular uses of Mode are those involving an application- or session-wide Connection
object. You may find it necessary in some parts of your application to set the permissions on
the connection to, say, read-only (for instance, when an administrative section of the application
is opened).

N O T E Mode can only be set while the connection is closed. When the connection is open, Mode
is read-only. ■

Table 21.2 summarizes the most popular modes of Mode.

Table 21.2 Mode Values

Value Description
adModeUnknown This value means that the Mode is unknown.
adModeRead This sets read-only permissions.
adModeWrite This sets write-only permissions.
adModeReadWrite This sets read and write permission for the connection.

ConnectionTimeout
The ConnectionTimeout property has a very simple job: to set the amount of time the connec-
tion object should wait to connect to the provider. However, knowing what value to set this
property to can be anything but simple. Listing 21.4 sets the property’s value.

Listing 21.4 21code04.ASP—Setting the ConnectionTimeout property.

Set DataConn.ConnectionTimeout = 30

The code above sets the ConnectionTimeout property to 30 seconds. This means that the
Connection object, DataConn, will wait 30 seconds before returning an error when it tries to
connect to a provider.

N O T E The ConnectionTimeout property’s default value is 15 seconds. ■

Generally, you do not need to change this value. However, you may design applications that
Part
reside on a server that has a slow pipe to the client. (For instance, your corporate presence
server may reside on a 128KB ISDN line.) In cases like these, you may want to bump up the V
value of the ConnectionTimeout property to 30 seconds, or even a minute. Ch

21

27.1389-6 CH21 457 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 458 Chapter 21 Working with ADO’s Connection Object

TIP Occasionally, you may even want to have the Connection object wait forever for the connection to
work out. If this is the case, set the ConnectionTimeout property to 0. However, if you do this, you
better be very sure your connection will eventually appear. Otherwise, your user will eventually get
tired of waiting for the page to appear. (Or you will get a script timeout, unless that, too, is set to wait
forever, which I really don’t recommend.)

The DefaultDatabase Property This interesting property sets the default database for a con-
nection. This can be confusing. Most developers are used to the idea of one datasource (or
provider), meaning one database. However, some providers may contain multiple databases.
When this is the case, you need to specify the database in all of your SQL statements. When no
database is specified, the default database is used.

TIP If you are dealing with a provider that uses multiple databases, I still recommend specifying the
database in your SQL string. Why? It will produce self-documenting code.
For instance, imagine you have a database called inventory and one called salesmen. You could
set your default database to inventory. Then, when you are selecting the number of some item, your
SQL statement would look something like this: SELECT numberFruit FROM tblFruit. Instead,
you could specify the database with SELECT numberFruit from inventory.tblFruit. This
way, someone else new to your code will know instantly which database tblFruit is part of.

The IsolationLevel Property This property is used to set the level of isolation in the connec-
tion. It is fairly easy to use, as Listing 21.5 shows.

Listing 21.5 21code05.ASP—Setting the IsolationLevel.

Set DataConn.IsolationLevel = adXactIsolated

Listing 21.5 sets the isolation level very high. With this level of isolation, all transactions take
place completely separate from other transactions.

IsolationLevel is the degree to which transactions are isolated from one another. A transac-
tion can be completely isolated from other transactions, or it can be set to some of the other
transactions also taking place. Table 21.3 summarizes some of the values the IsolationLevel
property can be set to.

Table 21.3 IsolationLevel Property Values

Value Description
adXactUnspecified This value means that it is impossible to determine the level
of isolation.
adXactChaos This value means that you cannot overwrite changes pending
from other transactions.

27.1389-6 CH21 458 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Linking to Your Data with the Connection Object 459

Value Description

adXactBrowse With this value, you cannot browse values from pending
transactions.
adXactIsolated Complete isolation for your transactions.

Controlling the Connection

Now that you have configured the Connection object by setting its various properties, you can
open and close the connection. You can also execute a SQL statement against the data provider
and set a recordset. And most importantly, you can set up transaction processing with the
Connection object’s methods.

Open and Close Opening the connection is very simple. Closing it is even simpler. Listings
21.6 and 21.7 show just how easy these methods can be to use.

Listing 21.6 21code06.ASP—Opening the connection...

DataConn.Open

Listing 21.7 21code07.ASP—...and closing it.

DataConn.Close
Set DataConn = nothing

TIP Notice how I have set the Connection object, DataConn, to nothing after closing it. Although not
strictly necessary, taking this additional step may help with troubleshooting later. Setting the object to
nothing releases the memory used for it; closing it does not.

However, don’t think there is nothing to these methods. The Open method provides developers
with many options in terms of setting properties for the connection right as it is opened, as
opposed to before. Listing 21.8 shows how you can configure and open a connection with the
Open method.

Listing 21.8 21code08.ASP—Using Open to configure the Connection object. Part

DataConn.Open “DSN=customers”, “sa”, “password”
V
Ch

21

27.1389-6 CH21 459 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 460 Chapter 21 Working with ADO’s Connection Object

In Listing 21.8, the Open method is being used to not only open to connection, but to also con-
figure the ConnectionString, User, and Password properties.

Even though there are no additional ways to use it, with Close, there are several issues to keep
in mind. First and foremost is that Close does not eliminate the Connection object you have
created from memory. It is still there, ready to be re-opened should the need arise. Second, any
recordsets that were associated with the connection will be closed. Finally, any Command objects
associated with the connection will be cleared of any parameters, along with their
ActiveConnection property.

CAUTION
Also, be sure to note that if a connection is closed during a transaction, the transaction will be cancelled.

Executing the Connection Using the Execute method of the Connection object, you can
create a recordset that is ready to go. Using Execute is easy; Listings 21.9 and 21.10 show two
different ways you can use it in your code.

Listing 21.9 21code09.ASP—Using Execute.

Set rs_main = DataConn.Execute “Select * from tbl_phoneNumbers”

In Listing 21.9, a SQL statement is executed against the provider. This SQL statement returns a
recordset, which the recordset, rs_main, is then set to. This is a great way to set up a recordset
with some data, and will probably be how you set a recordset most of the time.

Listing 21.10 21code10.ASP—Further uses of Execute.

DataConn.Execute “Delete * from tbl_phoneNumbers”, intRecs, adCmdTezt

In Listing 21.10, there is no recordset being set. This is because the SQL statement used this
time is a Delete, which returns an empty recordset. However, as developer, you may want to
know how many records were deleted. That is where the integer intRecs comes into play. You
could later use this number to tell the user how many records were deleted. For instance, you
might want to delete all the records for a particular product which you no longer carry. But if
you note how many records were deleted, you could then return that information to the user as
follows:
Number of records deleted = <%=intNumRecs%>

By passing intRecs to the Execute method, I can later test how many records were affected by
the command (affected in this case meaning deleted). Finally, to optimize the speed of execu-
tion of the command, I sent the value adCmdTxt as the Option parameter of the Execute
method. The Option parameter tells the provider what kind of command you are executing,
and allows the provider to save time by not having to figure it out itself.

27.1389-6 CH21 460 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Linking to Your Data with the Connection Object 461

TIP Whenever possible, send a value for the Option parameter! It can substantially increase the
performance of your applications.

Table 21.4 summarizes the various values you can send in the Option parameter.

Table 21.4 Values for the Option Parameter

Value Description
adCmdText This tells the provider that the CommandText will be a SQL
string.
adCmdTable This tells the provider that the CommandText is a table.
adCmdStoredProc This value enables the parameter to prepare for a stored
procedure.
adCmdUnknown This says that the CommandText is unknown. Don’t use this
value.

Transaction Processing with the Connection Object As you saw in Chapter 20, transaction
processing with the ADO is easy. All it takes are three simple steps:

1. Call the BeginTrans method to begin the transaction.

2. Do whatever data processing you need to.
3. Check for errors and then call either CommitTrans or RollbackTrans based on whether
there were any errors.

A Transaction Processing Example When should you use transaction processing? At the very
least, you should employ it when you are faced with creating applications that rely on mission
critical data. A popular and easy to understand example of this would be the banking software
example you saw in Chapter 20. Let’s examine this example in a little more detail, and see how
to code this transaction.

Imagine the portion of a banking application that deals with the transfer of funds from one
account to another. This program would need to delete the funds from the first account and
then add them to the second. But what if the power goes out after the delete? The funds would
never be accredited to the second account, and there would be no real way to find out what
went wrong.
Part
With transaction processing, both the delete and the adding of the funds have to be completed
before the transaction is committed to disk. This means that if the lights go out, the whole V
shebang is rolled back. The money is left in the first account, where it can be tried again later. Ch

21

27.1389-6 CH21 461 10/26/98, 9:40 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 462 Chapter 21 Working with ADO’s Connection Object

In Listing 21.11, you can see this process in action. The recordset used in this example has two
records, the first record is the account to take the money out of, the second is the account to
deposit the money into.

Listing 21.11 21code11.ASP—Transaction processing in action.

Set intAmount = 200 ‘ the amount to be transferred is $200
DataConn.BeginTrans
‘delete the money from the first account
Set tmpAmount = rs_main(“Balance”) - intAmount
Rs_main.AddNew
Rs_main(“Balance”) = tmpAmount
Rs_main.update

‘move the recordpointer to the account to add the funds to

rs_main.movenext
Set tmpAmount = rs_main(“Balance”) + intAmount
Rs_main.AddNew Rs_main(“Balance”) = tmpAmount
Rs_main.update
‘Check for errors, rollback if there are any, otherwise commit
If DataConn.Errors.Count = 0 Then
DataConn.CommitTrans
Else
DataConn.RollbackTrans
End If

Transaction Processing with the Attributes Property BeginTrans, CommitTrans, and

RollbakcTrans are easy to use. But what if you want to automatically start a new transaction
after each transaction ends? You use the Attributes property.

By setting the Attributes property to adXactCommitRetaining, you will start a new transaction
right after a CommitTrans is called. If you set it to adXactAbortRetaining, a new transaction will
start after RollbackTrans is called. If you are going to use transaction processing with all of
your database activity, I would recommend setting the Attributes property to both of these
values.

Other Benefits to Transaction Processing Besides the obvious benefit of helping to keep
data from being corrupted, transaction processing offers another significant reason for its use:
better performance. When you wrap your data processing around the code listing found in
Chapter 20, you can see some pretty hefty speed increases in your application. Why?

By keeping the transaction completely in memory, transaction processing only writes to disk
when CommitTrans is called. Because RAM is much faster than the hard drive, this enables
optimized disk writes only when needed. When they aren’t needed, they don’t happen (that is,
the transaction is rolled back), and when a disk write happens, it writes all the data out at once,
also saving time due to less movement by the hard disk head going out and finding the correct
spot to write the data several times.

27.1389-6 CH21 462 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Error Handling with ADO 463

Error Handling with ADO

Although ASP’s error-handling capabilities are rather sparse at this time, ADO provides its own
error handling through the Errors collection and Error object. The Error collection belongs to
the Connection object. It is further made up of none to many Error objects, one object for each
error.

Why is the Error collection part of the Connection object? Because the errors that are cap-
tured and stored in the Error collection are those errors that the data provider sends out.

The Properties and Methods of the Error Collection

While the Errors collection doesn’t offer many properties or methods, it does enable the devel-
oper to discover how many Error objects there are, and to specify any particular one. In addi-
tion, by using the Clear method you can clear the Errors collection of all Error objects cur-
rently in it.

■ Item—This property returns a specific Error object by number or name.

■ Count—This property holds an integer that is the number of Error objects (and thus
errors) in the collection.
■ Clear—This method clears the collection of all objects. An example of this method
would be:
dataConn.Errors.Clear

where dataConn is your Connection object, and Errors is how you refer to the collection.
It is a good idea to call this method before calling any transaction methods or the Filter
method of the Recordset object. This enables you to accurately determine the number of
errors those calls generate.

The Properties of the Error Object

The Error object holds information about a specific error that was generated. The Error object
has no methods, and has only a few properties that are used to gain more information about
the error raised.

■ Description—The text of the error.

■ Number—A number that corresponds to a unique error condition.
■ Source—The object that raised the error.
■ HelpFile and HelpContext—If they exist, these properties hold the help file and topic Part
that contain more info on the error.
V
■ SQLState and NativeError—These provide ANSI SQL codes that further describe the
Ch
error.

So, how do you use the Errors collection and Error object to intelligently handle errors? List-
21
ing 21.12 shows a small example.

27.1389-6 CH21 463 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 464 Chapter 21 Working with ADO’s Connection Object

Listing 21.12 21ASP12.ASP—Error handling with ASP.

MyRecordset.AddNew
MyRecordset(“firstName”) = “Heather”
MyRecordset(“lastName”) = “Urschel”
MyRecordset.Update

If dataConn.Errors.Count > 0 then

Response.Write(“The following errors occurred:<BR>”)
For each Error in dataConn.Errors
Response.Write(“Error #: ” & Error.Number & “<BR>”)
Response.Write(“Error Description: “ & Error.Description & “<BR>”)
Next
Else
Response.write(“Update Successful!”)

Table 21.5 provides a list of the error codes and what they mean. Use it to intelligently program
your error-handling routines to react based on what errors they receive.

Table 21.5 Error Types

Error Description
adErrInvalidArgument The arguments are of the wrong type or are out of
range.
adErrNoCurrentRecord There is no current record. In other words, BOF or
EOF is true.

adErrIllegalOperation The operation requested is not allowed.

adErrInTransaction You are trying to close a connection during a
transaction.
adErrFeatureNotAvailable The operation is not featured by the provider.
adErrItemNotFound The item you specified was not found.
adErrObjectNotSet The reference you are using no longer points to an
object.
adErrDataConversion You are using the wrong datatype.
adErrObjectClosed The object is closed, so the operation is not allowed.
adErrObjectOpen Same as above, only now the operation is not allowed
because the object is open.
adErrBoundToCommand This error is generated if the ActiveConnection of a
recordset attempts to change when the recordset
was created with a Command object.

27.1389-6 CH21 464 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 Revisiting the Employee Phone Application 465

Error Description
adErrInvalidParamInfo The parameter wasn’t defined correctly.
adErrInvalidConnection The operation requested won’t work, because there
is no connection, or the connection doesn’t allow it.
adErrProviderNotFound The provider specified was not found.

As you can see, the Error object and the Error collection have quite a few properties that can
help with error handling. If you remember from Listing 21.11, a simple error-handling routine
was in place to help with the transaction processing. This error handling could be beefed up to
examine any errors if they were created and try to find a way to re-execute the transaction
correctly the next time.

The Command Object and the Connection Object

The Command object can be optional. It isn’t needed to create a connection, nor to open a
recordset. It really has a fairly specialized purpose in comparison to its two sister objects, the
Connection and Recordset objects. It is used to pass parameters to stored procedures.

This means that the Command object can be used to create a recordset. And, it is also closely tied
to the Connection object. Listing 21.13 illustrates this close relationship.

Listing 21.13 21code12.ASP—The Command object.

Set DataCmd = Server.CreateObject(“ADODB.Command”)
Set rs_tmp = Server.CreateObject(“ADODB.Recordset”)
DataCmd.CommandType = adCmdText
Set DataCmd.ActiveConnection = DataConn
Rs_tmp.Open DataCmd, , 0, 1

In Listing 21.13, the Command object’s property ActiveConnection is set to the Connection
object DataConn. Then the recordset rs_tmp is opened with one of the parameters being the
Command object. Both the Command and Recordset objects are created and utilized through the
Connection object DataConn.

Revisiting the Employee Phone Application Part

Now let’s revisit the phone application you created in Chapter 20. You can probably see several
places where you can improve the code. Using your newly found knowledge of the Connection
V
object, examine this new and improved version of results.asp (see Listing 21.14). Ch

The changes to this file from the code in Chapter 20 have been made bold. As you can see, the 21
big change is that you are now more closely tuning the Connection object’s properties to
match your situation (in this case the higher than usual traffic that our site will be getting)
instead of relying on defaults.

27.1389-6 CH21 465 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 466 Chapter 21 Working with ADO’s Connection Object

Listing 21.14 21code13.ASP—A new and improved results.asp.

<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<HEAD>
<TITLE>
Search For Employee - By Last Name
</TITLE>
</HEAD>
<BODY>

<%
Set DataConn = Server.CreateObject(“ADODB.Connection”)

‘Set our connection and command timeouts higher, due

‘to all the heavy traffic we will be getting
DataConn.ConnectionTimeout = 60
DataConn.CommandTimeout = 60

DataConn.Open “DSN=Employees;User=Keith;Password=HiElizabeth”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
DataCmd.CommandText = “SELECT * FROM tbl_employees WHERE LastName = ‘“ &
Request.Form(“Last_Name”) & “‘“
Set DataCmd.ActiveConnection = DataConn
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
set rs_main = DataCmd.execute
%>

<TABLE>
<TR>
<TD>First Name</TD>
<TD>Last Name</TD>
<TD>Phone</TD>
</TR>

<%
do While not rs_main.EOF
Response.write(“<TD>”)
Response.write(rs_main(“FirstName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“LastName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“Phone”))
Response.write(“</TD>”)
Rs_main.moveNext
loop
%>
</TABLE>

</BODY>
</HTML>

27.1389-6 CH21 466 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 From Here… 467

The connectionTimeout value has been raised to enable the heavy traffic we know this appli-
cation will take. The connection closes itself at the end of the code.

From Here…
In this chapter, you have learned how to use the Connection object to connect to your data-
bases. You are now ready to begin implementing the basics of error handling with the Error
collections. Further, you have explored the many properties and methods of the Connection
object and how to use them in your applications. Using the Session and Application objects in
conjunction with the Connection object, you have learned how to implement connection pool-
ing in your applications.

■ Chapter 22, “Using Command Objects,” will show you how and when to use this object for
better performance and ease of development.
■ Chapter 23, “Working with ADO’s Recordset Object,” will cover how to use the work-
horse of ADO—the recordset.

Part
V
Ch

21

27.1389-6 CH21 467 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 468 Chapter 21 Working with ADO’s Connection Object

27.1389-6 CH21 468 10/26/98, 9:41 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH21 lp#3
 22
469

C H A P T E R
Part
V
Ch

22

Command Objects
Using Command Examine the properties and meth-
ods of the Command object.

Objects The Parameters Collection

Using the Parameters collection can
speed up your programs and your
development time.

Chapter 21, “Working with ADO’s Connection Object,”

showed you how to create an active connection to your
data with the Connection object. Using the Session and
Application objects and connection pooling, you saw
how to optimize the link to your data and keep traffic
down. Further, you saw how to execute a SQL statement
against the datasource with the Connection object.

In Chapter 23, “Working with ADO’s Recordset Object,”

you will learn how to manipulate the recordset objects you
have been creating in these last two chapters. There,
you will find the true richness and usefulness of ADO
and ASP.

In this chapter, you will learn how to execute commands

against your datasource in an even more controlled and
optimal fashion. You will create a collection of parameters
that can be fed to a stored procedure in your database.
You will also see how we can bypass creating a Connec-
tion object and implicitly create one with the Command
object.

What is the Command object? It is the ADO object that

represents a command you want to execute against your
database. With it, you can create recordsets quickly and
easily—whether that recordset is based on a table, a SQL
statement, or a stored procedure. ■

28.1389-6 CH22 469 10/26/98, 9:41 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 470 Chapter 22 Using Command Objects

The Command Object

There is one ADO object that your program wouldn’t ever have to use: the Command object. The
Command object has a slightly more specialized use than the Connection and Recordset objects.
Its best use (and what it was designed for) is sending parameters to stored procedures and
queries in your database.

You can create and use the Command object in several different ways. In this section, I examine
in greater detail the properties and methods of the Command object, and how they can help you.

Creating a Command Object

There are two major ways to create a Command object. The first is to explicitly create a Connec-
tion object. The second is to implicitly create a Connection object. When you explicitly create
a Connection object, you use the Server.CreateObject method, and set a variable to its return.
From then on, you can manipulate the connection via that variable. When you implicitly create
a Connection object, there is no call to Server.CreateObject, and you do not have the ability
to manipulate your connection (due to the fact that you have no variable to use).

Listing 22.1 shows the creation of a Command object, without first creating a Connection object.

Listing 22.1 22code01.ASP—Creating a Command Object without the

Connection object.
Set DataCmd = Server.CreateObject(“ADODB.Command”)

‘Customers is the DSN

DataCmd.ActiveConnection = “Customers”
DataCmd.CommandText = “qryListAllCustomers”

Set rs_tmp = Server.CreateObject(“ADODB.Recordset”)

rs_main = DataCmd.Execute

Listing 22.2 shows the creation of a Command object using a predefined Connection object.

Listing 22.2 22code02.ASP—Creating a Command object with a

Connection Object.
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN=Employees”

DataCmd.ActiveConnection = DataConn
DataCmd.CommandText = “qryListAllCustomers”

Set rs_tmp = Server.CreateObject(“ADODB.Recordset”)

Set rs_main = DataCmd.Execute

28.1389-6 CH22 470 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Command Object 471

In Listing 22.1, after creating the Command object, the ActiveConnection property is set to a Part
Datasource(DSN) called Customers. Then, its CommandText property is set to a query in the
database called qryListAll. This query lists all the customers in the database, and sorts them
V
by region. The recordset rs_main is then set to the results of that query. Ch

N O T E Don’t think that there wasn’t a Connection object created in Listing 22.1. One was
22
created; it was created implicitly, which means you have no real way to access it for things
such as error handling. ■

TIP There is really no reason to not first create a Connection object before creating a Command object. If
you don’t, you won’t have access to the Error collection, or be able to optimize and configure the
Connection object that is implicitly created.

In Listing 22.2, a Command object is created. Its ActiveConnection property is then set to the
Connection object DataConn, which was created earlier. After that, it is very similar to the other
listing; a query is run that results in the recordset rs_main.

Properties and Methods of the Command Object

Table 22.1 lists the properties and methods you can use to configure the Command object.

Table 22.1 Command Object Methods and Properties

Method/Property Description
ActiveConnection Used to set which connection the Command object is part of.
CommandText Used to specify the command the Command object will use.
CommandTimeout This property specifies how long to wait before timing out a
command.
CommandType This property tells the provider what type of command is stored in
CommandText.

Prepared Used to tell the provider to create a compiled version of the com-
mand for later use.

ActiveConnection This property indicates the Connection object that the Command object will
be a part of. The string that this property is set to can be either a Connection object, or the
definition of a connection.

Listing 22.3 demonstrates setting the ActiveConnection property to an already existing

Connection object.

28.1389-6 CH22 471 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 472 Chapter 22 Using Command Objects

Listing 22.3 22code03.ASP—Setting the ActiveConnection to a

Connection object.
Set dataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open = “myDSN”
DataCmd.ActiveConnection = DataConn

Listing 22.4 shows setting a Command object’s ActiveConnection property with a connection
string.

Listing 22.4 22code04.ASP—Setting the ActiveConnection with a

connection string.
DataCmd.ActiveConnection = “DSN=Customers;User=sa;password=bob”

Notice in Listing 22.3 that the ActiveConnection property is set to the Connection object
DataConn. In Listing 22.4, a connection string is used. This connection string should look famil-
iar to you. It accepts the same arguments as the Connection object’s ConnectionString prop-
erty. This string is passed to the ConnectionString property of the Connection object that is
implicitly created.

Usually, you will want to set your Command object’s active connection property to an already
existing connection. Not doing this should be the rare exception.

TIP Here’s a neat tip for you: Imagine you are setting up an ASP application on a remote machine, and you
don’t have a way to create an ODBC datasource on the machine. You can set the
ConnectionString to the following to create on-the-fly in your code:
“DRIVER={Microsoft Access Driver (*.mdb)}; DBQ=c:\path\database.mdb”

Here, you specify the driver needed for the database—and you also tell the system the location of the
database file itself.

CommandText CommandText is a string value that represents the command you want to ex-
ecute against the provider. It can be a SQL statement, the name of a table, or the name of a
stored procedure.

In Listing 22.5, you can see an example of setting the CommandText property to a SQL statement.

Listing 22.5 22code05.ASP—Setting the CommandText property to a

SQL statement.
DataCmd.CommandText = “SELECT * FROM tbl_customers”

28.1389-6 CH22 472 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Command Object 473

In Listing 22.6, you can see an example of setting the CommandText property to a table name. Part
V
Listing 22.6 22code06.ASP—Setting the CommandText property to a Ch
table name.
DataCmd.CommandText = “tbl_customers”
22

In Listing 22.7, I set the CommandText property to the name of an Access query called
qryCustomers.

Listing 22.7 22code07.ASP—Setting the CommandText property to a query.

DataCmd.CommandText = “qryCustomers”

Each of the above listings does the same thing. Each creates a recordset based on the records
in the table named tbl_customers. Later, when you examine the Prepared property, you will
see how you can have ADO compile the CommandText for you.

When would you use a stored procedure? When do you use a table name? And, when is it a
good idea to use SQL text? In general, try not to use table names. This can result in way too
much information being transported around.

The big choice is between stored procedure and SQL statement. Stored procedures on data-
base servers such as SQL Server can execute more quickly. SQL statements, however, are
more readable. Then again, you can always comment your code.

CommandType Now that you have set the CommandText property to either a SQL statement,
stored procedure, or table name, you need to set the CommandType property. This property tells
the provider what type of command is being set. This allows for a faster execution time when
the provider executes the command.

N O T E If the CommandType does not accurately reflect the command held in CommandText, an
error will be returned when the Execute method is called.
Why? Knowing what type of command is coming down the pike means that the provider won’t have to
spend any time trying to figure it out. In Listing 22.8, you can see what to set the CommandType to if
the CommandText property is a SQL statement. ■

Listing 22.8 22code08.ASP—Setting the CommandType property.

DataCmd.CommandType = adCmdText

Table 22.2 lists the possible values that CommandType can take. You should always use these
values, as opposed to their numeric equivalent, because the integer they represent may
change.

28.1389-6 CH22 473 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 474 Chapter 22 Using Command Objects

Table 22.2 Values for the CommandType Property

Value Description
AdCmdText This tells the provider that the CommandText will be a SQL string.
AdCmdTable This tells the provider that the CommandText is a table.
AdCmdStoredProc This value enables the parameter to prepare for a stored procedure.
adCmdUnknown This says that the CommandText is unknown. Don’t use this value.

CommandTimeout The property CommandTimeout specifies the amount of time to wait for the
command to execute. This property is fairly straightforward. Often, you will not need to
change it; however, there may be circumstances where you will need to set it to something
other than its default. Listing 22.9 sets the property’s value.

Listing 22.9 22code09.ASP—Setting the ConnectionTimeout property.

Set DataCmd.CommandTimeout = 30

In Listing 22.9, the CommandTimeout property is set to 30 seconds. The application will now wait
30 seconds for the command to execute before it times out.

N O T E The CommandTimeout property’s default value is 15 seconds. ■

If your application is residing on a Web server that is running slowly (perhaps from the mil-
lions of hits you are receiving), it might be worth your while to increase this property’s value to
30 seconds. Generally, however, you will want to keep this value set to its default.

TIP Occasionally, you may even want to have the Command object wait forever for the connection to work
out. If this is the case, set the ConnectionTimeout property to 0. However, I would highly recom-
mend against doing this.

Prepared This is a Boolean value that tells the provider to prepare a compiled version of the
command. This will slow down the command the first time it is run, but after that, the provider
will execute the command more quickly. Listing 22.10 shows how to set this property.

Listing 22.10 22code10.ASP—Using the Prepared property.

DataCmd.Prepared = True

N O T E Notice that not all providers will be able to create a compiled version of the command. If
the provider doesn’t support that functionality, it will ignore any attempts to set this
property. In that case, if queried, the Prepared property will return a value of False. ■

28.1389-6 CH22 474 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Command Object 475

Execute This is the method that actually executes the command contained in CommandText. It Part
returns a recordset. If the command is an update or some other command that isn’t supposed
to return any records, a null recordset will be returned. Listing 22.11 is an example of how to
V
call Execute. Ch

22
Listing 22.11 22code11.ASP—Executing the command.
Dim inNumRecs
Set Rs_main = DataCmd.Execute intNumRecs, , adCmdTExt

Listing 22.11 sets the recordset rs_main to the results of the command. In addition, the variable
intNumRecs is set to the number of records affected by the command. You can later examine it
and see how many records were deleted (for instance, assuming that was the command sent).
The last value, adCmdText, tells the provider what kind of command is coming. This is the same
value you set the CommandType to.

The general format used when you call Execute is:

command.Execute RecordsAffected, Parameters, Options

where RecordsAffected is a variable you can later query to find the number of records in-
volved in your command, Parameters is a set of parameters to feed the command, and Options
tells the provider what type of command will be executed.

Using the Command Object to Access Data

Now that you have examined some of the properties and methods of the Command object, you
are now ready to use it to create recordsets. Listing 22.12 is an example of how to set a
recordset with a Command object.

Listing 22.12 22code12.ASP—Accessing data with the Command object.

Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open “DSN=Employees”

DataCmd.ActiveConnection = DataConn
DataCmd.CommandText = “SELECT * FROM tbl_employees where fname = Joe”

Set rs_main = Server.CreateObject(“ADODB.Recordset”)

Set rs_main = DataCmd.Execute intRecs, , adCmdTxt

In Listing 22.12, all employees with the first name Joe are put into a recordset called rs_main.
Now, when the Execute method is called, it includes the arguments intrRecs and adCmdTxt.
The argument intRecs can be queried later to find out how many records are in the newly
created recordset, and adCmdTxt told the provider that it was being sent a SQL string to
execute.

28.1389-6 CH22 475 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 476 Chapter 22 Using Command Objects

The Parameters Collection and the Parameter Object

The Command object is best used for sending parameters to stored procedures or queries of the
providers. To provide this functionality, the Command object includes a collection of Parameter
objects. Using the CreateParameter method, you can very easily take advantage of stored
procedures for faster performance.

A stored procedure (or query) is a command that has been compiled by the data provider and
stored in the database. There are two types of stored procedures: parameterized and non-
parameterized queries.

A parameterized query is a stored procedure that is expecting arguments. For example, the
procedure could list all the doctors in the table tbl_doctors from a specific region. The param-
eter in this case would be the region.

Methods and Properties of the Parameters Collection

The Parameters collection contains several methods and properties that you can use to add
and delete new parameters. Some of these, like Item, are common to all collections in ADO
(see Fields collection in Chapter 23 for an example), whereas others are very specific to the
Parameters collection, such as CreateParameter (see Table 22.3).

Table 22.3 Parameters Collection Methods and Properties

Method/Property Description
Item This property returns the parameter specified.
Count This property returns the number of parameters.
CreateParameter This method is used to create new parameters.
Append This method is used to add a new parameter to the collection.
Delete This method is used to delete a parameter from the collection.
Refresh This method is used to refresh the collection.

Item This property returns a specific Parameter object by number or name.

Count This property holds an integer that is the number of Parameter objects in the collection.
CreateParameter This method creates a Parameter object. You can set most of the attributes
of the object with this method or you can choose to set them later, before you append it to the
collection. The syntax for this method is
Set parameter = command.CreateParameter(Name, Type, Direction, Size, Value)

Notice how this method takes five different values. These values are the Name of the parameter,
the Type of the parameter, the Direction of the parameter (input or output), the Size of the
parameter, and the Value.

28.1389-6 CH22 476 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Parameters Collection and the Parameter Object 477

Part
CAUTION
If the datatype can be of variable length, you must set the Size attribute also. If you don’t, an error will
V
occur. Ch

22
Append This method is used to insert the parameter in the collection after you have created
the parameter with the CreateParameter method. You must specify a type for the parameter
before you append it, or you will receive an error. The following is the form the Append method
should take:
dataCmd.Append myNewParameterObject

Delete The Delete method is used to remove a parameter from the collection. You must
refer to the parameter by name when using this method:
DataCmd.Delete NewParameterObject

Refresh This method queries the provider for information on the parameters the stored
procedure in the CommandText property will take. If you try accessing the Parameters collection
before calling this method, it will be called for you. The following code shows an example of
calling this method:
DataCmd.Refresh

CAUTION
When Refresh populates with variable length parameters, you should still explicitly set the size value.

Methods and Properties of the Parameter Object

The methods and properties of the Parameter object enable you to configure to a great degree
of detail the parameter you will send. For instance, you can set the direction of the parameter,
and even the size in bytes of the parameter’s value (see Table 22.4).

Table 22.4 Parameter Object’s Methods and Properties

Method/Property Description
Attributes This property sets what type of values the parameter will accept.
Direction This property specifies in which direction the parameter is.
Name This property is the parameter’s name.
NumericScale This property sets the how many decimal points the parameter may
contain.
Precision This property specifies how many digits the parameter may be.
Size This property specifies how big the parameter may be.
continues

28.1389-6 CH22 477 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 478 Chapter 22 Using Command Objects

Table 22.4 Continued

Method/Property Description
Type This property is used to set the data type of the parameter.
Value This property is used to set the value of the parameter.
AppendChunk This method is used to append a chunk of bytes to the parameter.

Attributes This property sets what kind of parameter attributes the parameter has. It takes a
ParameterAttributesEnum value, of which the following are possible values. Table 22.5 lists the
various possible values.

Table 22.5 Values for the Attributes Property

Value Description
AdParamSigned This is the default value. It indicates that the parameter accepts
signed values.
AdParamNullable This indicates that the parameter will accept NULL values.
AdParamLong This value indicates that the parameter will accept long binary data.

Direction This property indicates whether the parameter is an input, output, or return value.
It can take any of the values listed in Table 22.6.

Table 22.6 Values for the Direction Property

Value Description
AdParamInput This value indicates that the parameter is an input parameter.
AdParamOutput This indicates that the parameter is an output parameter.
adParamInputOutput This value indicates that the parameter is both an input and an
output parameter.
adParamReturnValue This value indicates that the parameter is a return value from the
stored procedure.

Name This is the name of the parameter. Before you append the parameter, this value is
read-write. Once you have appended it, it is read-only.

Numeric Scale The value of the Numeric Scale property sets how many points to the right of
the decimal place the parameter will be resolved to. So, for instance, let’s say it is pi, which goes
on forever to the right of the decimal place. By setting this value to 2, you would get back 3.14.

28.1389-6 CH22 478 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Parameters Collection and the Parameter Object 479

Precision The Precision property works much like the Numeric Scale property. However, Part
Precision determines the total number of digits. In the previous example, if the Precision
property were set to 2, the value for the parameter would be 3.1.
V
Ch
Size This property sets the maximum size of the parameter. This value will be an integer that
represents the size in bytes or in characters, as the case may be. 22
Type This value for this property is the type of data the parameter is. Table 22.7 summarizes
the various types of values this property can be set to.

Table 22.7 Values for the Type Property

Value Description
adBigInt An 8-byte signed integer.
adBinary A binary value.
adBoolean A Boolean value.
adBSTR A null-terminated character string.
adChar A string value.
AdCurrency A currency value.
adDate A date value.
adDBDate A date value in the format yyyymmdd.
adDBTime A time value.
adDBTimeStamp A date-time stamp.
adDecimal An exact numeric value.
adDouble A double-precision floating-point value.
adEmpty No value was specified.
adError A 32-bit error code.
adGUID A globally unique identifier.
adIDispatch A pointer to an IDispatch interface on an OLE object.
adInteger A 4-byte signed integer.
AdIUnknown A pointer to an IUnknown interface on an OLE object.
adLongVarBinary A long binary value.
adLongVarChar A long string value.
AdLongVarWChar A long null-terminated string value.
adNumeric An exact numeric value.
continues

28.1389-6 CH22 479 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 480 Chapter 22 Using Command Objects

Table 22.7 Continued

Value Description
adSingle A single-precision floating-point value.
adSmallInt A 2-byte signed integer.
adTinyInt A 1-byte signed integer.
adUnsignedBigInt An 8-byte unsigned integer.
adUnsignedInt A 4-byte unsigned integer.
adUnsignedSmallInt A 2-byte unsigned integer.
adUnsignedTinyInt A 1-byte unsigned integer.
adUserDefined A user-defined variable.
adVarBinary A binary value.
adVarChar A String value.
adVariant An OLE Automation Variant.
adVarWChar A null-terminated Unicode character string.

Parameters in Action
Now that you have covered the properties and methods of the Parameter object and collection,
let’s look at an example of implementing the great functionality this object gives us. Listing
22.13 creates a Parameter object and then appends it to the collection.

Listing 22.13 22Code13.ASP—A new and improved results.asp.

‘setting the name and value of the parameter
StrName = “last_name”
StrValue = “Jones “

‘setting the type, direction, and size

intType = adChar
intSize = 50
intDirection = adParamInput

‘creating the parameter object

dataParam = DataConn.CreateParameter(strName, intType, intDirection, intSize,
strValue)

‘appending the parameter to the collection

DataConn.Append DataParameter
Set rs = DataConn.Execute(,,adCmdStoredProc

28.1389-6 CH22 480 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 The Parameters Collection and the Parameter Object 481

As you can see, I created a Parameter object by using the CreateParameter method of the Part
Command object. I then appended this to the Parameters collection. After this, I call the Execute
method of my Command object, which returns a recordset that I call rs. Now, I can use this
V
recordset as I would any other recordset. Ch

22
Revisiting the Phone Application
Let’s see how to apply your new and improved knowledge of the Command object to the em-
ployee phone database from the previous two chapters. So far, this application has been relying
on SQL statements to create the recordset rs_main. Now, try using a parameterized query,
passing the parameters as part of the Parameters collection.

In Listing 22.14, you have the new and revised results page of your phone-book application.
This time around, you are using the Parameter object to speed up your development time and
your program’s execution.

Listing 22.14 22Code14.ASP—A new and improved results.asp.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>
Search For Employee - By Last Name
</TITLE>
</HEAD>
<BODY>

<!-- #include file=”adovbs.inc” -->

<%
Set DataConn = Server.CreateObject(“ADODB.Connection”)

‘Set our connection and command timeouts higher, due

‘to all the heavy traffic we will getting
DataConn.ConnectionTimeout = 60
DataConn.CommandTimeout = 60

DataConn.Open “DSN=Employees;User=Keith;Password=HiElizabeth”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
Set DataCmd.ActiveConnection = DataConn

‘setting the name and value of the parameter

StrName = “last_name”
StrValue = Request.Form(“last_name”)

‘setting the type, direction, and size

intType = adChar
intSize = 50
intDirection = adParamInput

continues

28.1389-6 CH22 481 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 482 Chapter 22 Using Command Objects

Listing 22.14 Continued

‘creating the parameter object
set dataParam = dataCmd.CreateParameter(strName, intType, intDirection, intSize,
strValue)

‘appending the parameter to the collection

dataCmd.Parameters.Append dataParam
dataCmd.CommandText = “qryLastName”
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
Set rs_main = DataCmd.execute
%>

<TABLE>
<TR>
<TD>First Name</TD>
<TD>Last Name</TD>
<TD>Phone</TD>
</TR>

<%
do While not rs_main.EOF
Response.write(“<TD>”)
Response.write(rs_main(“FirstName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“LastName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“Phone”))
Response.write(“</TD>”)
Rs_main.moveNext
loop
%>
</TABLE>

</BODY>

The above code will perform even faster than it did before. Why? When you use a query that is
part of the data provider, the time it takes to gather the data drops significantly.

From Here…
In this chapter, you delved further into the Command object. With it, you can directly link to data
providers, instead of having to explicitly create a Connection object. You can even connect to a
provider that hasn’t been set up as an ODBC datasource.

With the Parameters collection, you have seen how to take advantage of your data provider’s stored
procedures and queries. This little object can speed up your applications quite impressively.

28.1389-6 CH22 482 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 From Here… 483

In the following chapter, you will examine the Recordset object. This object will be the one you Part
work with most of the time. Its powerful methods and properties will allow you to manipulate
your data with greater ease than you have ever experienced.
V
Ch
Later chapters will discuss how to optimize your datasources, and use the Advanced Data
Connector and Microsoft’s Transaction Server in your applications. 22
■ Chapter 23, “Working with ADO’s Recordset Object,” will prepare you to use this object
in your applications.
■ In Chapter 24, “Managing Your Database,” you will examine how to optimize your
database with Microsoft’s Transaction Server, ODBC 3.0, and look at other issues related
to databases.

28.1389-6 CH22 483 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 484 Chapter 22 Using Command Objects

28.1389-6 CH22 484 10/26/98, 9:42 AM

Brands/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH22 lp#3
 23
485

C H A P T E R

Part
Retrieving Data V
Working with ADO’s Examine the issues that surround
creation of a Recordset object.
Ch

23
Recordset Object Recordset Properties
Learn the properties of the
Recordset object.

Navigating the Recordset

See how to swim through your data.
The Recordset object contains some
In the past two chapters, you have examined the great methods for accessing the
data you need.
Connection and Command objects. These objects exist for
one reason: to create a Recordset object that you will
Manipulating Your Database
manipulate and navigate through.
Discover the methods to manage
Now, you can manipulate your data with the Connection and manipulate your data.
and Command objects, but there is only so much you can
do. Inserting data is often easy with Parameter collection,
but when you need to display lots of data and then manipu-
late several records, the Recordset object is really the
interface to use.

This chapter will introduce you to the methods and prop-

erties of the Recordset object. As you read through the
next few pages, keep in mind that ADO was designed for
many different and disparate types of data—not just the
tabular data many of us are use to. ■

29.1389-6 CH23 485 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 486 Chapter 23 Working with ADO’s Recordset Object

Retrieving Data into a Recordset

There are several issues that should be examined before creating a recordset. Basically, it is a
battle between performance and integrity. You, as the developer, need to find a happy medium
for the applications you develop.

When developing for an Internet site that will be hit by thousands (or even hundreds of thou-
sands) of people every day, speed will probably be much more of an issue than it is when devel-
oping for a small intranet. By understanding how your database works and how ADO retrieves
data, you can save yourself major headaches.

Integrity versus Performance

When examining the issue of data integrity versus application performance, there is one overwhelm-
ing thought that always comes to mind (of course, this is just my opinion): If your data isn’t protected
and becomes corrupt, it is useless. And computers get faster all the time.
In other words, choose integrity over performance, because you can always find faster hardware, but
you can’t always retrieve corrupt data.
On the other hand (and this is where it gets sticky), if your application is too slow, no one will use it.
Happily, ADO enables developers to strike a happy, nice medium.

To help you in this task, I will examine isolation of data and cursor types. Isolation is how much
you insulate your data connections from each other. The type of cursor you choose to create
will be the major way in which you decide the performance versus integrity battle.

Isolation
ADO’s Connection object includes a property called IsolationLevel that is used to specify
exactly how much isolation a transaction will have and give. Table 23.1 summarizes the various
values this property can be set to.

As you may remember, isolation is an attribute of transactions. When a transaction is com-

pletely isolated, it cannot see what other transactions are doing, and other transactions cannot
see it.

For instance, in the midst of transaction #1, you may change the number of widgets on hand
from 100 to 99. In transaction #2, you may ask for the number of widgets on hand. If the trans-
actions are both completely isolated, then transaction #2 will think there are 100 widgets still.
This example seems to portray isolation as a bad thing, but there are times when you will want
your transactions isolated from one another.

29.1389-6 CH23 486 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Retrieving Data into a Recordset 487

Table 23.1 Popular IsolationLevel Values

Value Description
AdXactUnspecified This value means that it was impossible to determine the level
of isolation.
AdXactChaos This value means that you cannot overwrite changes pending
from other transactions. Part

AdXactBrowse With this value, you cannot browse values from pending V
transactions. Ch
Complete isolation for your transactions.
AdXactIsolated
23
By choosing the appropriate level of isolation, you can avoid dirty reads and phantom records.
Phantom records are records that used to exist that have been deleted by other transactions.
Dirty reads occur when you read a value that has been changed by another transaction.

Locking
Locking is where the developer specifies how much and to what level the data in the database
will be locked up and inaccessible by other recordsets. Why should you care about locking?
Probably the best way to answer this question is to examine the various types of locks you can
apply.

To begin with, you create a lock when you open up the recordset. One of the arguments in the
Open command is the type of locking you want to employ. The following snippet of code creates
an optimistic lock (don’t worry, I will explain what that is in a moment).
RsMaim.Open DataCmd, DataConn, adOpenForwardOnly, adLockOptimistic

Here’s an example of how to correctly apply locking. Suppose you are editing the number of
widgets in your factory from 100 to 99 because someone just bought one. You will probably
want to lock the record that contains the number of widgets before you actually reduce the
number, and then unlock it once you are done.

If you didn’t do this, you might first read in the number currently available (100) and then
reduce it by one and write that new number (99) back to the database. But what if in between
that time, someone else bought a widget and they reduced the number from 100 to 99 already?
Then you would write 99 again, and the count would be off by one! Instead, you lock the record
before getting the number currently available, and keep it locked until you have written the
new number. Then, the next transaction can do the same thing, this time reducing 99 to 98.
Thus everything remains accurate.

Identifying Cursor Types

There are four types of cursors in ADO. By using a combination of IsolationLevel and the
appropriate cursor type, a developer can exert a great level of control over the access and
manipulation of his data. Table 23.2 lists these cursor types.

29.1389-6 CH23 487 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 488 Chapter 23 Working with ADO’s Recordset Object

Table 23.2 Cursor Types for Recordsets

Value Description
Forward-Only This is the default type of cursor. It provides for forward-only
and read-only movement through the recordset. It is also the
quickest cursor type.
KeySet This cursor is read- and write-enable. It also enables move-
ment both ways through the recordset. It is unlike a dynamic
cursor in that you cannot see deletions and additions of other
recordsets.
Dynamic Like the KeySet cursor, this cursor is read-and write-enabled.
Unlike the KeySet, it can see the deletions, additions, and
changes of other recordsets. Using a dynamic cursor exacts
the highest overhead.
Static A static cursor is read and write-enabled. It can also move
throughout the recordset set in either direction. However,
additions and deletions by other users are not visible to this
cursor type.

Let’s examine each of these in a little more detail and see how to create each type of cursor. I
hope a more detailed examination will also help you see when to employ each type of cursor.

Forward-Only Cursor The forward-only cursor is the default. When you create recordsets
with the Connection or Command objects, it is this type of cursor that you received back. In the
line of code below, a recordset with this type of cursor is opened:
RsMaim.Open DataCmd, DataConn, adOpenForwardOnly

Creating any type of cursor is just as easy. Great, you now know how to create a forward-only
cursor. But when and why would you use one?

The forward-only cursor is often referred to as the firehose cursor. If all you want to do is dis-
play some data for read-only access, this is the cursor for you. For example, imagine you have
an Internet site that displays the closest regional office to its location based on the ZIP code the
Web surfer enters. This information is just for display. The surfer is not going to edit it in any
way. The fastest and easiest cursor to use in this situation would be the forward-only cursor.

Static Cursor The static cursor is created in almost the exact same manner as the forward-
only cursor. The only difference is, instead of the adOpenForwardOnly, you supply
adOpenStatic. Besides the slight difference in creation, a static cursor is much like a
forward-only in use.

29.1389-6 CH23 488 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Recordset Properties 489

The major difference is that a static cursor can move both forward and backward in the
recordset. When is this of value? Suppose you want to create a report on the sales the midwest
section of the business did in the month of August. This report might include totals, along with
breakdowns by product, salesman, and week.

Instead of using a forward-only cursor and having to run from the beginning to the end of the
recordset while gathering each type of data, the static cursor enables you to move forward and
backward to look for the correct data. At the same time, you will not incur the additional over- Part
head of a read-and-write cursor type such as KeySet. V
KeySet Cursor The KeySet cursor is a read-write cursor type that can move throughout a Ch
recordset. Its largest limitation is its inability to see the additional records that are created. You
still see changes and deletions, but additional records are a no-no.
23
The KeySet cursor is created in the same manner as the previous two cursors, with the value
adOpenKeySet forthe cursorType argument.

You want to use the KeySet cursor for applications where users will be modifying and deleting
data, such as a customer service phone agent application. The typical scenario in this type of
application is that the customer has called and wants to change his address. The phone agent
would then bring up his record and enter the change.

There would be no reason to see additional records (customers) created during this session,
because all you care about in this model is the data on one particular customer. The adding of
others customers or new information on other customers by other users will not affect how you
program this type of application.

Dynamic Cursor The dynamic cursor is the heavyweight of cursor types. It is the most pow-
erful. But this power comes at the cost of slower performance. Basically, use the dynamic
cursor when no other cursor type will work.

The value to send to the open command for this cursor is adOpenDynamic.

Using the dynamic cursor, you can move both forward and backward in the recordset. You can
also see the deletions and additions made by other recordsets.

For instance, you may have a recordset use this cursor when moving through a recordset
based on all the employees in a table with the last name “Smith.” If another recordset adds a
new Smith before you are done, you will be able to see this new record and display it also.

Recordset Properties
Now that you have created your Recordset object and set its cursor, you are ready to examine
some of the other properties of the Recordset object that you will use in your development.

29.1389-6 CH23 489 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 490 Chapter 23 Working with ADO’s Recordset Object

BOF/EOF
When running through a recordset looking for data, you need a way to check for the end of the
recordset. Slightly less popular, but still very useful, is the capability to check for the beginning
of the recordset. Both of these conditions can be tested for with the BOF and EOF properties.

BOF stands for Beginning Of File. EOF stands for End Of File. This property is used to find
whether the current record position is before the first record or after the last record. If the
property returns a value of True, then it is before or after the recordset. The following code
shows a test for EOF.
If rs_main.EOF then
We are at end of the recordset
else
do something
end if

What is meant by the current record position? The following figures show a recordset and what
is meant by the position of the current record. In Figure 23.1, the pointer is before the first
record; in Figure 23.2, the pointer is after the last record. In other words, it is impossible to
continue forward in the recordset.

FIG. 23.1 Current

Record
At the beginning of the
recordset object... Record 1

Record 2

Record 3

Record 4

Record 5

Record 6

Record 7

Record 8

•••

Record N

29.1389-6 CH23 490 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Recordset Properties 491

FIG. 23.2
...and at the end. Record 1

Record 2

Record 3
Part

Record 4
V
Ch
Record 5 23
Record 6

Record 7

Record 8

•••

Record N

Current
Record

RecordCount
This property returns the number of records in the recordset. There are many uses for this
property, but one of the best is for looping through the recordset without having to check for
EOF. The following code loops through a recordset.

IntNumRecs = rs_main.RecordCount
Do While I < intNumRecs
...something interesting
Loop

CAUTION
Be cautious about using the RecordCount property. If your recordset contains 500,000 records, imagine
how long it will take for the server to return the number of records if you also have 500,000 surfers!
A better way to set up your loops would be to check for EOF or BOF as you move through the recordset.

29.1389-6 CH23 491 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 492 Chapter 23 Working with ADO’s Recordset Object

AbsolutePosition
The AbsolutePosition property enables the developer to designate exactly which record to
set the current pointer to. If the current record is the first in the recordset, it is equal to 1. The
following line of code sets the absolute property to the fifth record.
Rs_main.AbsolutePosition = 5

Figure 23.3 shows where the current record pointer would now reside.

FIG. 23.3
Setting the current Record 1
record with
AbsolutePosition.
Record 2

Record 3

Record 4

Current
Record Record 5

Record 6

Record 7

Record 8

•••

Record N

TIP Some of the records in a recordset are cached in memory. (The size of this cache is set with the
CacheSize property.) When you set the AbsolutePosition property, all the records in the cache
are reloaded, even if the record you are setting up to point to was in the cache.

AbsolutePage
There is a way to group records in a recordset that is called pages. By setting the PageSize
property, you can set (for instance) four records per page. You can then use the AbsolutePage
property in much the same manner as AbsolutePosition. Figure 23.4 shows this grouping.

29.1389-6 CH23 492 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Recordset Properties 493

FIG. 23.4
Pages simplify data Record 1
management.
Record 2
Page 1

Record 3
Part

Record 4
V
Ch

Record 5 23
Record 6
Page 2

Record 7

Record 8

•••
Page N

Record N

In the following code, I set the page size to 4 and the AbsolutePage property to 3.
Rs_main.PageSize = 4
Rs_main.AbsolutePage = 3

Figure 23.5 shows the results of setting these properties.

Filter
Perhaps the most interesting of all the recordset properties is Filter. This property sets the
current record pointer to the first record that satisfies its conditions. For example, if you want
to filter the recordset and look just at the records where the last name is Smith, the code would
look like the following (assume the field name for last name is Last_name).
Rs_main.Filter “Last_Name = ‘smith’”

The current record pointer would now be set at the first record that satisfies this condition,
namely, that has the last name field matching “smith.” Moving through the recordset would
now take you to the next record that satisfies the filter.

What if you want to turn off the filter and run the recordset in the normal manner again? The
following code does just that.
Rs_main.filter = adFilterNone

29.1389-6 CH23 493 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 494 Chapter 23 Working with ADO’s Recordset Object

FIG. 23.5
Setting the current Record 1
record pointer to a new
position.
Record 2
Page 1

Record 3

Record 4
Current
Record
Record 5

Record 6
Page 2

Record 7

Record 8

•••

Record N

Besides enabling navigation based on criteria for fields, Filter also has some interesting pre-
sets, as summarized in Table 23.3.

Table 23.3 Filter Settings

Name Description
AdFilterNone Removes the filter.
AdFilterPendingRecords This enables a view of the records that have been
changed but not sent to the server.
AdFilterAffectedRecords This shows the records affected by the last ReSync,
Delete, UpdateBatch, or CancelBatch.

AdFilterFetchedRecords This shows you the records in the cache.

TIP While Filter is a great tool, be wary. If you know that your recordset only needs people with the last
name Smith to begin with, don’t pull everyone into the recordset and then filter. Moving that much
data, especially on a well-surfed Internet site, will put a huge load on the server.

29.1389-6 CH23 494 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Navigating the Recordset 495

Instead, try to narrow down your recordset with a SELECT statement. For instance, SELECT * FROM
tbl_people WHERE last_name = “smith”.

Navigating the Recordset

Now that you have studied some of the properties you can set with the Recordset object, it’s Part
time to learn how to move through the recordset. You have already seen how to move to cer-
tain recordsets with the AbsolutePosition and AbsolutePage properties, but you will generally V
want to move in more conventional manners. Figure 23.6 shows a recordset with five records Ch
that you will navigate through as I talk about MoveFirst, MoveLast, MoveNext, and
MovePrevious.
23

FIG. 23.6
This recordset contains John Smith
five records.
James Jones

Tim Edwards

Larry Carlson

Joe Schmuck

MoveFirst
This method moves the current record pointer to the first record. In this position, BOF would
return a value of True. When you first create the recordset, you won’t need to use this method,
because the current record pointer will already be set to the first record.

However, there may be times when you will want to reuse code, so it is good practice to begin
using this method before you start moving though the recordset.

The following line of code moves to the first record in the recordset:
Rs_main.MoveFirst

Figure 23.7 shows the result of this line of code, with the record pointer set at the first record.

MoveLast
This method acts much like MoveFirst. MoveLast, however, moves beyond the last record. So,
EOF would return True after running this method.

29.1389-6 CH23 495 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 496 Chapter 23 Working with ADO’s Recordset Object

FIG. 23.7
Current “John Smith”
At the first record, ready
Record
to go. Pointer
“Larry Jones”

“Steve Smith”

“Jill Hill”

“Carl Bae”

TIP You won’t really use MoveLast much except to append records onto the end of the recordset. So it is
a good idea to make sure that you have a cursor type (like KeySet or dynamic) that supports changing
the recordset.

The code below sets the record pointer to EOF.

Rs_main.MoveLast

Figure 23.8 shows the results of this line of code, with the record pointer now set beyond the
fifth (and last) record.

FIG. 23.8
At the end of the John Smith
recordset.
James Jones

Tim Edwards

Larry Carlson

Joe Schmuck

Current
Record

MoveNext
MoveNext is probably the most executed method in ADO. It is also one of the easiest to use.
When you run this method, all you are doing is moving the record pointer one record forward.
The following line of code does just that:
Rs_main.MoveNext

29.1389-6 CH23 496 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Navigating the Recordset 497

Assuming that the record pointer was BOF before you ran this method, the record pointer
would now look like Figure 23.9.

FIG. 23.9
At the second record. John Smith
Current
Record
James Jones Part
V
Tim Edwards Ch

23
Larry Carlson

Joe Schmuck

CAUTION
Always be sure to check for end of file when using MoveNext. If you attempt to move forward when at EOF,
an error will be generated.

MovePrevious
MovePrevious works just like MoveNext, only backward. It moves the current record pointer
back one record. So, if you just moved to the second record as in seen in Figure 23.9, this line
of code
Rs_main.MovePrevious

would result in the recordset looking like Figure 23.10.

FIG. 23.10 Current

Record
At the begining of the
John Smith
recordset, again.

James Jones

Tim Edwards

Larry Carlson

Joe Schmuck

29.1389-6 CH23 497 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 498 Chapter 23 Working with ADO’s Recordset Object

CAUTION
The same caution about checking for EOF before executing MoveNext also applies to MovePrevious.
Only, now, you should check for BOF. In the example above, running MovePrevious again would result in an
error that you have tried to beyond BOF, so always check for BOF first.

Manipulating the Recordset

Moving through a recordset can be a lot of fun. If your application only needs to see informa-
tion, it is also all you need to do. But often you will want to edit or manipulate the information in
your recordset. For those times, ADO includes some great methods for doing just that. The
methods enable the developer to easily add, delete, and change values in the database.

Basically, there will be three actions you will want to perform on a recordset:

■ Update an existing record.

■ Add a new record.
■ Delete an existing record.

Update
This method is used to update an existing record in the recordset. Using this method is as easy
as ABC. The code below updates the phone field to a new phone number.
NewPhone = “555-1212”
Rs_main.Update phone, newPhone

In Figure 23.11 you can see an example of a recordset before the update.

FIG. 23.11
Before an update. John Smith
555-1111

Larry Edwards
555-0000
•
•
•

Joe Schmuck
555-1211

In Figure 23.12, you see the recordset after we have updated the values.

The parameters that Update can take can be a single field in the database or several fields.
Generally, you will set several fields to their new values all at once, and then call the Update
method.

29.1389-6 CH23 498 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 Manipulating the Recordset 499

FIG. 23.12
After an update. John Smith
555-1212

Larry Edwards
555-0000
•
•
• Part

Joe Schmuck
V
555-1477 Ch

23
AddNew
AddNewis used to add a new record to the recordset. The following code adds a new record for
Jim Smith, 555-1121.
Rs_main.AddNew
Rs_main(“firstName”) = “Jim”
Rs_main(“lastName”) = “Smith”
Rs_Main(“phone”) = “555-1121”
Rs_main.update

After you create a new record with AddNew, you need to call Update to post the information to
the database. Until Update is called, the current record will be the one you added. Additionally,
the record won’t exist in the database until you call Update.

Figure 23.13 shows the recordset from Figure 23.12 after I have added a new employee.

FIG. 23.13
Adding a new employee John Smith
555-1212
with AddNew.
Larry Edwards
555-0000
•
•
•

Joe Schmuck
555-1222

Jim Smith
555-1121

TIP Many people (including those at Microsoft) will say that you should use the Supports method first to
make sure your recordset is writeable and that you can, in fact, add a new record. In reality, a good
programmer should know what type of recordset he has created. I would consider calling Supports
pretty redundant in most situations.

29.1389-6 CH23 499 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 500 Chapter 23 Working with ADO’s Recordset Object

Delete
Delete removes the current record from the recordset. Of course, the recordset must support
deletions (in other words, it can’t be read-only) or an error will be generated. Delete can also
delete all records that satisfy the filter with an optional tag. The following code shows both
examples.
Rs_main.Delete ‘This deletes the current record
Rs_main.Delete adAffectGroup ‘ this deletes all records in the recordset that
satisfy the filter

Supports
The Supports method is used to determine whether the recordset supports a particular action
or not. You call it and send it a CursorOption, which represents the activity you want to try. It
then will return a Boolean value indicating whether it will work. The following code shows an
example of the Supports method in action:
If rs_Main.Supports(adAddNew) then
‘we can add a new record!
Rs_main.AddNew
Rs_main(“firstName”) = “Jim”
Rs_main(“lastName”) = “Smith”
Rs_Main(“phone”) = “555-1212”
Rs_main.update
End if

The Recordset in Action

Now that you have examined the properties and methods of the recordset, let’s look at the
employee drill-down application you have been developing in the last two chapters and try to
enhance it with your new found knowledge and skill.

First, let’s look at the form you will use to gather the last name to search on (see Listing 23.1).

Listing 23.1 Entering the last name to search on.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<META NAME=”GENERATOR” Content=”Microsoft Visual InterDev 1.0">
<META HTTP-EQUIV=”Content-Type” content=”text/html; charset=iso-8859-1">
<TITLE>Employee Search</TITLE>
</HEAD>
<BODY>

<H1> Employee Search </H1><BR>

* Required
<FORM NAME=”searchForm” Action=”results.asp”>

<TABLE>
<TR>

29.1389-6 CH23 500 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 The Recordset in Action 501

<TD>
Last Name:
</TD>
<TD>
<INPUT TYPE=”text” NAME=”lastName”>*
</TD>
</TR>
<TR>
<TD> Part
First Name:
</TD> V
<TD>
<INPUT TYPE=”text” NAME=”firstName”> Ch

</TR>
</TD>
23
<TR>
<TD>
<INPUT TYPE=”submit”>
</TABLE>

</FORM>

</BODY>
</HTML>

The form in Listing 23.1 asks for the last name and, optionally, the first name of the employee
to search on. You could add a simple VBScript or JavaScript on the page to check for this field
before submiting, if you want.

Listing 23.2 checks the variable submitted by the form and opens a recordset of those records
that satisfy the condition.

Listing 23.2 Displaying the first five records.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<TITLE>
Search For Employee - By Last Name
</TITLE>
</HEAD>
<BODY>

<!-- #include file=”adovbs.inc” -->

<%
Dim i
i=0

Set DataConn = Server.CreateObject(“ADODB.Connection”)

continues

29.1389-6 CH23 501 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 502 Chapter 23 Working with ADO’s Recordset Object

Listing 23.2 Continued

‘Set our connection and command timeouts higher, due
‘to all the heavy traffic we will getting
DataConn.ConnectionTimeout = 60
DataConn.CommandTimeout = 60

DataConn.Open “DSN=Employees;User=Keith;Password=HiElizabeth”
Set DataCmd = Server.CreateObject(“ADODB.Command”)
Set DataCmd.ActiveConnection = DataConn

‘setting the name and value of the parameter

StrName = “last_name”
StrValue = Request.Form(“last_name”)

‘setting the type, direction, and size

intType = adChar
intSize = 50
intDirection = adParamInput

‘creating the parameter object

set dataParam = dataCmd.CreateParameter(strName, intType, intDirection, intSize,
strValue)

‘appending the parameter to the collection

dataCmd.Parameters.Append dataParam
dataCmd.CommandText = “qryLastName”
Set rs_main = Server.CreateObject(“ADODB.Recordset”)
Set rs_main = DataCmd.execute
%>

<TABLE>
<TR>
<TD>First Name</TD>
<TD>Last Name</TD>
<TD>Phone</TD>
</TR>

<%
do while i < 5 AND not rs_main.EOF
Response.write(“<TR>”)
Response.write(“<TD>”)
Response.write(rs_main(“FirstName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“LastName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“Phone”))
Response.write(“</TD>”)
Response.write(“</TR>”)
Rs_main.moveNext
i = i +1
loop
%>
<TR>

29.1389-6 CH23 502 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 The Recordset in Action 503

<TD>
<form action=”next.asp”>
<input type=”submit” name=”next” value=”next”>
</form>
</TD>
</TR>
</TABLE>

<% set Session(“rs”) = rs_main %> Part

</BODY>
</HTML> V
Ch
As you can see, after you open up the connection, you create a SQL command to select the 23
records that satisfy the search criteria. Then, you open up a recordset based on that Command
object. Notice how I have optimized the code by hard coding what type of command (text, in
this case) this will be. Once you have created the recordset, you then loop through and display
the first five records. If the user wants to see the next five records, he presses the Next button,
which will take him to this same page, with the next five records displayed. This is possible
because you have set the Recordset object rs_main to the Session.

When the user presses the Next button, the next page is called, which pulls the recordset out
of the session, and then displays the next five. Listing 23.3 shows this page.

Listing 23.3 Displaying the next five records.

<%@ LANGUAGE=”VBSCRIPT” %>

<HTML>
<HEAD>
<META NAME=”GENERATOR” Content=”Microsoft Visual InterDev 1.0">
<META HTTP-EQUIV=”Content-Type” content=”text/html; charset=iso-8859-1">
<TITLE>Search Employees</TITLE>
</HEAD>
<BODY>

<% set rs_main = Session(“rs”)

dim i
i = 0
%>

<TABLE>
<TR>
<TD>First Name</TD>
<TD>Last Name</TD>
<TD>Phone</TD>
</TR>

<%
do while i < 5 AND not rs_main.eof
Response.write(“<TR>”)

continues

29.1389-6 CH23 503 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 504 Chapter 23 Working with ADO’s Recordset Object

Listing 23.3 Continued

Response.write(“<TD>”)
Response.write(rs_main(“FirstName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“LastName”))
Response.write(“</TD>”)
Response.write(“<TD>”)
Response.write(rs_main(“Phone”))
Response.write(“</TD>”)
Response.write(“</TR>”)
Rs_main.moveNext
i = i +1
loop
%>
<TR>
<TD>
<form action=”next.asp”>
<input type=”submit” name=”next” value=”next”>
</form>
</TD>
</TR>
</TABLE>

<% set Session(“rs”) = rs_main %>

</BODY>
</HTML>

</BODY>
</HTML>

From Here…
Now that you have seen the simple, yet powerful, ways in which you can use the Recordset
object, you have everything you need to create some really great Web-based applications.

You have seen how to navigate and manipulate your data with the Recordset methods. You
have also examined the issues revolving around creation of Recordset objects. Hopefully, you
are as excited as I was when I finally understood ADO and the Recordset object.

The following chapters will now probably be of interest to you:

■ Chapter 24, “Managing Your Database,” will delve deeper into database issues such as
ODBC 3.0, Transaction Server, and concurrency.
■ Chapter 25, “Accessing Databases with Remote Data Services,” will show you how to
create data-driven applications that use these ActiveX objects on the client.
■ Chapter 26, “Putting It All Together: Creating a Dynamic Database Application,” will
show you how to put this disparate information together. You will see how to create a
powerful and functional Web-based application.

29.1389-6 CH23 504 10/26/98, 9:43 AM

Brands/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH23 lp#3
 24
505

C H A P T E R

Understanding Databases
Managing Your Examine database basics.

Database Transaction Processing with

Microsoft Transaction Server
Part
See how to implement robust data-
driven applications with MS Trans- V
action Server.
Ch
Concurrency Issues 24
Examine the issues surrounding
Now that you have examined how to integrate your ASP concurrency and database applica-
applications with your data, it’s time to examine in more tions.
detail how to manage your database. Database manage-
ODBC 3.0
ment is not database administration. Database manage-
ment is the tools and techniques you employ to design and Find out what’s new with ODBC 3.0,
secure a robust and well-behaved database. and how it can help you develop
more robust and portable code.
When developing application access to a database, two
very large problems loom over developers. The first is Designing and Configuring
transactions. The second is concurrency. Databases
Learn guidelines for designing and
Transactions are the idea of grouping actions against the
configuring your Web databases.
database to enable faster and more robust execution.
Without transactions, you cannot guarantee the reliability
of your database.

By definition, a Web-based application is one that will have

multiple users. With multiple users, however, comes the
idea of concurrency. How do you protect the integrity of
your data when several (perhaps thousands of) users are
accessing it at once? In this chapter, you will look at some
of the issues surrounding concurrency, and how to best
protect your application depending on how you need to
use it.

30.1389-6 CH24 505 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 506 Chapter 24 Managing Your Database

This chapter will also discuss issues surrounding the design of your database. A well-designed
data model can mean a much more efficient and higher-performing application. In reality,
creating a robust data model can be quite simple, as long as you follow some simple
guidelines. ■

Understanding Databases
In the most simple terms, a database is the permanent storage of data. However, how you store
your information is anything but simple. Databases must be able to support thousands, if not
millions, of records. At the same time, there must be ways to quickly access and manipulate
this information.

Another issue that arises is the capability to quickly port your applications from one database
to another. This means that there needs to be a standard upon which databases are built.

Finally, modern business applications need to enable multiple users to access the same data-
base over a network. This type of application development is called client/server development.
You can save yourself many headaches in the future by designing applications that take into
consideration a client/server architecture from the start.

Types of Databases
Many different types of data storage solutions have been created over the years, including
network, hierarchical, and flat databases. Each of these has pretty much had its day, and due to
one imperfection or another, has been replaced with far more effective solutions. Today, most
serious database work is done with either relational database systems or with object-oriented
database systems.

Object-Oriented Databases Object-oriented databases (ODBMS) are fairly new to main-

stream database application development. With an object-based database, there is one-to-one
mapping of the objects found in the application (and the real world) with the objects stored in
the database. In relational databases, there is a “flattening” of real-world data to make it fit that
database structure. With ODBMS, a more natural structure of data is achieved.

Object-oriented databases implement an object model. A popular object model is the ODMG
object model, which was developed by the Object Database Management Group. This object
model provides for the implementation of such concepts as the following:

■ Attributes and properties of database objects. This means you can specify in more detail
what type of data the various objects hold.
■ Object methods and exceptions. You can create methods for your data objects to use.
■ Multiple inheritance for objects. You can inherit a Toyota data object from a general car
object.
■ Object lifetimes. This means you can control how long a data object lives in the database.
■ Concurrency control and object locking. You can have even greater control over the
locking and control of the data.

30.1389-6 CH24 506 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 Understanding Databases 507

Relational Databases Relational databases are the world standard for database architecture.
With a relational database, data is stored in sets of two-dimensional arrays called tables. Col-
umns in the table contain properties. Rows are the individual records in the database. Figure
24.1 shows an example table from such a database.

FIG. 24.1 Emp # first name last name phone

An example table. 101 Ted Smith 555-1212
102 John Johnson 555-1211
201 Sara Scott 555-1111
202 Michael Wilson 555-1110
300 Lara Cooper 555-1010

Where relational databases become interesting is the idea of having multiple tables that link to Part
each other. For example, one table might list all of the employees in a business. It might con-
tain columns for employee number, name, title, and department. Another separate table may
V
contain a daily summary of hours worked by each employee. It might contain columns for Ch
employee number, date, and hours worked. Figure 24.2 shows a graphical representation of 24
how this might look.

FIG. 24.2 first name last name phone Emp # Emp # Date Hours
Two linked tables. Ted Smith 555-1212 100 100 9-1-97 8
John Johnson 555-1211 101 100 9-2-97 7.5
Sara Scott 555-1111 200 101 9-1-97 8
Michael Wilson 555-1110 201 200 9-1-97 10
Lara Cooper 555-1010 300 • • •
• • •
Dan Edgar 555-0000 301 • • •
Alex Mundlin 555-1000 302 302 10-4-97 7

SQL
A major goal of application development is portability. This is the notion of creating an applica-
tion that can be easily ported to various other environments with its behavior remaining the
same. When applied to databases, portability means that your applications should behave the
same regardless of which database they hook up to.

If each database vendor created its own proprietary method of accessing data in the database,
then this goal of portability would be very difficult. This is where the Structured Query Lan-
guage (SQL) comes into play.

SQL is a language designed for accessing and manipulating data. There are many different
dialects of SQL, but most SQL implementations support the most common SQL statements.
The list below summarizes the major SQL keywords and their uses.

SELECT—This keyword is used to select data from the database.

UPDATE—This keyword is used to update existing records.

30.1389-6 CH24 507 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 508 Chapter 24 Managing Your Database

INSERT—This keyword is used to add new records to the database.

DELETE—This keyword is used to delete records from the database.

To quickly illustrate each of these keywords, I will create a small table containing a list of em-
ployees in my building. This table contains a bare set of information: employee ID, name, and
phone number. The Figure 24.3 shows this table graphically.

FIG. 24.3
A table of employees.

Suppose you want to look at, and possibly manipulate, the records in this table where the em-
ployees’ first names are “John.” The select statement to do this would be:
SELECT employeeID, firstName, lastName, phone FROM tblEmployees WHERE firstName =
➥“John”

The results of this statement can be seen in Figure 24.4.

FIG. 24.4
SELECTing all records
with the first name of
John.

Now imagine that you want to delete every record that begins with John. That SQL statement
would look like the following:
DELETE * FROM tblEmployees where firstName = “John”

After executing this statement, the table looks like Figure 24.5.

FIG. 24.5
DELETEing all the
Johns.

30.1389-6 CH24 508 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 Understanding Databases 509

Now suppose you want to update a person’s record. Imagine that your employee Susie Harris
was married and her last name is now “Smith.” The SQL statement to update this record would
be the following:
UPDATE tblEmployees SET lastName = “Smith” WHERE firstName = ”Susie” and lastName
= “Harris”

The results of this operation would look like Figure 24.6.

FIG. 24.6
Susie, with her new last
name.

Part

Finally, suppose that a new employee is hired to fill the shoes of all the Johns that were deleted.
V
The INSERT statement to handle this is Ch

INSERT INTO tblEmployees (firstName, lastName, phone) VALUES (“Elizabeth”, 24

➥“Ballinger”, “555-8765”)

This statement results in the table shown in Figure 24.7.

FIG. 24.7
Adding a new
employee.

Client/Server Development
When developing a database-driven application, it will be a very rare case where many users do
not access the database at once over a network or the Internet. This type of application is called
client/server, and it refers to a system in which there is a client program that handles some
tasks, and a server program that handles others. The Web is an example of a client/server
application. The Web server handles certain tasks, such as finding the appropriate page, and
the browser displays the page contents. With client/server database programming, there is, at
the very least, a server where the database resides, and several client machines on which a
client program is used to access the database server. Figure 24.8 shows this type of two-tier
setup.

When developing client/server systems, you must consider which tasks and processes are
performed by the server, and which ones the client should handle. The very best applications
are a marriage of tasks between these two components.

30.1389-6 CH24 509 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 510 Chapter 24 Managing Your Database

FIG. 24.8
A simple client/server
setup.

Server

Client machines

Some of the tasks best suited for the client include the following:

■ Display of result sets from the database

■ Simple data validation
■ Gathering data before querying the database
■ Formatting and filtering result sets from the database

Meanwhile, the server is often best used for these tasks:

■ Complex data validation

■ Indexing data
■ Finding and returning result sets to the client
■ Maintaining database integrity
■ Managing transactions

Transaction Processing with Transaction Server

Microsoft Transaction Server is an exciting product that can reduce application development
by up to 40 percent. It enables developers to concentrate on developing the business logic of
their applications, and not worry about object brokering, transaction processing, and other
issues such as connection management.

For more information on these subjects, please see Chapter 28, “Working with the Microsoft
Transaction Server.”

30.1389-6 CH24 510 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 Transaction Processing with Transaction Server 511

Transactions
Before I go into more detail on how Transaction Server can help ease development and in-
crease the security of your system, let’s examine transactions themselves in a little more detail.
You cannot write a database application without considering transactions.

Imagine the Bank of Mongolia has hired you to create an intranet-based teller application. One
of the main functions of this application will be to transfer money (right after deposit and with-
drawal of funds).

First, the teller enters the account to take the money out of your account. Then, he enters the
account to deposit the funds into (Account 2), along with the amount, and then submits/con-
firms the changes. Behind the scenes, the application is going to first withdraw the money
from Account 1. Then, after ensuring that it has done this successfully, it will need to deposit
the funds into Account 2. But what happens if there is a system failure between the first step Part
(taking the money out of Account 1) and the second (deposit into Account 2)? V
To prevent this from happening, developers have created what is called transaction processing, Ch
which enables you to specify certain tasks to be completed together. If any task fails, then they
all fail.
24
There is a more formal, specific definition of transactions called the ACID test. ACID is an
acronym that stands for Atomicity, Consistency, Isolation, and Durability. Figure 24.9 illustrates
this concept.

FIG. 24.9 Atomic

ACID. Consistent
Transaction
Isolated
Durable

Atomicity means that no matter how many objects or procedures are called, the transaction is
treated as a single unit. If any piece of the transaction fails, the transaction as a whole fails. If
everything succeeds, then the transaction as whole succeeds.

Consistency means that the transaction will work the same way every time. It also means that it
won’t change aspects of the system that should be invariant.

Isolation means that nothing external to the transaction should affect its execution. At least, no
other transaction should. Transactions are totally isolated from one another, and have no
knowledge of each other’s successes or failures.

Durability means that that the results of the transaction will be durable and lasting, even if the
system the database is on is shut off.

Inside Transaction Server

Transaction Server is a complex system made up of many different components. There are
four key elements that are especially important: the Transaction Server executable, resource
managers, resource dispensers, and the Distributed Transaction Coordinator.

30.1389-6 CH24 511 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 512 Chapter 24 Managing Your Database

The Transaction Server executable is a service that provides for the context and thread man-
agement of business objects that are registered with Transaction Server. It can be thought of
as the “brain” of Transaction Server.

The Microsoft Distributed Transaction Coordinator (DTC) actually precedes Transaction

Server (it first showed up in Microsoft SQL Server). DTC is used to manage transactions that
take place on several servers, as opposed to all transactions occuring on just one machine.

To explain further, think back to the bank example. Imagine that you have a separate database
on a different server that is used to log all transfers and the information about them, such as
the people involved, the time, the teller, and so on. With DTC, you can be assured that, if the
transaction fails to write to either the database that contains the funds information or the new
activity database, the transaction as a whole will fail.

Resource managers provide the data that is durable to the transaction. By using resource man-
agers, such as SQL Server, the transaction can be guaranteed to be isolated and atomic. There
are two phases to a transaction: the prepare phase and the commit. The prepare phase is where
the provider prepares for the transaction. The commit phase is where it actually executes the
transaction. The resource manager helps to ensure that if there is an error, it will roll back any
changes.

Resource dispensers are similar to resource managers. The basic difference is that with a dis-
penser, you don’t have the guarantee of durability. There are two major resource dispensers
right now: the ODBC resource dispenser and the shared property resource dispenser.

The ODBC resource dispenser manages ODBC connections to your various datasources. By
using the ODBC resource dispenser, your system can benefit from connection pooling that this
dispenser employs.

Concurrency Issues
Now return to the example from the last section, the bank teller application. But now, think
about happens when you have 100 bank tellers all accessing the database at once.
Suppose the bank teller transfers funds from your account into your mom’s. At the same time,
Mom is at her local branch trying to close her account. If the money is deposited before she
finishes closing the account, but after they have tallied up how much money she has left, your
poor mother will be out the several hundreds of dollars you were sending her.

To deal with these problems of concurrency, or multiple users accessing the same data, several
different techniques have been developed. Most of these involve locking the database, or a
section of it, during a transaction. Let’s look at some of these techniques, particularly locking.

When developing a multi-user database application, you should strive to develop an application
that behaves exactly the same as if it were a single-user application. To help to this end, data-
bases provide a feature called locking. Locking is the means to prevent all but the lock-issuing
process access to the data that has been locked.

30.1389-6 CH24 512 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 ODBC 3.0 513

There are two major types of locks: shared locks and exclusive locks (Figures 24.10 and 24.11
show these types of locks). A shared lock is one that can be shared by several processes at
once. For instance, if all you and your mother want to do is see how much money is in her
account, both of the processes that have this information would be able to share a lock. This
type of lock only works for read-only connections.

FIG. 24.10 transaction #1

Locks can be shared...
transaction #2

transaction #3 Data

Lock
Part
Processes that are going to update, insert, or add information to the database issue exclusive V
locks. In the case of the funds transfer, each process would create an exclusive lock for mom’s
Ch
account.

Besides the classification of locks into exclusive or shared, there is also the issue of just how
24
much data to lock. Most databases enable several levels of granularity in terms of what data
can be locked. These range from locking the entire database to locking a single row.

FIG. 24.11 transaction #1

...or exclusive.
transaction #2

transaction #3 Data

Lock

ODBC 3.0
The Open Database Connectivity 3.0 standard is the most popular way to connect to a database
in the Windows world. By using ODBC, developers need not concern themselves with the
particular database they are hooking up to. Instead, by using ODBC, you can deal with a level
of abstraction above the particular database.

Specifically, ODBC is an API for accessing, manipulating, and creating databases. It is based on
the X/Open Call-level Interface and uses SQL.

As a Web developer, you don’t need to know the exact API for ODBC. However, a good under-
standing of ODBC will do you a world of good.

When you access a database through ODBC, that database must be registered as an ODBC
datasource. By registering the database as a datasource, the application only needs to know its
datasource name. The location of the database makes no difference, nor even what type of data-
base it is.

30.1389-6 CH24 513 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 514 Chapter 24 Managing Your Database

There are three types of datasources you can create: System, User, and File. Figure 24.12
shows how to access any of these from the ODBC control panel.

FIG. 24.12
Datasources can be
system wide, user
specific, or located in
a file.

A system datasource is one that is available to any user of the system. (Of course, the database
itself could be password protected.) A user datasource is one that is only available to that user.
Finally, a file datasource is a description of the database. It can be used to hook up to the data-
base without having to register the database itself with the system.

Generally, with Web applications you want to create system datasources. Only rarely would you
use a file or user datasource. But if you do find the need, they are available to you.

Designing and Configuring Databases

Relational databases enable very powerful functionality. For instance, Oracle databases can
store millions of records. Even with tools such as Access, anyone can develop a database to
hold his or her information. Using ASP with ADO, you can publish that information to the Web
quickly and easily. However, it is very easy to start running into problems with your database.

There are two aspects of database design that I will look at here. The first is data modeling; the
second is data access. If done wrong, each can have its own difficulties, ultimately rendering a
program useless. If done right, your application will run more efficiently and more quickly.

Data Modeling
The goal of a good data model is to avoid duplicate data. Also high on the list is creating a
model that is understandable and easy to develop for. Another very important piece of the data
modeling puzzle is a database structure that won’t slow down the application.

Another name for a database that doesn’t repeat data is a normalized database. The opposite of
a normalized database is one that is flat, or non-relational. The best way to learn how to design
is to see a design in action.

30.1389-6 CH24 514 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 Designing and Configuring Databases 515

Imagine that you are building an application to store the client companies for a fictional com-
pany, along with any key contacts at the company. Now, you don’t really know how many con-
tacts you will record for each client; some might have one, while others have 10. You could
guess, and build a database structure that will hold up to 10 contacts per client. In that case, the
database would look something like that shown in Figure 24.13.

FIG. 24.13
A flat database
structure.

Part

This type of database structure is a compromise. It is not flexible, and it requires writing code V
to figure out how many contacts for each salesman have been written to see what slots are still Ch
available. Additionally, every client record that doesn’t have 10 contacts would be wasting
space. This is a classic example of a flat database.
24
A relational design, on the other hand, would enable you to have as many contacts, or as few, as
the business client has. That would not only mirror real life more, it would also be easier to code.

To implement this, you need to create one more table. This table will be called tblContact, and
it will link each record in the contact table with a record in tblClient. You link tables together
through the use of keys. A key is a unique identifier for each record in a table. Figure 24.14
shows the new and improved database structure.

FIG. 24.14
Relationships can exist
between tables in a
database.

Notice in Figure 24.14 that there is a line drawn between clientID in the table tblClient, and
the clientID field in tblContact. This line has a “1” on one end, and the sign for infinity on the
other. This denotes a one-to-many relationship. In other words, for every one record in
tblClient, there are (or can be) many records in tblContact. This more closely mirrors what
you have in real life. For each one client your company has, there can be many contacts.

Another situation you will run up against is the possibility of adding salesmen to the database.
You will probably want to link your salesmen to the accounts they are in charge of. However,
you may have several salesmen attached to a single client. And, most salesmen will have sev-
eral clients.

30.1389-6 CH24 515 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 516 Chapter 24 Managing Your Database

To handle this, add two new tables. The first will contain information about the salesmen. The
second will be a link between salesmen and clients. Figure 24.15 shows this structure.

FIG. 24.15
Salesmen and their
clients.

This design enables you to have as many salesmen per client as you need, and salesmen can
also have as many clients as you want to assign them.

This idea of normalized databases can be measured through what are called levels of normal-
ization. Generally, third-level normalization is the goal to shoot for. It provides the best compro-
mise between data redundancy and development ease.

The following is a list of the various levels of normalization you should care about.

■ Not at all
■ First-level
■ Second-level
■ Third-level

A database that is not normalized at all is shown in Figure 24.16.

FIG. 24.16
A non-normalized
database.

This database contains a list of employees, but all the employees for a certain department are
found in a single column, which is a very awkward way to store this information. It would bet-
ter to give each employee his own record, and record their location in another column for that
row. Figure 24.17 shows a revised version of this table. This is the first level of normalization.

FIG. 24.17
First-level normaliza-
tion.

30.1389-6 CH24 516 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 Designing and Configuring Databases 517

To achieve second-level normalization, you need to give each record a unique identifier. This
can be done by adding a new column for employee number. Figure 24.18 shows this added
change.

FIG. 24.18
Second-level
normalization.

To achieve full third-level normalization, you need to eliminate the data redundancy of naming
the department that each employee is in by keeping departments in their own table with their
own unique record identifier. You can then refer to this identifier in the employee table. If Part
something changes within your department, such as location, you only need to update the V
record in the department table. Making this change gives us third-level normalization, as
Ch
shown in Figure 24.19.
24
FIG. 24.19
Third-level
normalization.

Keys and Indexes

There are two different unique identifiers in the last example: one for the employee table (em-
ployee number) and one for the department table (department number). In each of these
tables, this column is called the key. In the employee table, you also have the department num-
ber. Because this a key in another table, it is called a foreign key.

By using keys, you can group data together in a logical manner. Departments have their own
table. Employees have their own table. You can add other parts of the business, such as clients,
and give that set of data its own table. By linking these keys together, you create relationships.

Using SQL statements across multiple tables, you can quickly pull a result set of data that
groups this information together for the user. For instance, the following SQL statement:
SELECT tblEmp.employeeName, tblDept.location WHERE tblEmp.deptNum =
➥tblDept.deptNum

creates a result set that looks like the one shown in the Figure 24.20.

Indexes are used in some databases to speed up access to data. Instead of having to scan every
record in a table, if the table has an index, that index can be used to quickly narrow down which
records to select. With SQL Server, you can have multiple columns be used as indexes. For in-
stance, you can index both zip code and last name of a table that includes customer information.

30.1389-6 CH24 517 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 518 Chapter 24 Managing Your Database

FIG. 24.20
A result set from two
different tables.

Indexes can be quite helpful, and can improve performance dramatically. However, they take
up additional space in the database. So before implementing indexes, it is always a good idea to
understand the use of the table involved. Is the customer database searched often? If so, what
are the usual search criteria? Do users search by zip code? If they do, create an index based on
the zip code.

But what if last name or phone is used to search far more often? In that case, creating a zip
code index would probably be a waste of time and space. A better choice would be an index
based on those criteria.

Data Access
The other major issue to consider besides how you design your data is to think about how you
access it. This discussion will deal with SQL statements. Basically, there are two issues to con-
sider when designing your SQL statements.

The first issue deals with the amount of information that you move. When dealing with lots of
hits, you need to be aware of the huge amount of data that might be shuffled around by your
processes. For instance, in the database that you designed in the last section, you might want
to pull out a list of all customers. All this statement needs to do is display a list of names. Here
is one way to grab that list:
SELECT *
FROM tblClients

This statement will return all the fields in the table tblClients (see Figure 24.21).

FIG. 24.21
Selecting too much
information.

As you can see, for as simple a request as client name, you are grabbing way too much informa-
tion. For an application that isn’t used a lot, this isn’t a big problem. But imagine the implica-
tions of this SQL statement being hit thousands of times an hour, and you can see the problem.

It is good coding practice to always specify the exact fields you want in a SQL statement. This
will avoid the problem of grabbing too much information. The following statement does just
that:
SELECT clientName
FROM tblClients;

This statement results in Figure 24.22, which is much more economic.

30.1389-6 CH24 518 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 From Here… 519

FIG. 24.22
Only pull as much
information as you
need.

Database Security
One of the major facets of developing Web-based databases that you must consider is data
security. A secure database is a must. If just anyone is allowed to access, delete, and manipulate
your data, it can become useless. Many databases enable you to set up permissions for various
levels of access to the database.

In particular, Microsoft’s SQL Server gives the database administrator and developer a high
level of control over who can access what information. If you have seen connection string argu-
Part
ments that specify usernames and passwords, it is probably because the database being ac-
cessed is a database that requires user logins, such as Microsoft SQL Server or Sybase SQL V
Server. Ch
To configure a new login and its access privileges, follow these steps: 24
1. Start SQL Server Enterprise Manager.
2. Select the server you want to configure.
3. Type in the login and password of the user.
4. Choose which database operations this user is allowed to perform.
5. Click the Add button.

By carefully considering which permissions to give each type of user, you can secure your
database from allowing just any user to manipulate it. In an ASP application distributed over the
Internet, this may be anyone on the earth with a browser and Internet connection. You don’t
want unauthorized users typing in your database, or possibly even deleting sensitive and impor-
tant information from it.

From Here…
In this chapter, you examined some of the basics of databases. You learned the differences
between relational and object-oriented databases, and you saw how SQL can help you develop
portable, database-driven applications.

Further, you have seen how Microsoft’s Transaction Server can help you implement transac-
tion processing in your applications. With this powerful new application, you can have your
transactions span multiple servers, which enables distributed processing.

You have also seen how concurrency, or multiple users accessing your application simulta-
neously, can lead to the loss of data integrity. You have seen how locking can help alleviate this
situation, and what amount of locking to apply, based on the situation.

30.1389-6 CH24 519 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 520 Chapter 24 Managing Your Database

You have further examined ODBC 3.0. This standard of database access enables developers to
forget about the exact nature of the database they are accessing, and instead spend their devel-
opment time on the business logic of the application.

Finally, you have seen some guidelines for data modeling and data access that will help you to
develop your applications faster, and to develop faster applications. For related information,
please see the following chapters:

■ Chapter 26, “Putting It All Together: Creating a Dynamic Database Application,” will
show you how to create a sample application with ASP and ADO.
■ Chapter 28, “Working with the Microsoft Transaction Server,” will give you more in-
depth information on this powerful new product from Microsoft.

30.1389-6 CH24 520 10/26/98, 9:45 AM

Brands03/Stone3 SEU Active Pages Server #1389-6 SRing 11/13/97 CH24 lp#3
 25
521

C H A P T E R

Remote Data Services

Accessing Databases Overview
An overview of how the Remote

with Remote Data Data Services can help you create

powerful applications more quickly.

Services Using the RDS Components

Examine the components that make
up Remote Data Services.

Limitations of the Remote

Data Services Part
When Remote Data Services are not V
enough.
Formerly named the Advanced Data Connector, Remote Ch
Extending the Phone Book
Data Services enable you, the developer, to create Web 25
pages that are bound to a data source on the Web server. See how the simple phone book
This exciting technology is integrated with IE 4, but can application from earlier chapters
also be used with IE 3. can be written for RDS.

Using this technology, you can bind data to your Web

pages and then cache it on the client browser, which re-
leases the browser from having to make unnecessary trips
to the server. This means that you can create full-blown,
database-driven programs that run from within the
browser. You can do this because Remote Data Services
utilize existing technologies that you are already familiar
with: ActiveX Data Objects, OLE DB, VBScript, and
more. ■

31.1389-6 CH25 521 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 522 Chapter 25 Accessing Databases with Remote Data Services

Understanding Remote Data Services

Using the Remote Data Services, developers can now create powerful applications using the
familiar three-tier approach. By embedding components like drop-down and list boxes, you can
turn a Web page into a full-featured application. Figure 25.1 shows a graphical example of tradi-
tional three-tier architecture.

FIG. 25.1
Traditional three-tier Some
architecture. Client Protocol Middle Back-end
(VB, C++) Tier Server

Figure 25.2 shows three-tier architecture with RDS. Notice that the browser technology creates
a stateless and connectionless environment for the application, which is substantially different
from traditional client/server architecture.

FIG. 25.2
Client Middle tier
Three-tier architecture HTTP
(Web page of RDS
with RDS. with RDS Business Database
Components) Objects

With a Web page, updates are easy. Change the page, and next time your users visit, it will be
different. At the same time, the downside of Web pages—their static nature—is avoided with
client-side data caching. This means that the browser won’t have to make trip after trip to the
server for more data. It can be sent once, manipulated and navigated, and then finally, any
changes can be sent back to the server at the end of the transaction.

Of course, this isn’t always the best way to handle data. There will be times when you will need
to query the server for data, and then requery often before your user is done with his transac-
tion. In particular, a very active application that is used by multiple people will probably need to
check for data changes more often than a single user or less-used application.

There are three key objects that are used in most RDS applications:

■ AdvancedDataControl
■ AdvancedDataSpace
■ AdvancedDataFactory

AdvancedDataControl is a non-visual ActiveX control that is used to bind data to other visual
controls on a Web page. It is scriptable to enable navigation and data manipulation with an
already familiar language such as VBScript.

31.1389-6 CH25 522 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Understanding Remote Data Services 523

N O T E What is a non-visual ActiveX control? As you know, an ActiveX control is a control that you
embed in an HTML page that delivers more functionality than what HTML alone can deliver.
Most ActiveX controls are visual elements like enhanced tree views of Web sites and the like.

However, not all ActiveX controls have to be visual. For instance, there is a control that can be used as
a timer on a Web page. This control fires off an event at some specified time, as set up by the page
author. This control doesn’t actually appear on the page—it is non-visual. The
AdvancedDataControl is the same way. ■

The AdvancedDataSpace object is a constructor used to create DCOM marshaling proxies,

HTTP proxies, and Automation objects.

The AdvancedDataFactory object is the default middle tier in RDS applications. It is used to
send SQL queries and updates to the back-end ODBC-compliant datasource.

Creating Business Objects

Business objects are components that encapsulate application logic and business rules. By
componentizing the logic and rules of the application, you can more quickly and efficiently
develop and deliver Web-based applications. Part
For example, you may have an application that is used in online ordering. A business object in V
that application might be a set of code that is used to validate credit cards and make sure they
Ch
have enough money in available credit. This set of code is a business object.

Business objects can be created with any tool that can create ActiveX objects. Generally, the
25
business components of RDS will be created with either Visual C++ or Visual Basic. By using
this model, many benefits can be realized. The biggest of these (in my opinion) is that business
objects are, for the most part, independent of the interface and database. The logic of these
components can be deployed in a variety of situations, enabling for faster development of future
versions of the application.

In creating an RDS application, the following principles apply well:

■ Use the client for user interface and most data validation. This means that the focus
during the development of the client should be on a usable interface, not on the internal
logic of the program. (By simple data validation, I mean such functions as checking for
illegal characters [such as $, %, #, and so on] in a name field, or for only numbers in an
area code.)
■ Encapsulate business logic into components using a black-box approach, where you can
later change the component and the program still works. This means true encapsulation
and data hiding. The server-side components should be independent of the front-end.
Changing either the server-side objects or the client browser shouldn’t impact the
program.

31.1389-6 CH25 523 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 524 Chapter 25 Accessing Databases with Remote Data Services

Understanding IE 4 and Dynamic HTML

Internet Explorer 4.0 includes a new feature called Dynamic HTML. With Dynamic HTML,
Web pages can respond and interact with users in complex manners previously only seen in
stand-alone applications. There is even a version of the classic arcade game Asteroids written
completely in JavaScript and Dynamic HTML (DHTML).

What exactly is meant by Dynamic HTML, and what does it have to do with Remote Data Ser-
vices? With IE 4, Microsoft has made every element of a Web page scriptable, which means
that using JavaScript or VBScript, a page author can manipulate the letters and words on a
page—changing them dynamically.

Using a combination of Remote Data Services and DHTML, you can create a Web page that
displays, for instance, a list of employees in your department. Next, you can filter and sort the
list by clicking on the column you want to sort by. This is possible through two new tags that
Microsoft has introduced with DHTML: DATASRC and DATAFLD.

Active Server Pages and RDS

ASP and RDS go together for two reasons. The first is that, as an ASP developer, you are al-
ready familiar with ADO, which is used on the client for navigation and manipulation through
the recordset bound to the AdvancedDataControl.

The second is the capability to dynamically create the appropriate page based on information
the user has previously entered. For instance, the user may enter a login password that enables
him both read and write access to the database, while others may only be allowed read access.
This can be achieved by checking the password on the server with ASP, and then writing a
page based on the permissions of that login. Later in this chapter, you will see an example of
this.

Using the RDS Components

You have already briefly looked at the major objects that are used with the Remote Data Ser-
vice in the previous section, “Understanding Remote Data Services.” Now, let’s look more in
detail at these and the other components that you will utilize in your RDS development.

Advanced Data Virtual Table Manager

The Virtual Table Manager (VTM) takes care of caching data that is sent to the client, and also
manages the updates to that data. The VTM is a set of data structures that are used to buffer
information on table structure along with that actual information in the table.

To be able to manage the caching and updating of data that the VTM does, it has the capability
to create and delete temporary tables. It also can provide schema information such as base
tables and key columns.

31.1389-6 CH25 524 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Using the RDS Components 525

One of the best features of the Virtual Table Manager is its capability to send only the records
that have been updated. This means that unnecessary traffic to and from the browser can be
greatly reduced when using RDS.

How does the VTM marshal (transfer) the data it sends? It uses tablegrams. Tablegrams are a
special MIME type that was designed for transporting table-based information over HTTP. One
of the nicest features of tablegrams is their capability to normalize the information they send.
This means that if there are certain rows of information that are repeated throughout a table,
they will send the actual information only once; thereby saving bandwidth once again.

AdvancedDataControl Object
The AdvancedDataControl object is a non-visible ActiveX control that binds data from a
recordset (which is, itself, the result of a query such as a Select statement) to various text
fields and other visible ActiveX controls on the Web page.

How do you use the AdvanceDataControl in a Web page? Basically, you embed it in the same
way as any other ActiveX control, and give it a Width and Height of 1 to avoid losing any space
to it. The code below shows an example of how to do this.
<OBJECT CLASSID=”clsid: 9381D8F2-0288-11d0-9501-00AA00B911A5"
ID=”ADC1"> Part
<PARAM NAME=”Bindings” VALUE=”SGrid;”>
<PARAM NAME=”SQL” VALUE=”SELECT * FROM tbl_phone”>
V
<PARAM NAME=”Connect” VALUE=”DSN=EmployeeDB;”> Ch
<PARAM NAME=”Server” VALUE=”http://hr-www; “>
</OBJECT> 25
TIP When using the Sheridan grid control with RDS, make sure to create the AdvancedDataControl last
on the page. There is a known bug that will cause your application to act weird otherwise.

The code above should look very familiar. It is the standard method in which ActiveX controls
are embedded into a Web page. There are four major parameters that you will set when using
this control (see Table 25.1).

Table 25.1 Parameters of the AdvancedDataControl

Parameter Description
Binding Specifies which controls to display the recordset with on the page. Can
range from a single data grid that handles all of the columns of data to
several controls that each handle a column of data.
SQL This specifies the SQL statement to use in the creation of the recordset.
Connect This property specifies the DataSource to use. This parameter can also
take a user name and password if the database requires one.
Server This specifies the IIS sever and protocol used to connect to the server-side
objects.

31.1389-6 CH25 525 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 526 Chapter 25 Accessing Databases with Remote Data Services

N O T E If you will be binding your data to single-threaded ActiveX controls, you should use a
ClassID of 259381D8F2-0288-11D0-9501-00AA00B911A5 for the
AdvancedDataControl. If
you are using apartment-threaded controls, you should use a ClassID of
C5C18AE2-4D6E-11D0-9823-00C04FC29E30. ■

AdvancedDataSpace Object
This object is used as a proxy between the client and server-side business objects by using the
CreateObject method of the AdvancedDataSpace object to create these proxies. When using
this method, send the URL and ProgID of the business object. The URL is of the server that
contains the business object.

The CreateObject method returns a pointer to the server-side business object. The proxy
created is then responsible for packaging and marshaling (transporting) the data to the busi-
ness object. The following code shows an example of how to use the AdvancedDataSpace ob-
ject.
<OBJECT CLASSID=”clsid: 99586D40-DB60-11cf-9D87-00AA00B91181"
ID=”AdvancedDataSpace”
</OBJECT>
Set rsBusinessObject = AdvancedDataSpace.CreateObject(“AdvancedDataFactory”,
➥“http://hr-www;”)

Notice how two pieces of information were sent with the CreateObject method. The first was
the ProgID of the business object—in this case the AdvancedDataFactory. The second was the
URL of the server where this object is located.

AdvancedDataFactory Object
The AdvancedDataFactory is the default server-side object in RDS. It provides for read and
write access to the database. By default, this is the object that the AdvancedDataControl uses
to gather and send data to and from the server.

The following code shows an example of the AdvancedDataFactory at work.

<OBJECT CLASSID=”clsid:9381D8F5-0288-11d0-9501-00AA00B911A5"
ID=AdvancedDataFactory1 HEIGHT=1 WIDTH=1>
</OBJECT>
dim rs_temp
set AdvancedDataFactory1 = ADS1.CreateObject(“AdvancedDataFactory”, “http://hr-
➥www”)
set rs_temp = AdvancedDataFactory1.Query(“DSN=Employees;”, “SELECT * FROM
tbl_employees”)
ADC1.Recordset = rs_temp

Notice how the preceding code uses both the AdvancedDataControl (or, more properly put, the
instantiation of it—ADC1) and AdvancedDataSpace. By default, most of this work can be done in
the instantiation of the AdvancedDataControl.

31.1389-6 CH25 526 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Extending the Phone Book Application 527

Replacing the AdvancedDataFactory

One of the first things you will notice when reading through the Remote Data Service documentation,
or any materials from Microsoft on RDS, is how often Microsoft mentions that the
AdvancedDataFactory is generic and can be replaced with a custom object that can also contain
various business rules.
While the AdvancedDataFactory is actually a nice piece of work, and works just fine, I would take
Microsoft’s advice if you at all have the resources. Why?
The business rules of your application have to exist somewhere. If they don’t exist in the server-side
component, they will most probably exist on your client (browser) in the form of scripting. For simple
applications, this approach is fine. But if you are creating an anti-brevatious application, it would be
better to handle complex business logic in a compiled form running on the server as opposed to an
unmanageable amount of client-side scripts.
In other words, if you have 3,000 lines of code per Web page, it’s going to be a nightmare to update.
And no one will.

Understanding the Limitations of RDS Part

The biggest limitation with the Remote Data Service is its lack of browser independence. The
only browser that can use RDS with no problem is Microsoft’s Internet Explorer 4.0. Netscape
V
won’t be of any help, because it has no support for ActiveX, which is the cornerstone technol- Ch
ogy of RDS.
25
The next big limitation of RDS is the absence of third-party server-side components for use
with RDS. Of course, this can also be thought of as an opportunity. Instead of using the
AdvancedDataFactory or creating your own business object, wouldn’t it be nice if there were a
set of other objects for such things as user registration and bug tracking?

Extending the Phone Book Application

Now, let’s create a real-live application by applying some of this newfound knowledge. Let’s
return to our phone book application of earlier chapters. Only now, let’s use Remote Data Ser-
vices to deliver the data to our users.

This example will run on an IE 4 client with IIS 3.0 for the server. First, I’ll create a read-only
version of the application for anyone to use. Then, I’ll introduce some ASP into the application
to enable for creating pages based on a login the user will provide. After all, someone has to
delete, modify, and add people to the phone database, right?

There are several steps involved in creating this application.

1. Create the grid that will display the data, and the AdvancedDataControl object that it will
bind to.
2. Add HTML buttons and text fields.

31.1389-6 CH25 527 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 528 Chapter 25 Accessing Databases with Remote Data Services

3. Add code for searching through the recordset.

4. Add code for navigating through the recordset.

Actually, before adding the code for the grid and AdvancedDataControl, you need to make
sure there is a datasource named Employees on the server. This can be done by opening the
ODBC Control Panel on the server and adding a new datasource, as shown in Figure 25.3 and
Figure 25.4.

FIG. 25.3
Adding a datasource.

To check your datasource, follow these steps:

1. Click the Start button, and choose Control Panel.

2. Double-click the ODBC32 icon.
3. Click the System DSN tab.
4. Under the System DSN tab, you should find the datasource you need. If not, go ahead
and click the Add button.

FIG. 25.4
Adding a datasource.

Now, let’s look at the code necessary to add the grid and AdvancedDataControl. As you may
remember, adding the AdvancedDataControl is easy. Listing 25.1 shows how.

Listing 25.1 25code01.ASP—Adding the AdvancedDataControl.

<OBJECT classid=”clsid:BD96C556-65A3-11D0-983A-00C04FC29E33”
ID=”AdvancedDataControl1” HEIGHT=1 WIDTH=1>
</OBJECT>

31.1389-6 CH25 528 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Extending the Phone Book Application 529

Notice how the HEIGHT and WIDTH of the control is set to 1. This in effect enables the control to
take up no space. It is already invisible, but you don’t want it to take up any room, either.

Next you can add the datagrid that the data will be bound to. In this example, you will use the
Sheridan Data Grid. Listing 25.2 shows how to embed it into the application.

Listing 25.2 25code02.ASP—Adding the data grid.

<OBJECT CLASSID=”AC05DC80-7DF1-11D0-839E-00A024A94B3A”
ID=”DataGrid” HEIGHT=”200” WIDTH=”600”
CODEBASE=”http://hr-www/apps/phonebook/sheridan.cab”
DATASRC=”#AdvancedDataControl1”>
<PARAM NAME=”_Verson” VALUE=”131072”>
<PARAM NAME=”BackColor” VALUE=”-2147483643”>
<PARAM NAME=”BackColorOdd” VALUE=”-2147483643”>
<PARAM NAME=”BackColorEven” VALUE=”0”>
</OBJECT>

This object is bound to the data that the AdvancedDataControl will gather. You bind the two
together with the DATASRC tag of the grid. DATASRC is one of two new HTML tags in DHTML. So
this tag could actually be used in a table. Part
Now that you have added the grid, add some HTML buttons that will enable you to navigate V
through the recordset (see Listing 25.3).
Ch

Listing 25.3 25code03.ASP—Buttons.

25
<INPUT TYPE=”button” NAME=”First” VALUE=”First” onClick=”MoveFirst”>
<INPUT TYPE=”button” NAME=”Last” VALUE=”Last” onClick=”MoveLast”>
<INPUT TYPE=”button” NAME=”Next” VALUE=”Next” onClick=”MoveNext”>
<INPUT TYPE=”button” NAME=”Prev” VALUE=”Prev” onClick=”MovePrev”>

Now that you have the buttons, you need to write the code that will run when the user clicks on
the buttons, along with some initialization code to run when the page loads up (see Listing 25.4).

Listing 25.4 25code04.ASP—Code to navigate through the data.

Sub Init
Server.Value = “http://hr-www”
Connect.Value = “DSN=”Employees”
SQL.Value = “SELECT * FROM tbl_employees”
AdvancedDataControl1.Server = Server.Value
AdvancedDataControl1.Connect = Connect.Value
AdvancedDataControl1.SQL = SQL.Value
End Sub
Sub MoveFirst
AdvancedDataControl1.MoveFirst
End Sub

continued

31.1389-6 CH25 529 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 530 Chapter 25 Accessing Databases with Remote Data Services

Listing 25.4 Continued

Sub MoveLast
AdvancedDataControl1.MoveLast
End Sub
Sub MoveNext
On Error Resume Next
AdvancedDataControl1.MoveNext
If ERR.Number <> 0 Then
AdvancedDataControl.MoveLast
End If
End Sub
Sub MovePrev
On Error Resume Next
AdvancedDataControl1.MovePrevious
If ERR.Number <> 0 Then
AdvancedDataControl.MoveFirst
End If
End Sub

Now you can combine all of this code into one ASP page and give it a name. Listing 25.5 shows
the entire page, Y.asp.

Listing 25.5 25code05.ASP—The phone book.

<HTML>
<HEAD>
<TITLE>
Phone Book Application
</TITLE>
<SCRIPT LANGUAGE=”VBSCRIPT>
Sub Init
Server.Value = “http://hr-www”
Connect.Value = “DSN=”Employees”
SQL.Value = “SELECT * FROM tbl_employees”
AdvancedDataControl1.Server = Server.Value
AdvancedDataControl1.Connect = Connect.Value
AdvancedDataControl1.SQL = SQL.Value
End Sub
Sub MoveFirst
AdvancedDataControl1.MoveFirst
End Sub
Sub MoveLast
AdvancedDataControl1.MoveLast
End Sub
Sub MoveNext
On Error Resume Next
AdvancedDataControl1.MoveNext
If ERR.Number <> 0 Then
AdvancedDataControl.MoveLast
End If
End Sub
Sub MovePrev

31.1389-6 CH25 530 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Extending the Phone Book Application 531

On Error Resume Next

AdvancedDataControl1.MovePrevious
If ERR.Number <> 0 Then
AdvancedDataControl.MoveFirst
End If
End Sub
</SCRIPT>
</HEAD>
<BODY BGCOLOR=”white”>
<OBJECT CLASSID=”AC05DC80-7DF1-11D0-839E-00A024A94B3A”
ID=”DataGrid” HEIGHT=”200” WIDTH=”600”
CODEBASE=”http://hr-www/apps/phonebook/sheridan.cab”
DATASRC=”#AdvancedDataControl1”>
<PARAM NAME=”_Verson” VALUE=”131072”>
<PARAM NAME=”BackColor” VALUE=”-2147483643”>
<PARAM NAME=”BackColorOdd” VALUE=”-2147483643”>
<PARAM NAME=”BackColorEven” VALUE=”0”>
</OBJECT>
<BR>
<INPUT TYPE=”button” NAME=”First” VALUE=”First” onClick=”MoveFirst”>
<INPUT TYPE=”button” NAME=”Last” VALUE=”Last” onClick=”MoveLast”>
<INPUT TYPE=”button” NAME=”Next” VALUE=”Next” onClick=”MoveNext”>
<INPUT TYPE=”button” NAME=”Prev” VALUE=”Prev” onClick=”MovePrev”>
<BR> Part
<OBJECT classid=”clsid:BD96C556-65A3-11D0-983A-00C04FC29E33”
ID=”AdvancedDataControl1” HEIGHT=1 WIDTH=1> V
</OBJECT>
</BODY> Ch
</HTML>
25
Figure 25.5 shows what the phone book page should now look like.

FIG. 25.5
The phone book
application.

Now let’s create a login page for the phone book. This page will ask just for a login name. It will
then check this login to see if it is the administrator of the phone book. If it is, it will enable the
user to make changes to the phone book and update the database accordingly.

Listing 25.6 shows the ASP page that asks for the user login.

31.1389-6 CH25 531 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 532 Chapter 25 Accessing Databases with Remote Data Services

Listing 25.6 25code06.ASP—Getting the user login.

<HTML>
<HEAD>
<TITLE>
Phone Book Login
</TITLE>
</HEAD>
<BODY BGCOLOR=”white”>
<FORM ACTION=”y.asp” METHOD=”POST>
Phone Book Login:
<INPUT TYPE=”text” NAME=”login”>
<BR>
<INPUT TYPE=”submit” NAME=”Submit”>
</FORM>
</BODY>
</HTML>

What code do you have to add to enable updating the database? Actually, very little. The data
grid is already designed to enable edits to the fields. So all you really need to do is add a button
called Update and a function that actually does it (see Listing 25.7). These buttons will fire off
the built-in methods of the AdvancedDataControl, SubmitChanges, and Refresh.

Listing 25.7 25code07.ASP—Update: The button and the function.

Sub Update
AdvancedDataControl1.SubmitChanges
AdvancedDataControl1.Refresh
End Sub
.
.
.
<INPUT TYPE=”Button” Name=”Submit” VALUE=”Submit Changes” OnClick=”Update”>

Listing 25.8 is the new code integrated into the phone book application, and Figure 25.6 shows
the new and improved page when it is the admin login.

Listing 25.8 25code08.ASP—New and improved phone book.

<HTML>
<HEAD>
<TITLE>
Phone Book Application
</TITLE>
<SCRIPT LANGUAGE=”VBSCRIPT>
Sub Init
Server.Value = “http://hr-www”
Connect.Value = “DSN=”Employees”
SQL.Value = “SELECT * FROM tbl_employees”
AdvancedDataControl1.Server = Server.Value
AdvancedDataControl1.Connect = Connect.Value

31.1389-6 CH25 532 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 Extending the Phone Book Application 533

AdvancedDataControl1.SQL = SQL.Value
End Sub

<% IF Request.Form(“Login”) = “Admin” Then %>

Sub Update
AdvancedDataControl1.SubmitChanges
AdvancedDataControl1.Refresh
End Sub
<%End IF%>
Sub MoveFirst
AdvancedDataControl1.MoveFirst
End Sub
Sub MoveLast
AdvancedDataControl1.MoveLast
End Sub
Sub MoveNext
On Error Resume Next
AdvancedDataControl1.MoveNext
If ERR.Number <> 0 Then
AdvancedDataControl.MoveLast
End IF
End Sub
Sub MovePrev
On Error Resume Next Part
AdvancedDataControl1.MovePrevious
If ERR.Number <> 0 Then V
AdvancedDataControl.MoveFirst
End If Ch
End Sub
</SCRIPT> 25
</HEAD>
<BODY BGCOLOR=”white”>
<OBJECT CLASSID=”AC05DC80-7DF1-11D0-839E-00A024A94B3A”
ID=”DataGrid” HEIGHT=”200” WIDTH=”600”
CODEBASE=”http://hr-www/apps/phonebook/sheridan.cab”
DATASRC=”#AdvancedDataControl1”>
<PARAM NAME=”_Verson” VALUE=”131072”>
<PARAM NAME=”BackColor” VALUE=”-2147483643”>
<PARAM NAME=”BackColorOdd” VALUE=”-2147483643”>
<PARAM NAME=”BackColorEven” VALUE=”0”>
</OBJECT>
<BR>
<INPUT TYPE=”button” NAME=”First” VALUE=”First” onClick=”MoveFirst”>
<INPUT TYPE=”button” NAME=”Last” VALUE=”Last” onClick=”MoveLast”>
<INPUT TYPE=”button” NAME=”Next” VALUE=”Next” onClick=”MoveNext”>
<INPUT TYPE=”button” NAME=”Prev” VALUE=”Prev” onClick=”MovePrev”>
<% IF Request.Form(“Login”) = “Admin” Then %>
<INPUT TYPE=”Button” Name=”Submit” VALUE=”Submit Changes”
➥OnClick=”Update”>
<%End If%>
<BR>
<OBJECT classid=”clsid:BD96C556-65A3-11D0-983A-00C04FC29E33”
ID=”AdvancedDataControl1” HEIGHT=1 WIDTH=1>
</OBJECT>
</BODY>
</HTML>

31.1389-6 CH25 533 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 534 Chapter 25 Accessing Databases with Remote Data Services

FIG. 25.6
The phone book with
the update.

Of course, this example doesn’t have any real security, as anyone who knows to log in as
Admin can make changes to the database. A more secure example would have both a login and
a password. You could also have a password on the database itself, and send that information
when setting up the connection to the database.

From Here…
In this chapter, you have examined the Remote Data Services and seen how RDS can help you
create data-bound Web pages. Leveraging your knowledge of ADO, you have seen how to
utilize RDS even further. The applications you can create are endless.

You have also seen some of the limitations of Remote Data Services, most significant of these
being the lack of browser independence. If you use RDS, you are playing in the realm of
Internet Explorer.

However, Remote Data Services offer developers a new level of freedom in building Web pages
driven by data. Using this technology, combined with ASP and Dynamic HTML, there is, liter-
ally, very little you cannot have your Web site do.

For related information, please see the following chapters:

■ Chapter 24, “Managing Your Database,” for more information on database development
and management.
■ Chapter 30, “Application Development for the Corporate Intranet/Extranet,” to uncover
the major issues involved in creating Web-based applications for these two arenas.

31.1389-6 CH25 534 10/26/98, 9:46 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH25 lp#3
 26
535

C H A P T E R

Use the ADO object to Load

Putting It All Together: and update Database
Information

Creating a Dynamic Learn how to implement the differ-

ent ADO objects to connect and
extract information from the data-

Database Application base and how to use the ADO object

to update information in the
DataWidget database.

Implement Visual InterDev’s

Data Control and Data Form
Wizard
Utilize the design-time data control
I n this chapter, you will convert the Amazing DataWidgets to visually create the database
WareHouse Web site from a static Web site into a connections and create ASPs to
database-driven Web site. Specifically, you will use the administer database information.
Active Data Object (ADO) to load the DataWidgets prod- Part
uct into application-level variables to be used throughout Configure Web Site and Appli-
the online catalog system. Database code will be devel- cation Security Settings V
oped to add and remove items from the current inventory. Discover how to control and man- Ch
age who can access your Web site.
This chapter will also demonstrate how to use Visual 26
InterDev to generate an administration HTML interface
that enables new products to be added to the Products
table. In addition, I will demonstrate how to control access
to the HTML administration interface to control access
permissions to the HTML administration interface. ■

32.1389-6 CH26 535 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 536 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Loading the Products Table

The Amazing DataWidget WareHouse Web site provides an online catalog and order entry
system. This catalog system was demonstrated in Chapter 19, “Putting It All Together: Creat-
ing an Online Registration System.” That chapter demonstrated how the catalog information
was originally loaded from variables stored in Active Server Pages into application-level vari-
ables. If new products needed to be added to the site or product name, description or pricing
information changed, the changes would have to be made directly in the ASP code. Now you
are going to expand the Web site’s functionality by retrieving the information from a database
when the application starts. From the user perspective, the site will not change appearance
when viewed by a browser. However, the changes will be evident to the Web administrators
who are responsible for the site’s upkeep and maintenance.

Create the ActiveX Data Object Connection

The first step in retrieving product information from the Amazing DataWidgets database is to
establish the ODBC connection to the database. The ActiveX Data Object (ADO) is responsible
for establishing the database connection. The ADO object provides a programmable communi-
cation mechanism to manage and manipulate database information through the RecordSet
object. The RecordSet object gives the programmer the ability to utilize multiple database
connections, support stored procedures and triggers, and implement advanced data control
and management through multiple cursor types and cache management to support scaleable,
distributed Web-based applications.

To use the ADO object, you will create an ASP function called LoadProductsTable. The
LoadProductsTable function actually performs two roles. The first role is to retrieve the avail-
able catalog information. This information will be used to write what products are currently
available for sale on the Web page. Second, the same RecordSet object is going to be used to
create a temporary, empty shopping basket for each individual user visiting the site. The
session-level variables are needed to enable the ASP code to be modified according to the
users’ actions. The application-level product listing is needed to remain constant throughout
the entire application. This dual approach provides the capability to load application- and
session-level variables with just one connection to the database.

To tap into this functionality for DataWidget’s Web site, first create the ADO Connection ob-
ject. Use the CreateObject method to reference the ADO object, as shown in the following
example.
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open strODBC

This code example creates an object named DataConn to reference an instantiated ADO con-
nection using the CreateObject method. After the connection object is created, the ADO opens
the ODBC datasource name represented by the variable strODBC. The Connection object rep-
resents a direct interface with a specific datasource to the DataWidgets database. For more
information on using the CreateObject method, please see Chapter 12, “Controlling the Server
with the Server Object and Using Transaction Processing with the ObjectContext Object.”

32.1389-6 CH26 536 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Loading the Products Table 537

Next, create Command and Recordset objects to send commands to the database and store the
returned information, as shown in the following snippet:
Set cmdConn = Server.CreateObject(“ADODB.Command”)
Set cmdRS = Server.CreateObject(“ADODB.Recordset”)

The Command object is used to provide database command strings that are executed against a
datasource. The Recordset object is used as a container to store the results of the Command
object.

Note that the Recordset object does not need explicitly defined Connection and Command ob-
jects to retrieve information from the database, as shown above. The Recordset object auto-
matically creates a new Connection object for every request sent by the Recordset object, even
if the same connection string is used to establish the database connection. Therefore, in order
to maximize your connection resources, the Connection object should be used to minimize the
amount of database connections created throughout the application.

The Command object enables you to explicitly control the ADO object. For example, the follow-
ing code example creates a SQL statement that selects specific fields from the Products table
and explicitly defines that the Command object is a text-based SQL statement.
cmdConn.CommandText = “SELECT ProductName, Description, UnitPrice FROM Products”
cmdConn.CommandType = 1 ‘=== Evaluates CommandText as a textual
➥definition of a command.

After the command string has been set, the following code sets the active datasource connec-
tion and opens the instantiated Recordset to contain the results of the command string.
Set cmdConn.ActiveConnection = DataConn
cmdRS.Open cmdConn, , 0, 1 ‘=== Constants: adOpenForwardOnly, adLockReadOnly Part
This code generates a forward-only, read-only RecordSet object. V
Note that in order to call the ADO constants by name, the Adovbs.inc must be included in the Ch
ASP. You can reference the file by using a server-side include:
26
<!--#INCLUDE FILE = “adovbs.inc”-->

After the Recordset is created, you have to test to see if any records have been returned into
the Recordset. If no records have been returned, the record’s BOF and EOF (begin of file and
end of file) properties are True, as shown in the following If...Then statement.
If cmdRS.EOF And cmdRS.BOF Then
‘=== no records
Else
‘=== records found
End if

If results are returned into the Recordset, the script iterates through the recordset creates a
customized user array. The array extracts the ProductName, Description, and UnitPrice fields
from the Recordset and creates a fourth array item. This last item will represent the amount of
each product item the user has ordered.

32.1389-6 CH26 537 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 538 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Do While Not cmdRS.EOF

i = i + 1
aProductArray(i, 1) = cmdRS(“ProductName”)
aProductArray(i, 2) = cmdRS(“Description”)
aProductArray(i, 3) = cmdRS(“UnitPrice”)
aProductArray(i, 4) = 0
cmdRS.MoveNext
Loop

After the array is populated, the RecordSet and Connection objects are closed and the entire
local array is passed as a variable back to the calling statement:
cmdRS.Close
dataconn.Close
LoadProductsTable = aProductArray

The entire LoadProductsTable function is displayed in Listing 26.1.

Listing 26.1 Returning a customized Recordset via an array.

Function LoadProductsTable(strODBC)

‘=== Loads the ProductName, Description and Price Fields from the Product
table
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open strODBC
Set cmdConn = Server.CreateObject(“ADODB.Command”)
Set cmdRS = Server.CreateObject(“ADODB.Recordset”)
cmdConn.CommandText = “SELECT ProductName, Description, UnitPrice,
UnitsInStock FROM Products”
cmdConn.CommandType = 1
Set cmdConn.ActiveConnection = DataConn
cmdRS.Open cmdConn, , 0, 1

If cmdRS.EOF And cmdRS.BOF Then

‘=== no records
Else
Dim icount
cmdRS.MoveFirst
Do While Not cmdRS.EOF
icount = icount + 1
cmdRS.MoveNext
Loop
cmdRS.MoveFirst
ReDim aProductArray(icount, 4)

Do While Not cmdRS.EOF

i = i + 1
aProductArray(i, 1) = cmdRS(“ProductName”)
aProductArray(i, 2) = cmdRS(“Description”)
aProductArray(i, 3) = cmdRS(“UnitPrice”)
aProductArray(i, 4) = 0
cmdRS.MoveNext
Loop

32.1389-6 CH26 538 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Loading the Products Table 539

End If
cmdRS.Close
Dataconn.Close
LoadProductsTable = aProductArray

End Function

Now that the LoadProductsTable function extracts the customized RecordSet variable from the
DataWidget’s database, you will develop the script to call the function. To implement this new
function, call the LoadProductsTable function as shown below:
Application(“aSoftware”) = LoadProductsTable(Application(“strODBC”))

This code creates an application-variable named aSoftware and assigns it the array value re-
turned by LoadProductsTable. However, you have to trap whether the LoadProductsTable
function returns an empty recordset. To eliminate the possibility that an empty string will be
assigned to the catalog array, use the IsArray VBScript function to determine the results of
the LoadProductTable function:
Dim aTestArray
aTestArray = LoadProductsTable(Application(“strODBC”))
If IsArray(aTestArray) Then
Application(“aSoftware”) = aTestArray
Set aTestArray = Nothing ‘=== Destroys array
Else
Err.Raise (1) ‘=== Generate an error
End if

Next, trap any errors that may occur to prevent additional errors from filtering throughout the
site. In this situation, you have already created an error.asp that accepts various arguments to Part
display errors messages to the user. V
If Err Then Ch
page = Request.ServerVariables(“SCRIPT_NAME”)
errortype = “Critical”
errormsg = “A Critical Error has occurred on page “ &page &”. <BR>Please
26
contact the System Administrator”
urlstr=”error.asp?errortype=” &Server.URLEncode(errortype) &”&errormsg=”
&Server.URLEncode(errormsg)
Response.Redirect urlstr
End If

Now all this logic can be applied together to form the script shown in Listing 26.2.

Listing 26.2 Loading the Recordset values into application- and

session-level variables.
<%@ LANGUAGE=”VBSCRIPT” %>
<% On Error Resume Next
‘===================== Procedures ==========================
Sub LoadInitValues()

continues

32.1389-6 CH26 539 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 540 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Listing 26.2 Continued

‘=== Load Catalog Values by hand
Dim aCatalog(2,2)
aCatalog(1,1) = “Software Components”
aCatalog(1,2) = “SoftwareCatalog.asp”
aCatalog(2,1) = “Hardware Components”
aCatalog(2,2) = “HardwareCatalog.asp”

‘=== Set General variables

Application(“aCatalog”) = aCatalog
Application(“strODBC”) = “DataWidget”
Application(“iTimeOut”) = 30
End Sub

‘===================== Begin Application Code =================

Application.Lock
Application(“aSoftware”) = LoadProductsTable(Application(“strODBC”))

If Err Then
page = Request.ServerVariables(“SCRIPT_NAME”)
errortype = “Critical”
errormsg = “A Critical Error has occurred on page “ &page &”. <BR>Please
contact the System Administrator”
urlstr=”error.asp?errortype=” &Server.URLEncode(errortype) &”&errormsg=”
&Server.URLEncode(errormsg)
Response.Redirect urlstr
End If

Application.UnLock
‘=== Now load the users empty basket
Session(“basket”)=Application(“aSoftware”) ‘=== Create user’s new, but empty
basket
Response.Redirect “main.asp”
%>

The application-level variables are used to produce the catalog listing shown in Figure 26.1.

Modifying the Inventory Database

Now it’s time to create the database capabilities to handle when a user places an order for a
product. You will create two functions to add and delete items from the Inventory table.

Deleting Items from the Inventory

To delete items from the current Inventory table in the DataWidgets database, a
DeleteProductInventory function is created to remove the ordered amount from inventory.
The DeleteProductInventory function accepts the ODBC connection name, the product name
to be deleted, and the amount of items to be deleted. You will build the function using the same
Connection, Command, and Recordset objects used in the LoadProductsTable example.

32.1389-6 CH26 540 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Modifying the Inventory Database 541

FIG. 26.1
The catalog list is
extracted from the
ActiveX Data Object.

However, there are a couple of differences to be pointed out with this new function. First is
obviously the SQL statement. The UPDATE statement is used to find the current value of an
existing field and modify the value. In this situation, the UnitsInStock field is being updated
for the specific product name:
sql = “UPDATE Products Set UnitsInStock = UnitsInStock -” Part
sql = sql & iAmount & “ WHERE ProductName = ‘“ & strProductName & “‘“ V
Furthermore, because the RecordSet object does not return any information from the query, Ch
the provider automatically closes the recordset. Therefore, the cmdRS.Close command is not
needed. The DelteProductInventory function is illustrated in Listing 26.3. 26

Listing 26.3 Deleting products from inventory.

Function DeleteProductInventory(strODBC, strProductName, iAmount)
‘=== This function delete inventory once an item is ordered
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open strODBC
Set cmdConn = Server.CreateObject(“ADODB.Command”)
Set cmdRS = Server.CreateObject(“ADODB.Recordset”)
sql = “UPDATE Products Set UnitsInStock = UnitsInStock -”
sql = sql & iAmount & “ WHERE ProductName = ‘“ & strProductName & “‘“
cmdConn.CommandText = sql
cmdConn.CommandType = 1
Set cmdConn.ActiveConnection = DataConn
cmdRS.Open cmdConn, , 0, 1
Dataconn.Close

continues

32.1389-6 CH26 541 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 542 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Listing 26.3 Continued

If Not Err Then
ret = True
End If
DeleteProductInventory = ret
End Function

This listing demonstrates the DeleteProductInventory function that accepts the ODBC string
name, the name of the product that is ordered, and how many of that item is ordered. With that
information, the function creates a new database connection and deletes the amount of product
from the inventory when a user orders an item. It is important to reemphasize that you want to
create and release the database connections as needed in the calling code. Your Web sites and
applications could possibly have thousands of concurrent users. Let ODBC 3.0 and the
Microsoft Transaction Server (demonstrated in Chapter 32, “Putting It All Together: Creating
an N-Tier Online Catalog Application”) handle the connection pooling for you.

You may be tempted to create a subroutine to perform this inventory functionality because no
information is being returned via the Recordset object. However, you actually do want to re-
turn whether the function was executed successfully. To call function from the client ASP, use
the following code:
ret = DeleteProductInventory(Application(“strODBC”), strProductName,
iUnitQuantity)

To examine the return value, you can check the value of the return value or examine the func-
tion inline, as shown below:
If DeleteProductInventory(Application(“strODBC”), strProductName, iUnitQuantity)
then
‘=== Record is deleted
End If

Adding Records to the Database

The AddProductInventory function is used to complement the DeleteProductInventory
functionality, but is responsible for adding items to the Inventory database. The
AddProductInventory, as shown in Listing 26.4, uses the UPDATE statement to create a query
string that adds the amount an amount of an item to the Products table.

Listing 26.4 Creating a function to add inventory via the ADO connection.
Function AddProductInventory(strODBC, strProductName, iAmount)
‘=== This function adds inventory to the database
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open strODBC
Set cmdConn = Server.CreateObject(“ADODB.Command”)
Set cmdRS = Server.CreateObject(“ADODB.Recordset”)

32.1389-6 CH26 542 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 543

sql = “UPDATE Products Set UnitsInStock=UnitsInStock + “

sql = sql & iAmount & “ WHERE ProductName = ‘“ & strProductName & “‘“
cmdConn.CommandText =sql
cmdConn.CommandType = 1
Set cmdConn.ActiveConnection = DataConn
cmdRS.Open cmdConn, , 0, 1
Dataconn.Close
If Not Err Then
ret = True
End If
AddProductInventory = ret

End Function

Likewise, the AddProductInventory function can be called using the following code:
ret = AddProductInventory(Application(“strODBC”), strProductName, iUnitQuantity)

To examine the return value, you can check the value of the return value or examine the func-
tion inline, as shown below:
If AddProductInventory(Application(“strODBC”), strProductName, iUnitQuantity)
then
‘=== Record has been added
End If

Using Visual InterDev to Access the Data

As you can see, the ADO provides a fully programmable interface to tailor database access to
your application needs. However, to speed up development time to create the database access, Part
you can use the visual database tools found in Visual InterDev. V
Ch
Using Visual InterDev’s Design Data Control to Retrieve Records 26
Visual InterDev uses ActiveX design-time controls to help speed up the development of Active
Server code. The database connection features are directly controlled by the Data Command
control. To start generating the database code, start a new Web project in Visual InterDev. To
start using the design-time controls, a database connection to the current project is needed.
This connection is used at runtime to provide a real-time connection to the datasource at de-
sign time. This design-time capability enables a direct interface to test and debug your data-
base applications. You will use the Data Command control to generate a function to retrieve the
current amount of in-stock items for a particular product.

To add a database connection, right-click the Visual InterDev project folder and select Add
Database Connection. A dialog box will appear that enables you to select from an existing file
or machine data source or create a new datasource, as shown in Figure 26.2.

32.1389-6 CH26 543 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 544 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

FIG. 26.2
The Visual InterDev
Connection wizard
enables you to create a
new datasource name
or select from an
existing database
connection.

After selecting the DataWidgets datasource, the Connection wizard prompts for information
about the data connection. Figure 26.3 illustrates the interface to set the database connection
properties, such as the connection and command timeout properties.

FIG. 26.3
The Data Connection
dialog box is used to set
general properties of the
database connection.

The Run-Time tab of the Data Connection dialog box is used to set the runtime properties of
the database connection, as shown in Figure 26.4. The runtime connection properties are used
to set and monitor access to the datasource. This information is stored on the Web server and
is transmitted to the database source when the data connection is embedded. This is actually
an important architecture aspect of access information from the Web server. By enabling the
Web server to keep and maintain runtime connection information, the intercepting of database
username and permissions is minimized because you don’t have to continuously transmit this
information over the Internet.

After you enter the connection information, the Connection wizard inserts the database con-
nection information into the Session_OnStart event of the GLOBAL.ASA file, as shown in Fig-
ure 26.5.

32.1389-6 CH26 544 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 545

FIG. 26.4
The Run-Time tab sets
the runtime database
permissions needed to
access the datasource.

FIG. 26.5
The Connection wizard
creates the database
connection information
in the
Session_OnStart
event of the
GLOBAL.ASA file.

Part
V
Ch

26

N O T E Keep in mind that Session objects are only applicable with cookie-compliant browsers
and networks. You might want to convert this database connection information into a non-
visual ASP file and track information using hidden HTML fields to enable non-cookie-supporting
environments to properly access the Web site. ■

For more information on using the Session object and cookies, please refer to Chapter 13,
“Managing the User Session: The Session Object.” Listing 26.5 is the listing of using the
Session_OnStart event in the GLOBAL.ASA file to set the database connection properties.

32.1389-6 CH26 545 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 546 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Listing 26.5 Setting database connection properties in the

Session_OnStart event.
<SCRIPT LANGUAGE = VBScript RUNAT = Server>
Sub Session_OnStart
‘==Visual InterDev Generated - DataConnection startspan==
‘--Project Data Connection
Session(“DataConn_ConnectionString”) = “DSN=datawidgets;DriverId=25;FIL=MS
Access;MaxBufferSize=512;PageTimeout=5;”
Session(“DataConn_ConnectionTimeout”) = 15
Session(“DataConn_CommandTimeout”) = 30
Session(“DataConn_RuntimeUserName”) = “”
Session(“DataConn_RuntimePassword”) = “”
‘==Visual InterDev Generated - DataConnection endspan==
End Sub
</SCRIPT>

Now that the database connection is created, you can use the design-time Data Control to
create ADO connections to the database. To use the Data Command control, right-click an ASP
and select Insert ActiveX Control. From the Design-Time tab control, select the Data Com-
mand control, as shown in Figure 26.6.

FIG. 26.6
Use the design-time
Data Command control
to create connections to
the Data Widget
database.

The Data Command Properties dialog box provides a visual interface to set the properties of the
data connection. The Control tab, as shown in Figure 26.7, enables you to set the name of the
connection, the datasource connection, the type of Command object, and the command string.

You can then use the SQL Builder button illustrated in Figure 26.7 to open Visual InterDev’s
Query Builder. The Query Builder provides a visual development environment to create and
test your database queries. Because you want to generate a query that selects the current
amount in inventory, a simple query can be created to extract this information, as shown in the
Figure 26.8.

After successfully testing and saving the query, the Data Control wizard generates the Active
server Page code, as shown in Listing 26.6, to retrieve the database information from the query
created in the Query Designer.

32.1389-6 CH26 546 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 547

FIG. 26.7
The Data Connection
Properties dialog box
displays a visual
interface to set the
properties of the
Command object.

FIG. 26.8
The Query Designer
provides a visual query
building and testing
interface.

Part
V
Ch

26

Listing 26.6 Creating ASP code to retrieve database information using

the Query Designer.
<%
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.ConnectionTimeout = Session(“DataConn_ConnectionTimeout”)
DataConn.CommandTimeout = Session(“DataConn_CommandTimeout”)
DataConn.Open Session(“DataConn_ConnectionString”),

continues

32.1389-6 CH26 547 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 548 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

Listing 26.6 Continued

Session(“DataConn_RuntimeUserName”), Session(“DataConn_RuntimePassword”)
Set cmdTemp = Server.CreateObject(“ADODB.Command”)
Set DataWidget = Server.CreateObject(“ADODB.Recordset”)
cmdTemp.CommandText = “SELECT UnitsInStock FROM Products WHERE (ProductName = “
& ProdName & “)”
cmdTemp.CommandType = 1
Set cmdTemp.ActiveConnection = DataConn
DataWidget.Open cmdTemp, , 0, 1
%>

Now, you can modify this generated code to create the GetProductInventory function shown
in Listing 26.7.

Listing 26.7 Creating a function from the ASP code generated by the
Query Designer.
Function GetProductInventory(strODBC, strProductName)
‘=== This function retrieve that current inventory for a product
Set DataConn = Server.CreateObject(“ADODB.Connection”)
DataConn.Open strODBC
Set cmdTemp = Server.CreateObject(“ADODB.Command”)
Set cmdRS = Server.CreateObject(“ADODB.Recordset”)
sql = “SELECT UnitsInStock FROM Products where ProductName= ‘“ &
strProductName & “‘“
cmdTemp.CommandText =sql
cmdTemp.CommandType = 1
Set cmdTemp.ActiveConnection = DataConn
cmdRS.Open cmdTemp, , 0, 1
If cmdRS.EOF And cmdRS.BOF Then
ret = 0
Else
ret = cmdRS(“UnitsInStock”)
End If
cmdRS.Close
Dataconn.Close
GetProductInventory = ret
End Function

This GetProductInventory function returns the amount of items in the current inventory. Now,
with the previously created AddProductInventory and DeleteProductInventory functions, you
can add and remove items from the user’s shopping basket and remove the appropriate items
from inventory, as shown in Listing 26.8.

Listing 26.8 Adding and removing items from the shopping basket
and inventory.
Call SetSessionVariables() ‘=== Set session
variables
If ValidateNumber(Session(“UserAccountNumber”)) Then ‘=== Is it a valid

32.1389-6 CH26 548 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 549

account number
aUsersBasket=Session(“basket”) ‘=== Create a local copy of the
users shopping basket
iProductsOrdered = UBound(aUsersBasket,1) ‘=== How many items are there?
For iCounter = 1 to iProductsOrdered ‘=== Start looping through
basket
iUnitQuantity=clng(aUsersBasket(iCounter,4)) ‘=== How many items are
ordered?
If iUnitQuantity > 0 Then
strProductName = aUsersBasket(iCounter,1) ‘=== Get name of
ordered item
If GetProductInventory(“datawidgets”, “Mind-Bender Control”) > 15
Then ‘=== Restock inventory?
‘=== Everything is fine, remove inventory
ret = DeleteProductInventory(Application(“strODBC”),
strProductName, iUnitQuantity)
Else
‘=== Add more items to inventory
ret = AddProductInventory(Application(“strODBC”),
strProductName, 25) ‘=== Restock items
ret = DeleteProductInventory(Application(“strODBC”),
strProductName, iUnitQuantity)
End If
End If
Next
Response.Redirect “summary.asp” ‘=== Order is successful print
receipt
Else ‘=== Invalid Account number
dim errortype, returnpage, fieldname, urlstr
errortype=”Retry”
returnpage =”customer.asp”
fieldname =”Account Number” Part
urlstr=”error.asp?errortype=” &Server.URLEncode(errortype) & “&returnpage=”
&Server.URLEncode(returnpage) & “&fieldname=” &Server.URLEncode(fieldname)
V
Response.Redirect urlstr Ch
End If
26
Listing 26.8 first checks to ensure that the account number passes verification. If the account
number is correct, the script determines the amount of items in the user’s shopping basket.
Next, the amount of existing inventory for that product is determined to ensure that enough
items are in stock to fill the order. If the in-stock inventory is greater than a minimum thresh-
old value, the products are removed from the inventory. If not enough inventory is available,
the AddProductInventory function restocks the inventory and then the items are processed. If
the user enters an invalid account number, the error function is used to prompt the user for
valid information, as shown in Figure 26.9.

If the account number is valid, the script continues to process all the items in the shopping
basket and generates a virtual receipt for the order, as shown in Figure 26.10.

32.1389-6 CH26 549 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 550 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

FIG. 26.9
The order is not
processed if an invalid
account number is
entered.

FIG. 26.10
The inventory items are
adjusted and the virtual
receipt is generated.

32.1389-6 CH26 550 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 551

Using the Visual InterDev Forms Wizard to Create the Product

Administration Interface
Now that the user interface and ordering processing sections are complete, you are going to
create the HTML administration utilities to update the product information. This information
can then be updated from any browser, depending on proper authorization. To accomplish this,
you will use the Visual InterDev Forms wizard to quickly create Active Server Pages to provide
insert, update, and delete capabilities to the Products table.

To create the Web-based administration interface, create a new Visual InterDev project and add
the database connection, as described earlier in the chapter. To start the Forms wizard, select
New from the File menu. Now select the Data Form Wizard from the File Wizards tab. After
naming the file, the first page in the Data Form wizard appears, as shown in Figure 26.11. This
page of the wizard prompts for the appropriate datasource connection and for the title that
appears on the top of each page.

FIG. 26.11
The Data Form wizard is
used to collect the
information for the
database connection
and title of the HTML
page.

Part
V
Ch

26
After selecting the DataWidget datasource and entering Amazing DataWidget WareHouse for
the page titles, the second page of the wizard requests the source type of the housed informa-
tion. Because you want to provide administration access to the Products table, select Table.
After selecting Next, the third page of the Form wizard enables you to select which fields to
display on the page, as shown in Figure 26.12.

Page four of the wizard enables you to decide whether the user has the ability to insert, update,
or delete the displayed data, or is just allowed to view the data. After selecting the appropriate
permissions, the fifth page in the wizard prompts you to decide how the database information
will be viewed, as shown in Figure 26.13.

32.1389-6 CH26 551 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 552 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

FIG. 26.12
The Forms wizard
enables you to select
and control the heading
names for the displayed
fields.

FIG. 26.13
The Data Form wizard
enables you to decide
whether the table
information is going to
be displayed in a List
view or a Form view.

The List view provides a sequential list of records to be displayed on a Web page extracted
from a database table. The Form view provides a forms-based view of the individual records.

Finally, the last page of the wizard provides the capability to select a theme. Themes are imple-
mented by using Cascading style sheets (CSSs) to control the look and feel of your site. Keep
in mind that all browsers do not support CSS. After you select the theme, the Form Wizard
generates three Active Server Pages that are used to display the data.

The List view page, which is shown in Figure 26.14, displays the sequential listing of records in
the Product table. From this page, you can systematically navigate through all the records in
the Products table.

The List view page, as seen in Figure 26.14, enables you to drill down into the details of a data-
base record by selecting the product number. The drill-down capability enables the user to edit
the individual fields in the Form view page, as shown in Figure 26.15. Not only can you edit and
update records, but you can also delete, create, and provide filters on the existing data.

32.1389-6 CH26 552 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Using Visual InterDev to Access the Data 553

FIG. 26.14
The List view provides
the capability to
sequentially list and
filter information.

FIG. 26.15
The Form view
provides the capability
to edit the individual
fields in the specific Part
record.
V
Ch

26

For example, the Amazing DataWidget WareHouse has created a new design-time control that
is available for market. The site administrator, as shown in Figure 26.16, uses the Form view to
enter the new information. In this example, the CoffeeX ActiveX control is added to the data-
base via the HTML administration page.

32.1389-6 CH26 553 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 554 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

If you navigate back to the List view, you will notice that the new item is now displayed in the
product list, as shown in Figure 26.17.

FIG. 26.16
The Form view enables
new product informa-
tion to be added to the
database.

FIG. 26.17
The List view displays
the new product
information.

32.1389-6 CH26 554 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 Controlling Access to the Site 555

As a result, when new users browse the catalog application, the new CoffeeX control product is
displayed in the online catalog and is available for sale. Figure 26.18 displays the new product
as it appears in the catalog listing.

FIG. 26.18
The product information
is now automatically
displayed to new users
that browse the site!

Now the users can seamlessly order the new components without you, the developer, having to Part
change any ASP code on the Web site. Figure 26.19 illustrates two orders for the new CoffeeX V
control.
Ch
If you try to view the List page generated by the Data Form wizard and the following OLE DB
error occurs, make sure that a hard coded path statement is not defined in the database con- 26
nection string.
Microsoft OLE DB Provider for ODBC Drivers error ‘80004005’
[Microsoft][ODBC Microsoft Access 97 Driver] ‘(unknown)’ isn’t a valid path.
Make sure that the path name is spelled correctly and that you are connected to
the server on which the file resides.
/ASPDBstoreAdmin/Product AdminList.asp, line 266

Controlling Access to the Site

Although the capability to control the information in the Product table provides remote HTML
administration from any browser in the world, deciding who has access to change this informa-
tion is paramount. Likewise, for many intranet sites that need extranet capabilities, the capability
to grant and restrict access to their sites is critical to maintaining control over sensitive informa-
tion. As you saw in Chapter 10, “The Composition of an Active Server Application,” there are two
ways to provide access to your Web sites: at the Web server level and at the file level.

32.1389-6 CH26 555 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 556 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

FIG. 26.19
Active Server Pages
place Web site
maintenance back into
the hands of the
content providers.

The first layer of control is established through TCP/IP packet filtering at the Web server
level. Because you want to enable the database administration site to be maintained from de-
fined IP packets from within the DataWidgets domain and from any external IP addresses, you
will not restrict authorization at the IP level.

However, the second layer of control is to regulate the permission of the site at the file level.
Typically, when you set up the Internet Information Server, a user name IUSR_machinename is
created to represent the Internet guest account. When an anonymous user requests pages
from your public site, NT uses this security setting to verify access for the requested files.
Therefore, to restrict access to only authorized users or groups, use the NT file permission to
administer create and delete user rights. Use the Security tab, as illustrated in Figure 26.20, to
remove Internet Guest Account and add the proper group rights for the DataWidget HTML
administration site.

Now that the Internet guest account has been removed, when the Internet Information server
tries to access this file, the user will be prompted for proper identification. This technique was
applied to the Amazing DataWidget WareHouse administration Web site to only enable mem-
bers of the DataWidgets Operator’s group to have access to this site. Now, when any page on
the administration Web site is requested, the user is prompted with an authorization dialog
box, as shown in Figure 26.21.

32.1389-6 CH26 556 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 From Here… 557

FIG. 26.20
Remove the Internet
Guest Account to
eliminate anonymous
access to your site.

FIG. 26.21
With anonymous
access permission
removed, the Amazing
DataWidgets
WareHouse administra-
tion site is only
accessible to defined
users.

Keep in mind that the username and password are being passed in non-encrypted form from Part
the Web browser to the Web server. A potential hacker could intercept the packet and retrieve
the NT username and password. To prevent this situation from occurring, only transmit V
usernames and passwords over a Secure Socket Layer or other encrypted channels. Use the Ch
HTTPS server environment variables to ensure information is being transmitted over a secure
channel. For more information the server variables, please refer to Chapter 15, “Retrieving
26
User Information: The Request Object.”

From Here…
In this chapter, you learned how the ActiveX Data Object is used to add database-driven infor-
mation on the Amazing DataWidgets WareHouse site. The ADO’s Connection object, the
Command object, and RecordSet object all work together to provide a robust, light-weight, fully
programmable interface to any ODBC-compliant database. In addition to coding the ADO
objects by hand, Visual InterDev’s design-time Data Command control was used to demon-
strate how to use the design-time control to generate Active Server ADO script. Finally, setting
access permissions to the Data Widgets administration site was also demonstrated by control-
ling the user permission at the file level.

32.1389-6 CH26 557 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 558 Chapter 26 Putting It All Together: Creating a Dynamic Database Application

For related information, please see the following chapters:

■ Chapter 28, “Working with the Microsoft Transaction Server,” illustrates the use of the
Transaction Server to manage components within your ASP applications.
■ Chapter 30, “Application Development for the Corporate Intranet/Extranet,” demon-
strates how to properly use your network topology to create intranet and extranet
applications.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
demonstrates how to use the various features of the Internet Information Server to help
manage your sites and applications.

■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,”
demonstrates how to build ActiveX components to create an n-tier distributed application
for the Amazing DataWidget WareHouse Catalog.

32.1389-6 CH26 558 10/26/98, 10:04 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/14/97 CH26 lp#3
 P A R T

VI
Active Server Pages and the Enterprise
Solution
27 Visual InterDev: The Active Server Pages Integrated
Development Environment 561

28 Working with the Microsoft Transaction Server 585

29 Working with the Message Queue Server 613

30 Application Development for the Corporate

Intranet/Extranet 639

31 Using Internet Information Server 4.0 to Manage Your

Web Applications 653

32 Putting It All Together: Creating an N-Tier Online

Catalog Application 673

33.1389-6 Part VI 559 10/26/98, 10:05 AM

Brands SEU Active Server Pages #1389-6 Part VI SRing 10/28/97 lp#1
33.1389-6 Part VI 560 10/26/98, 10:05 AM

Brands SEU Active Server Pages #1389-6 Part VI SRing 10/28/97 lp#1
 27
561

C H A P T E R

Using Visual InterDev

Visual InterDev: Visual InterDev provides an inte-
grated development environment

The Active Server for creating, developing, managing,

and deploying Active Server Page-
based applications.

Pages Integrated Interfacing with the Databases

Learn how to use the visual data-
Development base tools integrated into Visual
InterDev to provide both design-
time and runtime connections to
Environment any ODBC database.

Increase Productivity with

Design-Time Controls and
Wizards
Expedite development time for your
Active Server Page applications by
using ActiveX design-time controls
Visual InterDev provides the first integrated development and wizards to generate ASP code.
environment for creating Active Server Page applications.
Visual InterDev provides a single development environ-
ment to help you create your Active Server Page applica-
tions and manage the content, various file types needed, Part
database connections, and source control features. This VI
integrated development environment provides the first
Ch
opportunity for Rapid Application Development (RAD) for
Active Server Page applications. This chapter presents an 27
overview of Visual InterDev, so for a more detailed discus-
sion I recommend Special Edition Using Visual InterDev
from Que. ■

34.1389-6 CH27 561 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 562 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

A Development Environment for Active Server Pages

The joys of tapping into the power and extensibility of Active Server Pages can be quickly
doused when you consider that the only Active Server Page development environment is any
ASCII-based editor. Although using an ASCII editor enables developers to create pages on a
variety of platforms, developers that are accustomed to advanced development environments
such as Visual Basic, C++, Delphi, and PowerBuilder find this unacceptable. To prevent the
powerful functionality of Active Server Pages from going to waste, Microsoft created Visual
InterDev as the Active Server Page development environment.

An Introduction to Visual InterDev

Visual InterDev is actually categorized into two components: the client- and server-side compo-
nents, as shown in Figure 27.1. The client component is further separated into four modules:
The Visual InterDev Client, the Image Composer, the Music Producer, and the Media Man-
ager. The Visual InterDev IDE is the actual development environment used to create Active
Server Pages, and it also includes the HTML layout editor, which assists in placing and layer-
ing Active controls. The Image Composer, Music Producer, and Media Manager are utilities
that help you create your own multimedia aspects to integrate with your Web sites.

FIG. 27.1
The Visual InterDev
client and server
components.

The second component of Visual InterDev is the collection of server-side components. The
server-side components consist of the Personal Web Server, Active Server Pages, and the
FrontPage extensions. The Personal Web Server enables developers to install an HTTP server
on any Windows 95 machine, which gives you the ability to create and test your ASP pages on
any machine without having to rely on having the NT Server installed. The Active Server
Pages component installs the Active Server engine, which is responsible for processing Active
Server scripts on the HTTP server. The FrontPage extensions are used to establish the com-
munication link between the requesting client development tools and the Web server. This

34.1389-6 CH27 562 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 A Development Environment for Active Server Pages 563

communication link is needed to enable both FrontPage developers and Visual InterDev devel-
opers to manage the same project files, directories, and permissions on the server over the
stateless HTTP protocol.

FIG. 27.2
The Visual InterDev
integrated development
enviroment.

Keep in mind that in most situations, content creators such as the marketing and human re-
sources departments will use a GUI-centered HTML creation tool such as FrontPage. Applica-
tion developers will create ASP scripts in a more code-centric development tool such as Visual
InterDev to add the logic needed for the ASP application. Even though these two tools are
tailored to meet the demands of two different groups, the projects created in FrontPage and
Visual InterDev are completely interchangeable, as shown in Figure 27.3. A Visual InterDev
project is a FrontPage web and, conversely, a FrontPage web is a Visual InterDev project. Part
VI
Creating Projects and Workspaces Ch
Visual InterDev uses two different motifs to manage projects. Visual InterDev and Active
Server Pages identify projects as individual applications on the Web server that reside in a
27
virtual directory. A Visual InterDev project consists of creating the virtual directories of Active
Server projects on both the Web server and the client development workstation. These Active
Server projects are the same format as, and can be accessed by, FrontPage clients. A Visual
InterDev workspace is created on the client machine to help house and manage multiple ASP
projects within one environment, as shown in Figure 27.4.

34.1389-6 CH27 563 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 564 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

FIG. 27.3
A Visual InterDev project
can access a FrontPage
web and vice versa.

FIG. 27.4
Visual InterDev can
manage multiple
projects from one
developer workspace on
the client’s machine.

This capability to have multiple projects open at the same time is not just limited to Active
Server Page or FrontPage projects; it can also accommodate Visual C++ and Visual J++ projects
in the same workspace. This multiple-project development environment enables you to work on
your Java applets and COM components without having different development environments
open. Figure 27.5 shows an Active Server project and a Java project open in Visual InterDev.

34.1389-6 CH27 564 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 A Development Environment for Active Server Pages 565

FIG. 27.5
Visual InterDev can
manage and house
various projects within
the same IDE.

To create a Visual InterDev project, select File | New from the Visual InterDev toolbar. From
this menu, most of the productivity features can be accessed for all Visual Studio projects.
Remember, Visual Studio is the collection of all the Microsoft Visual Development tools,
whereas Visual InterDev is part of the Visual Studio collection that is used to create Active
Server Page applications. The New dialog box that appears enables you to select the different
options available: New Files, New File Wizards, New Projects, New Workspaces, and New
Other Documents. Each category consists of subcatogies that enable you to drill down into
detailed items. For example, to create a new ASP Web project, select the Web Project Wizard
from the Projects tab, as shown in Figure 27.6.

The Visual InterDev Environment Part

Visual InterDev is comprised of different panes, as illustrated in Figure 27.7. Each pane has a VI
specific purpose. The Project window is responsible for representing a hierarchical navigation Ch
display of source code files, help files, and database objects. The Project window is comprised
of three views: the Info View, the Data View, and the File View. The Info View displays the 27
available help document for any Visual Studio tool installed on the desktop. The Data View
displays the live, design-time ODBC connections. The File View represents the physical file
structure of the current Web project.

34.1389-6 CH27 565 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 566 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

FIG. 27.6
Visual InterDev uses
wizards to reduce
redundant coding and
help speed develop-
ment.

FIG. 27.7
The Project window,
the Document window,
and the Output win-
dow provide different
interfaces to the Visual
Studio project.

Project window

Document window

Output window

34.1389-6 CH27 566 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 A Development Environment for Active Server Pages 567

The Document window is basically an enhanced version of Notepad. The Document window
supports color highlighting to help differentiate between standard HTML and ASP script. By
default, ASP script is highlighted in yellow.

The Project window provides a familiar Explorer organization model of the various project
items needed for the Web application. The Project window is comprised of three views: the File
View, the Data View, and the Info View. The File View, as shown in Figure 27.8, represents the
physical file structure of the current project and is the default tab when a new project is
opened. The File View uses folders to create logical storage locations to help manage the
project files.

FIG. 27.8
The File View organizes
the physical file
systems for your Visual
Studio projects.

Part
The Data View window also provides a hierarchical organization of database objects. Not only
does Visual InterDev provide database access at runtime, but the Data View window also pro- VI
vides design-time access to any heterogeneous datasource via ODBC. This interface provides Ch
direct design-time access to any ODBC data storage unit without relying on external native
database interfaces to access the data store, as shown in Figure 27.9.
27
The exciting feature of the Data View is that the listed objects are a live connection to the
datasource. This real-time connection enables you to have direct access to the structure and
actual data within the database. For example, Figure 27.10 illustrates executing a query from
an Access database and directly viewing the results.

34.1389-6 CH27 567 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 568 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

FIG. 27.9
The Data View provides
a design-time
connection to any
ODBC database.

FIG. 27.10
The Data View provides
real-time access to the
information and
structure of the data
source.

The Info View tab displays the Visual Studio documentation, as shown in Figure 27.11. The
Info View tab is the default view when you open Visual InterDev. This view is a convenient
storage location for the multiple help files that are needed when you create Internet-based
applications. It is also a warehouse for all Visual Studio help files.

34.1389-6 CH27 568 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Visual Database Tools 569

FIG. 27.11
The Info View provides
a central location for
storing reference files
for the multitude of
technologies needed
for Web-based
applications.

Visual Database Tools

Although Visual InterDev does a good job of managing and organizing the multitude of differ-
ent files needed for Web-based applications, its real power is captured in the integration of the
visual database tools. Visual InterDev provides a rich set of database management tools that
come integrated as a part of Visual Studio.

The shared database tools provide a common interface point for data access. Furthermore, they
prevent developers from installing the native database administration tools on the workstations.
This direct database interface is centered on the Data View tab of the Project window. The Data
View, as shown in Figure 27.10, provides real-time information about database objects and Part
properties. Not only is the Data View good for displaying information about the database and its
objects, but it is also integrated with the Query Designer and the Database Designer. The Query VI
Designer enables you to graphically build and test queries on the database. The Database De- Ch
signer provides a visual interface for performing sophisticated database development, manage-
ment, and administration. The combination of these tools provides a powerful developer interface
27
for managing connections to any ODBC-compliant database.

The Integrated Data View

The Data View is created in the Project window after a database connection is created for the
application. The Data View provides a visual representation of the active database objects for
the project. Notice that the visual interface is similar to the graphical user interface found in
Microsoft Access or SQL Server Enterprise Manager. Through this interface, you can visually
design and test queries; directly insert, update, or delete data (permission depending); and
display relationships between tables to any ODBC-compliant datasource.

34.1389-6 CH27 569 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 570 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

To create the Data View tab, insert a database connection into the Web project by selecting
New Database Item from the Insert menu. Visual InterDev will prompt for a datasource name.
The datasource name (DSN) provides parameters to the ODBC driver manager to establish a
connection to a database. The DSN is stored in one of two connections. If the information is
stored in a file, the DSN is a File Data Source. If the datasource is stored within the Registry,
the DSN is a Machine Data Source. The File Data Source generates a DSN-free connection,
enabling greater flexibility for developers with the same drivers installed on their machines to
access the same database. The Machine Data Source connection must be physically configured
on the calling system. Figure 27.12 shows this configuration.

FIG. 27.12
Use a File or Machine
Data Source to insert a
data connection in an
Active Server Page.

After you establish a live connection to a database, the Data View can be used to explore all
database objects in your database connection. This enables you to not only drill down through
the database structures and view the database properties, but also to view and manipulate data
directly by opening views or tables or executing queries. Right-clicking on an object exposes its
shortcut menu to enable further manipulation of an object. For example, right-clicking the
Tables folder exposes the available table options, such as Open, Design, New Trigger, New
Table, Copy, Delete, Docking View, Hide, and Properties.

The Query Designer

One of the most exciting aspects of Visual Studio and Visual InterDev is the integration of the
Visual Database tools. The Visual InterDev database tools establish a live, design-time connec-
tion to any ODBC-compliant database. The Query Designer enables you to design, execute,
and save queries using a graphical format similar to that found in Microsoft Access, as shown
in Figure 27.13. Because the database connection established is a live database connection, you
can directly edit, insert, and delete information in a database.

Furthermore, Visual InterDev enables you to directly create, edit, and modify database objects,
such as tables, triggers, and stored procedures, at design time. When you make changes to the
existing database schema, a temporary schema is created. This temporary schema enables you

34.1389-6 CH27 570 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Visual Database Tools 571

to test a new database scheme before you actually change the real database schema. Alterna-
tively, you have the option to save the changed scripts that create the new database structure.
This is particularly useful if you do not have the proper permission to change the structure of
the database.

FIG. 27.13
The Query Designer
provides drag-and-drop
query creation
techniques to any
OBDC-compliant
database.

The Query Designer consists of four panes: the Diagram pane, the Grid pane, the SQL pane,
and the Results pane. The Diagram pane is used to display the tables in your database. You can
use the Query Designer to manipulate database tables by dragging the table from the Data
View onto the Diagram pane.

After the table is placed in the Diagram pane, the table’s fields are available database field Part
names. Once the items are selected, these fields are returned from the executed query. The
Diagram pane also enables you to create relationships between tables. Following this similar VI
drag-and-drop metaphor, joins can be created by dragging the fields from one table onto the Ch
fields in another table. The Grid pane is used to display the selected or output fields in a colum-
nar format. The SQL pane displays the SQL statement in SQL format. After you run and test
27
the query, the Results pane displays the results of the executed query in a spreadsheet-like
display.

The interesting point about the Query Designer is that because these four panes are synchro-
nized, you are not limited to using one particular query creation pane to create SQL state-
ments. The synchronized panes update automatically to reflect any changes in another pane.
For example, if you build a query by selecting columns in the Diagram pane, the SQL window

34.1389-6 CH27 571 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 572 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

automatically constructs the resulting SQL statement. If the field name from the SQL string is
deleted in the SQL pane, the Diagram and Grid panes automatically remove the same field
from their panes.

The Database Designer

The third component of the Visual Database tools is the Database Designer. The Database
Designer enables you to directly create, edit, and delete SQL Server 6.5 database components
using database diagrams. The database diagrams provide a method to visually represent tables,
columns, and relations for the whole or part of a database, as illustrated in Figure 27.14.

FIG. 27.14
The Database Designer
enables SQL Server
database structures to
be directly manipulated
and modified.

This visual representation provides developers with two main database development features
without having to leave the Visual InterDev development environment. The first feature of
the database diagrams is the capability to create a visual understanding of the database
schema without having to reverse-engineer a schema using external tools such as ErWin or
S-PowerDesigner. The second and often overlooked advantage of using database diagrams is
the capability to temporarily change the structure of the database and experiment with the
altered database without modifying the actual database structure.

For example, if the Database Designer is used to modify the structure of the database, the
changes are not made to the underlying database until the database diagram is closed and
saved. This temporary database diagram enables you to develop and optimize various data-
base schemas according to your development needs. The Database Designer provides fur-
ther flexibility to change the database diagram by presenting three options to the developer:
save the changes to the underlying database, disregard the proposed changes, or save the

34.1389-6 CH27 572 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Visual Database Tools 573

Transact-SQL code. The generation of the Transact-SQL code is of particular importance in a

team development environment.

Typically, the application developers do not have the database rights to change the structure of
the database. The generation of the Transact-SQL code provides a convenient means to save
the change script and pass it along to the database administrator. Remember, the previous two
tools, the Data View and Query Designer, are designed to connect with any ODBC-compliant
databases. The Database Designer, however, is focused specifically on creating and administer-
ing a SQL Server 6.5 database.

Data Form Wizards

The Data Form Wizard reproduces the data-binding concept of attaching a database field to a
client-side object, such as a text box, multi-line edit box, or combo box. However, remember
that this data binding is not a consistent connection as in traditional applications. The Data
Form Wizard simply associates an HTML text box with a given field in a database and updates
that field only on a Submit event in the HTML. The Data Form Wizard creates a total of three
Active Server Pages that provide the add, edit, delete, and filter functionality that is needed for
most basic database applications. The Data Form Wizard accomplishes this by generating a
List page, a Form page, and a Filter page after requiring the developer to complete a series of
steps, as shown in Figure 27.15.

FIG. 27.15
The Data Form Wizard
generates ASP to
provide edit, update,
insert, and delete
functionality.

Part
VI
Ch

27

The List page displays the results of a database query in a list form on an HTML page. The
List page enables the user to drill down to the Form page to view the details of a selected
record. The Filter page enables the user to tailor his list according to customized input. The

34.1389-6 CH27 573 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 574 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

real usefulness of Visual InterDev is that it not only creates the Active Server Pages for data-
base access and application workflow, but it also enables you to modify the wizard-generated
Active Server code to further tailor and customize your application. Furthermore, the Visual
InterDev Wizard Software Developer’s Kit (SDK) can be used to design your own wizards to
meet your company- or technology-specific development needs. The Wizard SDK is available at
the Microsoft Web site at www.microsoft.com/VInterDev.

Database Design-Time Controls

In addition to providing an environment for inserting and manipulating client-side ActiveX
controls into a HTML page, Visual InterDev provides controls that are specifically tailored to
be used at design time. A design-time ActiveX control represents a code object that creates
Active Server code based on a wizard-like approach. Design-time controls are analogous to
wizards because they provide a visual interface for generating ASP code. But unlike wizards,
design-time controls do not require a sequential navigation of steps to complete the intended
task. The design-time controls are also different from the wizards because a design-time con-
trol focuses on creating code for a particular section of a Web page, whereas the wizards often
focus on creating one or more related pages, as demonstrated with the Data Form Wizard.

The three database design-time controls that ship with InterDev are the Data Command, Data
Range Header, and Data Range Footer, as shown in Figure 27.16.

FIG. 27.16
Design-time controls
create specific code
modules within an
Active Server Page.

The Data Command control writes the Active Server script that creates a RecordSet object
based on the Active Data Object (ADO) on the Microsoft Internet Information Server. The
ADO is a logical combination of the RDO and DAO object models that have been shipping with
Visual Basic for some time. The ADO object model has been specifically optimized for database
access over the Web and provides an object model developers can program against to meet the
specific needs of their application.

The Data Range Header creates an ADO RecordSet object and begins a loop method to move
through the record in the result set.

The Data Range Footer is used with the Data Range Header to complete the loop method
started by the Data Range Header control to cycle through the recordset. Visual InterDev
extends its versatility by enabling third-party vendors to develop and distribute their own
ActiveX design-time controls. Any tool that can create ActiveX controls, such as Visual Basic
5.0, Visual C++, or Borland’s Delphi, can create the design-time controls.

34.1389-6 CH27 574 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Productivity Features 575

To insert a design-time ActiveX control, right-click an ASP file and click Insert ActiveX Control.
At this point you can either insert an ActiveX control onto the Web page that will get down-
loaded to the client browser, or you can insert a design-time ActiveX control that will be used
only at design time to automatically create a server-side script. Because all the design-time
controls center on the Data Command control, choose it from the Design-Time Control tab.
The Object Editor then displays the Design window and a property sheet to set the properties
of the control, such as the Name, Data Connection, Command Type, Command Text, and SQL
Builder, as shown in Figure 27.17.

FIG. 27.17
Design-time controls
generate ASP code
using the Query
Designer interface.

After a data connection is established, you can use the Query Builder, as mentioned above, to
create and test a query. After closing the Query Builder window, the Data Command control
creates the Active Server script that creates the database connection and cycles through the
result set based on your query. The important thing to remember is that no ActiveX runtime
component is needed to display the results of the database query. The Active Server processes
the database request and sends only HTML to the client, thus eliminating any platform or
browser dependence.
Part
VI
Productivity Features Ch
The Visual Database tools provide a direct interface for manipulating database objects that was
missing from traditional application development tools. Visual InterDev provides several productiv-
27
ity features to extend its functionality for creating database-driven Web applications. Developers can
use existing wizards or components, or build their own authoring wizards and components using
tools such as Visual Basic to develop applications with unique or business-specific needs.

An Active Server script can be added to a page in one of three ways. The first way to add a script
onto the page is to simply type the Active Server script in the Document window. The second way
to add an Active Server script to a page is through the use of design-time controls. Design-time
controls are ActiveX controls that are run at design-time, instead of runtime. From a functional
aspect, design-time controls are analogous to wizards, except that design-time controls are fo-
cused on generating smaller sections of Active Server script within a page instead of complete

34.1389-6 CH27 575 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 576 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

pages. The third method to add an Active Server script to your Web application is to use a wizard.
Wizards display a series of panels that solicit information from the developer. When finished, the
wizard generates one or more complete Active Server Pages.

Visual InterDev ships with seven wizards that minimize the development time that must be
focused on setting up sites, projects, and databases. The Web Project Wizard creates and man-
ages a Visual InterDev project on a server. The Template Page Wizard creates a page based on
a template or theme and adds the page to the current project. You can create custom themes
according to your own corporate standards. A theme is the visual combination of images, lay-
out, bullets, and line treatment that gives a site a particular look and feel.

The Template Site Wizard creates a site based on the selected templates and themes. The
Department Site Wizard creates a customized site at the department level within your organi-
zation. The New Database Wizard creates and administers a SQL Server database and inserts a
data connection into your InterDev project. The Input Wizard creates Active Server Pages that
are specifically designed to enable users to insert information into your database application.
The Data Range Builder Wizard creates Active Server Pages that enable users to search, up-
date, insert, or delete records in any ODBC-compliant database (see Figure 27.18).

FIG. 27.18
The Data Range Builder
Wizard generates Active
Server code to display
records returned from a
database query.

The Script Wizard

The Script Wizard provides a visual interface for generating client- and server-side VBScript
and JavaScript. The Script Wizard examines all the objects on an HTML page and determines
the objects’ properties, methods, and events. These objects can then be tied into the HTML
object model to become an integrated part of the HTML page. For example, the text in HTML
intrinsic text boxes can be set to a specific value every time a page loads by using the
Window.OnLoad event. To activate the Script Wizard, right-click an open Active Server Page in
the Visual InterDev editor. Alternatively, selecting the Script Wizard option from the Visual
InterDev View menu also starts the Script Wizard (see Figure 27.19).

This Script Wizard is the same Script Wizard that comes with the ActiveX Control Pad and
FrontPage editors.

34.1389-6 CH27 576 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Productivity Features 577

FIG. 27.19
The Script Wizard can
be used to produce
client and server
VBScript and
JavaScript.

The Visual InterDev Editor

The Visual InterDev default editor (see Figure 27.20) provides a text-based interface to code the
application logic into your Active Server applications. This text-based editor provides an equiva-
lent to an intelligent Notepad interface, similar to the source code editor found in FrontPage. The
Visual InterDev Editor provides color-coding to help you easily separate ASP scripts from stan-
dard HTML tags. The Visual InterDev Editor is also able to read the HTML tags to provide short-
cut menus when you right-click over code items such as design-time ActiveX controls.

FIG. 27.20
The Visual InterDev
Editor provides access
to HTML and ASP script
in an ASCII-based
environment.

Part
VI
Ch

27

34.1389-6 CH27 577 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 578 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

The Visual InterDev Editor is not the only editor available to create Active Server Pages. Visual
InterDev enables you to create and edit HTML or ASP code from any editor. You can import
existing code from other locations and activate any editor of your choice by using the Open
With option, as shown in Figure 27.21.

FIG. 27.21
Visual InterDev enables
you to use the editor of
your choice to create
and edit your content.

HTML Layout Control Editor

Visual InterDev also integrates the Layout Control Editor into the Visual InterDev IDE. The
Layout Control Editor provides a drag-and-drop interface for adding ActiveX controls to a page,
and a forms-based motif for controlling properties, methods, and events of an object within the
page. Furthermore, the Layout Control Editor provides the familiar look and feel environment
that developers are used to in Visual Basic, as you can in Figure 27.22. However, it requires
either the Internet Explorer browser or an ActiveX plug-in when viewed with other browsers.

FIG. 27.22
Visual InterDev
incorporates the Layout
Control Editor for
managing ActiveX
controls on a page.

34.1389-6 CH27 578 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Productivity Features 579

Integrated File Management Visual InterDev provides excellent file management capabili-
ties by providing hooks into any source code control mechanism, such as Visual SourceSafe or
PVCS. The “check-in” and “check-out” capabilities are mandatory for team-based development
and are particularly useful in Web-based application development. Because Web development
is a file-based delivery system, tighter permissions and controls on file management are criti-
cal. This capability to roll back changes to previous file versions or prevent files from being
overwritten is especially important when you consider the new format of the Web application
development team. The new Web development team is more than the traditional project man-
agers, application developers, and network engineers, but now includes content creators,
graphic artists, and multimedia specialists.

To regulate file management, Visual InterDev uses the combination of the FrontPage Server
extensions and HTTP to create and maintain file permissions on file transfer capabilities to the
Web server. The FrontPage Server extensions are installed on the Internet Information Server
to enable Visual InterDev and FrontPage clients to create a connection to the Web server. The
FrontPage extensions are used to provide administration features needed to manage Web
projects. For example, the FrontPage Server extensions are used to track who has the most
recent working copy of a document. Visual InterDev uses HTTP to transfer information be-
tween the Visual InterDev client and the server. Visual InterDev also enables you to use SSL to
encrypt your Active Server Pages as you post changes to the Web server.

When you create a Web project in Visual InterDev, you are creating a virtual directory on the
Internet Information Server. This virtual directory represents an Active Server Page applica-
tion. Not only are you creating a directory structure on the hosting Web server, but you are
also mimicking that directory structure on your local machine. The local directory structure is
used as a temporary cache to hold the files and reduce download times. This local caching
feature is different from the way a FrontPage client interfaces with the Web server. When a
FrontPage client interacts with the Web server, the FrontPage client only loads the page into
memory on the FrontPage client.

A FrontPage project can be opened by a Visual InterDev client and a Visual InterDev project
can be opened by a FrontPage client. This multi-product editing capability is needed to provide
access to the same information between the content creators and application developers. Part
FrontPage, as shown in Figure 27.23, is targeted at the content creators and Visual InterDev is VI
targeted at application developers.
Ch
Visual InterDev provides full drag-and-drop functionality to help maintain and organize the files
on your site. You can also import or add files or entire folders into Visual InterDev projects.
27
One of the design features of Visual InterDev is the capability to house multiple project types
within the same workspace. This means your Web project can consist of Java applets and entire
Visual J++ and Visual C++ applications. This feature is a direct result of the fact that Visual
InterDev is part of Visual Studio, Microsoft’s suite of visual design tools. This capability to
house different projects within the same IDE enables you to develop your COM-based compo-
nents, Java applets, and Active Server Pages without opening three different development
environments.

34.1389-6 CH27 579 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 580 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

FIG. 27.23
FrontPage is tailored for
content creators,
whereas Visual InterDev
is targeted for
application developers.

The Link View Furthermore, once all your files have been imported to the project, use the
Link View tools to display the valid and broken links from a page. For example, Figure 27.24
illustrates the Link View of the Intel Web site.

FIG. 27.24
The Link View tools
analyze the site for
broken HTML links.

34.1389-6 CH27 580 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 Productivity Features 581

After you save the files, you can view the pages in a variety of ways. The files can be viewed
directly within the Project window as the Document window is transformed into the Internet
Explorer viewer. Alternatively, Visual InterDev enables you to browse the page from outside
Visual InterDev by using the Browse With Preview option. You can configure Visual InterDev
to launch any HTML browser to view your content.

Microsoft Image Composer

The Microsoft Image Composer provides a fully featured image editor that enables you to
create and manipulate graphics. The Image Composer, shown in Figure 27.25, provides a
straightforward approach to image design that is found in traditional imaging tools. The Image
Composer is based on a technology called sprites. Sprite technology treats each image within
an image as its own separate object that has transparency information built in. This transpar-
ency feature enables images to be easily combined and merged to form a collective picture
without having to understand the details of channels or layering found in many editing tools.

FIG. 27.25
The Image Composer
enables you to create
exciting images tailored
for Web-site designs.

Part
VI
Ch

27
The Microsoft Image Composer is loaded with a wide variety of imaging tools and utilities that
are essential for delivering high-quality images. These tools provide a full range of image ma-
nipulation capabilities that enable you to arrange images or spites with ease. Additional func-
tionality is found in the built-in art effects, which enable you to generate exciting images that
catch the user’s eye by providing features such as embossing, charcoal, stained glass, and
chrome effects. Furthermore, the Image Composer enables you to use many of the third-party
imaging plug-ins that exist for Adobe Photoshop to add extract functionality not found natively
in the Image Composer.

34.1389-6 CH27 581 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 582 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

However, the reason the Image Composer ships with Visual InterDev is that it is tailored to
meet many of the requirements needed for creating images for the Web. The Image Composer
saves the spites in JPG and GIF (interlaced and transparent) formats and displays the created
image exactly how it would be displayed on the Web page.

Microsoft Music Producer

The Microsoft Music Producer, shown in Figure 27.26, is a client tool that is shipped with
Visual InterDev. It enables you to compose your own music score for Web sites or other multi-
media interfaces. The Music Producer enables you to combine different music themes to meet
the custom nature of your Web site or presentation by enabling you to control the different
themes, keys, personalities, and instruments that make up your composition. Not only can you
change the feel of the music, but you can also mix the instrumentation as desired. This mixing
enables you to control the effects of your composition. For example, not only can you control
the volume of your composition, but you can also control the strength of each instrument
within the composition.

FIG. 27.26
The Music Producer
enables custom-created
music based on
different themes,
instruments, and mixing
capabilities.

Microsoft Media Manager

The Microsoft Media Manager provides a centralized storage and indexing location for various
media files and other data files. In addition, the Media Manager enables you to create annota-
tions for the media files. These annotations are used to help maintain comments and organize
and catalog the media. Once the Media Manager is installed, the Media Manager creates an
annotation database that stores the user-defined image properties. The Media Manager is
directly integrated into the Window Explorer to provide thumbnail images of the files and the
ability to simply right-click the media file to expose its annotations. These annotations also stay
with the image even if the image is moved or renamed.

34.1389-6 CH27 582 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 From Here… 583

From Here…
This chapter quickly pointed out many of the developer specifics of Visual InterDev. One of the
most exciting aspects of Visual InterDev is the Visual Database tools. Clearly, as applications
are transitioned to the Web, the tight integration with database development tools is essential
for minimizing development time. The Data View within Visual InterDev is primarily respon-
sible for managing the database connections. The Data View is complemented by the addition
of the Query Designer. The Query Designer provides a graphic interface to create, edit, and
test queries. Furthermore, design-time ActiveX controls and wizards help speed development
time by generating Active Server script.

For related information, please see the following chapters:

■ Chapter 28, “Working with the Microsoft Transaction Server,” illustrates the use of the
Transaction server to manage components within your ASP applications.
■ Chapter 30, “Application Development for the Corporate Intranet/Extranet,” demon-
strates how to properly use your network topology to create intranet and extranet
applications.
■ Chapter 31, “Using Internet Information Server 4.0 to Manage Your Web Applications,”
walks you through the new features of Internet Information Server 4.0.

Part
VI
Ch

27

34.1389-6 CH27 583 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 584 Chapter 27 Visual InterDev: The Active Server Pages Integrated Development Environment

34.1389-6 CH27 584 10/26/98, 10:07 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH27 LP#3

 28
585

C H A P T E R

Distributed Transaction
Working with the Processing
Distributed transaction processing

Microsoft Transaction involves spreading the logical and

business rule processing over
multiple computers. This chapter

Server shows you how this works, and how

it requires you to design your appli-
cations differently from traditional
client/server applications.

Microsoft’s Transaction Server

Microsoft recently released its own
Transaction Server for use in build-
There has been a lot of discussion in the client/server ing distributed transaction architec-
tures. I’ll dig in to how Transaction
industry over the past few years about “second genera-
tion” and “three-tier” architectures. These are two differ- Server works, and how it uses
Microsoft’s Distributed Common
ent terms for the same application infrastructure. This is
Object Model (DCOM) technology
an architecture in which the application’s processing is not
to provide a transparent platform for
divided into the traditional two parts, the front-end user building distributed applications.
interface and the back-end database, but into at least three
Active Server Pages and
Transaction Server
One of the key advantages offered
by Microsoft’s Transaction Server is
the capability to build modules
using Visual Basic (along with
Microsoft’s other development
languages, including Visual C++ and
Visual J++) that can be easily called Part
from Active Server Pages. This
enables you to distribute some of VI
the processing required for your Ch
ASP application onto other servers
within your organization. I’ll look at 28
what is involved in integrating
Active Server Pages with Transac-
tion Server.

35.1389-6 CH28 585 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 586 Chapter 28 Working with the Microsoft Transaction Server

separate and distinct parts. There are several reasons given for why this application architec-
ture is superior to the original client/server model, such as:

■ Scalability. The capability of an application architecture to scale up to hundreds of

simultaneous users.
■ Manageability. The ease with which problems can be isolated, updates distributed, and
configurations managed.
■ Transparency. The capability to switch server processing from one particular computer
to another, as dictated by scheduled or unexpected system outages.

One of the key differences between traditional, two-tier client/server models and second gen-
eration, three-tier (also known as n-tier, where n is any number greater than two) client/server
architectures is in the use of what is known as middleware. Some of the most popular kinds of
middleware are transaction monitors and object request brokers (ORB). Microsoft’s Transaction
Server is a combination of these two technologies, providing you with a flexible and powerful
middleware component that can be used to build very large-scale distributed-processing appli-
cations with minimal coding and configuration effort on your part.

N O T E A transaction monitor is used to manage database access and to control transactional

consistency for data-entry type applications. An object request broker manages services
that are distributed across one or more computers, providing applications with a single service to
connect to, which then provides the applications with access to other services in a transparent
manner. This frees the application from the constraint of having to know where within the network each
service it needs to use is located. Both of these technologies will be explained in further detail later in
the chapter. ■

N O T E The typical three-tier architecture is composed of the following three tiers:

● The presentation tier, which consists primarily of the user interface running on the
desktop computer. This tier would typically be built using languages designed for
creating user interfaces, such as Visual Basic, PowerBuilder, and Delphi.

● The business-logic tier, which consists of shared application processing logic modules
running on one or more application servers. Each of these application servers is
connected to multiple desktop computers. In the past, modules in this tier would
typically be built using Visual C++, working with either an object request broker or a
transaction monitor. Now, with tools like Microsoft’s Transaction Server, components in
this tier can easily be built with languages such as Visual Basic and Delphi.

● The data server tier, also known as the database server. There is usually one database
server serving multiple application servers. If there are multiple database servers, they
are normally spread out over a large geographical area with the common data
replicated between them. This would typically be an enterprise-scale database server
such as SQL Server, Oracle, Sybase, or Informix. ■

35.1389-6 CH28 586 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Understanding Distributed Transaction Processing 587

This chapter takes an in-depth look at Transaction Server, how it works, and how you can use
it in building applications. I’ll also discuss what’s involved in using Transaction Server with
Active Server Pages, and what considerations need to be taken into account when building
Visual Basic objects that can be integrated into Transaction Server for use with Active Server
Pages to provide a large, distributed application. In addition, I’ll look at how Transaction Server
can be integrated into a Web site to provide the capability to build an integrated enterprise-wide
application system that stretches far beyond what most Web sites currently are capable of.

Understanding Distributed Transaction Processing

Imagine that your company has several independent database systems: one for maintaining the
current inventory in the company warehouse; one for maintaining all items that have been requi-
sitioned and should be arriving on the receiving dock; and a third database for maintaining cus-
tomer orders and shipping information, such as that shown in Figure 28.1. In this company, when
an order comes in, the order is entered into the order entry system, which produces a shipping
order. This order is then taken to the warehouse, where each item in the order is removed and
taken to the shipping dock to be packaged up and sent to the customer. At the warehouse, the
employee filling the order finds that one or two items are out of stock. The employee then takes
the order to the requisition department and has the missing items back ordered for later ship-
ping. At each of these points, the individual systems have to be updated so that they are up to
date. Even the shipping dock system has to be updated in order to produce an accurate shipping
bill-of-lading, which correctly reflects the back-ordered items.

FIG. 28.1
A typical catalog
company has systems Order
to track current Database
inventory, requisitioned
or back-ordered items, Order Entry
Shipping
and shipping orders.

Inventory Requisition
Database Database
Part
Warehouse Receiving VI
Inventory
Ch

This seems like a lot of wasted and duplicated effort. Wouldn’t it make a lot more sense to 28
connect all of these systems together so that they could all exchange the necessary informa-
tion to perform most of these tasks themselves? If you put a network in place and connect all of
these systems to the network, as shown in Figure 28.2, the potential for these systems to ex-
change the appropriate information can be realized.

35.1389-6 CH28 587 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 588 Chapter 28 Working with the Microsoft Transaction Server

FIG. 28.2
A network could be
used to connect all of Order
the company systems, Database
so that they could
exchange information Order Entry
Shipping
with each other.
Network

Inventory Requisition
Database Database

Warehouse Receiving
Inventory

Unfortunately, as anyone who has attempted to connect separate systems together in this way
knows, once you have all the systems on a network, the work is just beginning. The work of
enabling all of these systems to work together is part of what middleware is all about.

Another function of middleware is to reduce the processing load on the desktop, enabling
business-rule logic, which is likely to be updated and revised on a frequent basis, to be located
on central servers. This configuration makes the process of updating business-logic modules a
simple task of replacing the affected modules on a few servers, instead of updating the soft-
ware sitting on hundreds of desktops.

A third function of middleware is to place data-processing logic close to the database server in
order to minimize network traffic between the database server and the client applications run-
ning on the desktop. Most of the same network traffic is still created on the network segment
between the database server and the data processing modules (unless they are located on the
same computer). This still isolates the network traffic and congestion to that specific network
segment, and keeps most of it off the rest of the corporate network.

Typically, client/server types of applications don’t scale well to large numbers of users using a
two-tier architecture. Even if an application is using only a single database, if the application is
being used by more than a couple dozen users at once, it is advisable to use some sort of
middleware and a three-tier architecture to maintain reasonable application performance. In
the past, this usually required extensive redesign on most two-tier applications. With the latest
set of tools, the process of converting a two-tier application to a three-tier application is getting
much easier. However, there are still situations where the two-tier application was originally
designed such that extensive modifications are required to make the transition.

35.1389-6 CH28 588 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Understanding Distributed Transaction Processing 589

Transaction Monitors
The original idea behind the client/server computing model was to split the application process-
ing between two computers. This split was normally made so that the database processing was all
performed on the server, while all application processing was performed on the client, as shown
in Figure 28.3. This model worked reasonably well except for the fact that it was fairly easy to
overload the client system with data, if the application allowed the database to return more data
than the client computer could handle (remember that this was when the standard desktop com-
puter was a 386 with 4MG of RAM, and a 486 was a high-end workstation).

FIG. 28.3
The original idea
behind client/server
computing was to split Database
the application Server
Client
processing between the
client and the database
server.

As the number of active users of these early client/server systems increased, the workload on
the database would increase. A normal application would maintain an open connection to the
database server for the entire time the application was running (see Figure 28.4). This re-
quired the database server to have additional processing power in order to service all of those
open connections, even if they were sitting idle. This also increased the cost associated with
the database itself, as most traditional enterprise database license prices are based on the num-
ber of users that can be connected to the database simultaneously.

N O T E The practice of maintaining open database connections was developed for two primary
reasons. First, it made the development of database applications easier if the database
connections were maintained throughout the life of the application, instead of constantly having to
close and reopen the connection (not to mention having to keep track of whether the connection is
open or closed). The second reason was to avoid the overhead involved in opening and closing the
database connection. This overhead can add a couple of seconds (on a congested network, or to a
busy database server) to the time the application requires to perform certain actions. Users are
typically willing to wait through delays like this during the startup and shutdown of applications, but
not during the midst of the application. ■

As the number of databases that an application needs to interact with increases, the number of Part
open connections that the application must maintain also increases. This adds to the process-
ing load on the client computer, as well as the network management and configuration. The
VI
total number of connections that must be maintained can be calculated as the number of clients Ch
times the number of database servers to which the client connects, as shown in Figure 28.5. 28

35.1389-6 CH28 589 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 590 Chapter 28 Working with the Microsoft Transaction Server

FIG. 28.4
As the number of users
increase, the number of
active connections that
must be serviced by the Client
database increases.

Client Database
Server

Client

Client

FIG. 28.5
The total number of
Database
open connections that Server
must be maintained can
be calculated as the
Client
number of clients times
the number of servers.

Database
Server

Client

Database
Server

Client

Database
Server

Client

35.1389-6 CH28 590 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Understanding Distributed Transaction Processing 591

This is a lot of configuration information that has to be maintained for each client computer. If
one of the databases that an application uses has to be taken offline, and the backup database
brought online, the configuration of each and every client that connects to the database has to
be updated to reflect the new database server (with care taken not to update the wrong data-
base server information).

This is one of the primary problems that transaction monitors were designed to solve. A trans-
action monitor goes between the client systems and the database servers, as in you can see in
Figure 28.6. Each of the clients maintains a single connection to the transaction monitor in-
stead of the database servers. Likewise, the transaction monitor maintains connections to each
of the database servers. This enables the transaction monitor to act as a traffic cop, passing
each database query or update to the appropriate database, and to maintain only as many open
database connections as are currently required, which enables the database servers to run
more efficiently.

FIG. 28.6
A transaction monitor
Database
reduces the number of Server
open connections to
the number of clients
Client
plus the number of
database servers.
Database
Server

Client

Transaction
Monitor Database
Server

Client

Database
Server

Client

Object Request Brokers Part

Object request brokers (ORBs) fall into a different category of middleware than transaction VI
monitors. ORBs provide location transparency to application modules and services, which Ch
means that when an application needs to interact with a server process, the application does
not need to know where that specific server process is located on the network. The only pro- 28
cess that the application knows the location of is the ORB client stub located on the same ma-
chine as the application.

35.1389-6 CH28 591 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 592 Chapter 28 Working with the Microsoft Transaction Server

The ORB knows where all server processes are running, or on which machines the server
processes may be run. If a particular server process is not running when an application re-
quests access to it, the ORB will start the server process on one of the machines for which the
process is configured to run. The client application does not know on which server the re-
quested process is running or if the server process is even running. The ORB keeps track of all
server processes for the client applications. The ORB may even perform load balancing be-
tween two or more servers running the same process, so that each of the servers is servicing
around the same number of requests.

Returning to the catalog sales company example, it is reasonable to expect that the order entry
system would make use of various services that could be located on a series of servers on the
company network. One of these services could be a sales tax engine, which calculates the tax
for each of the customer orders. It would make sense to keep this set of calculations on a
server, as tax laws have a tendency to change. By keeping this processing module on the
server, it will be a lot easier to update in this one location every time the tax laws change, as
opposed to having to update every order entry workstation in the company.

Another service that each of the order entry systems would be likely to take advantage of is a
credit check application. By having all the credit authorization requests go through a single
server, it would be easy for that one server to maintain an open connection to the credit clear-
inghouse, as compared to outfitting each workstation with a modem and the software to call up
the credit authority (not to mention all of the additional phone lines this would require).

A third service that each of the order entry systems could use might be the business rules
engine. This would be the system that calculates shipping costs based on the quantity or
weight of the ordered items or enforces minimum purchase rules. By keeping this module on
one or two servers, you would once again be able to easily update the module as the powers-
that-be within the company change the business rules that this module has to enforce.

Considering that the functions provided by these server modules are critical to the core busi-
ness of our catalog sales company, it’s important to make sure that these modules are always
available for the order entry systems. In order to make sure that these systems have high
availability, they are probably loaded onto more than one system. If the order entry application
made direct accesses to these services, each copy of the application would have to know which
machines each of these services are running on at all times. By using an ORB, the individual
copies of the order entry application don’t know and don’t care what servers any of these ser-
vices are running on; the ORB takes care of making the connections and passing the results
back to the order entry application, as seen in Figure 28.7.

Introducing Microsoft Transaction Server

Microsoft’s Transaction Server is somewhat of a cross between a transaction monitor and an
ORB, although it tends to lean more toward the ORB set of functionality. If you are running
an application on a Windows NT system that has Transaction Ser ver installed, and you are
using ODBC to access a database, Transaction Server transparently inserts itself between

35.1389-6 CH28 592 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Introducing Microsoft Transaction Server 593

your application and the database to manage that connection (as well as all of the other open
connections to the same database). Transaction Ser ver also enables application functionality
to be built as a series of ActiveX DLLs and distributed across a network to run on any num-
ber of servers running on the corporate network.

FIG. 28.7
The ORB relieves the
Tax Engine
client application from
having to know which
server computer each
of the services is Order Request Credit
running on. Broker Verification
Client Application
Engine

Business
Rule
Engine

Managing Database Connections

When an application is using the ODBC interface to access a database, Transaction Server takes
control of the database connection to provide more consistent access, quicker connection, and
transaction control. By placing itself between the application and the database, Transaction
Server is able to open its own connection to the database, thereby providing the application with
a connection to Transaction Server instead of the database. This enables Transaction Server to
limit the actual number of database connections to only as many as are necessary to service all
the application requests, as seen in Figure 28.8. This relieves the work of maintaining all those
connections from the database, enabling the database to perform better and be more responsive.

FIG. 28.8
By inserting itself
between the applica-
tions and the database,
Transaction Server is Client Application
able to limit the
number of active
connections that the Part
database has to Transaction Database
service. Server VI
Client Application Ch

28

Client Application

35.1389-6 CH28 593 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 594 Chapter 28 Working with the Microsoft Transaction Server

N O T E Transaction Server performs this resource allocation through the Microsoft Distributed
Transaction Coordinator (DTC). The DTC performs resource allocation by communicating
with a resource manager (such as a database server) through the OLE-Transaction protocol. Not all
databases support this protocol, and many require the use of an additional transaction monitor that
supports the X/Open XA protocol, which can interact with the DTC through the OLE-Transaction/XA
translator. ■

N O T E Transaction Server is required to be running on each server on which there are components
that it needs to run, or that call components that run within Transaction Server. This means
that a minimal installation of Transaction Server needs to be on every client computer, even if all of the
components being called reside on other computers within the organization. ■

When an application closes its connection to the database, Transaction Server maintains the
open database connection so that the connection can be reused either by the application that
closed the connection, or by another application (or client) that needs the same connection to
the database. This enables application modules to be written in such a way that they only main-
tain open connections to a database for those periods of time that the application really needs to
have the connection open. The performance penalty for closing and re-opening a database
connection is removed, making it more attractive to write applications that release the associ-
ated resources while they are not needed.

N O T E Transaction packages are configurable for how long the components within the package are
kept open and maintained in memory. The default length of time is three minutes, but this
can be set to any number of minutes, or set to keep the components open constantly. When Transac-
tion Server closes the open components, because they have been idle for the specified time, it also
releases any database connections associated with those components by letting the resource manager
know that the connection is going to an inactive state and may not be needed for awhile.

The setting controlling how long Transaction Server maintains components in memory within a certain
package should be based on the typical usage patterns of the components in the package. For
instance, if the components in a package are called very frequently, leaving the setting at the default
three minutes may be appropriate. However, if the typical usage of an application will have the
components sitting idle for five or six minutes between calls, then configuring the package to keep the
components active for 10 minutes may be more appropriate. ■

Resource Management
The management and allocation of shared resources such as database connections is performed
through the use of resource managers and resource distributors. The resource managers are usually
part of the database management system, and the resource distributors are built into Transaction
Server. These two systems interact through the OLE Transaction protocol.

35.1389-6 CH28 594 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Introducing Microsoft Transaction Server 595

Managing Distributed Objects

Transaction Server provides a facility for building distributed applications by enabling you to
build functionality into a series of ActiveX server DLLs and distribute them across your net-
work. Transaction Server keeps track of where each of these DLLs is located, and performs all
the communications between them and your application. This enables you to move your func-
tionality modules to the most suitable computer on your network based on the processing load
that each module will be requiring in order to service all the requests from applications need-
ing the services of the module. You can even double up and place the same module on multiple
computers and enable Transaction Server to load balance between the copies. This provides
you with the flexibility to take a complex processing module that might take several minutes to
perform its tasks and place it on several computers. This way, if the copy on one computer was
sitting idle while the copy on another computer was servicing two or three requests, if a new
request came in, Transaction Server could direct the request to the idle copy of the module
instead of the copy that was already busy.

Transaction Server also provides you the capability to easily mix and match modules of func-
tionality that are built in a number of different programming languages. Any language that can
be used to build ActiveX server DLLs can be used to build modules to be used with Transac-
tion Server. This includes not just Microsoft’s Visual Basic, Visual C++, and Visual J++, but also
Borland’s Delphi, Symantec’s Café, and Sybase’s PowerBuilder. This enables you to pull to-
gether functional modules together into a large distributed application, regardless of what
language was used to build the individual modules, as shown in Figure 28.9. This capability of
Transaction Server enables you to build an extensive application using “best-of-breed” modules
and components.

FIG. 28.9
Using Transaction
Visual Basic Visual C++
Server, you can use ActiveX DLL ActiveX DLL
functional modules that
were built using many
Client Application
different languages in a
single application.

Transaction Database
Server

Part
PowerBuilder Visual J++ Delphi VI
ActiveX DLL ActiveX DLL ActiveX DLL
Ch

28

35.1389-6 CH28 595 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 596 Chapter 28 Working with the Microsoft Transaction Server

Transaction Coordination
One of the many beneficial features of Transaction Server is the way it provides coordinated
transaction control through many objects and over multiple databases. Transaction Server
accomplishes this by using the Microsoft Distributed Transaction Coordinator (DTC). The
DTC was first released as part of SQL Server 6.5, and is included with Transaction Server. It
provides a low-level infrastructure for distributed transactions, controlling and guaranteeing
the outcome (either commit or rollback) across multiple databases and database connections.
The DTC uses a two-phase commit protocol to ensure the outcome of these transactions.

An example of how this works is if you were running a bank and had two separate databases
for savings and checking accounts. With most transactions, only one database would be in-
volved, as most customers place money into or take money out of either their checking or
savings account. However, whenever a customer transferred money from his checking account
to his savings account, or vice versa, then both databases would need to be involved. This
transaction would require debiting the amount of money transferred from the account on one
database, and crediting the same amount on the other database. If either of these two actions
fails, for whatever reason, the entire transaction needs to fail, rolling back both databases to
their pre-transaction state. This is the task that the DTC handles. The DTC controls the trans-
actions for both databases, making sure that either both databases commit the transaction, or
both databases rollback the transaction. The DTC does not enable one of the two databases to
commit the transaction while the other is rolling back the same transaction, as this would leave
the databases in an inconsistent state (not to mention that it would likely result in a very happy,
or a very upset customer). This coordination of the transactions of two or more databases is
called two-phase commit.

Two-Phase Commit
A two-phase commit is where a data change (insert, update, or delete) to two or more databases
absolutely has to be successful in all or unsuccessful in all. If the situation dictates that the changes
to the data cannot be committed in one of the databases without being committed in the others,
then this is the situation where two-phase commit is necessary.

N O T E The Microsoft Distributed Transaction Coordinator provides the flexibility to extend

transactions to non-database resources. The Transaction Server SDK provides the tools and
documentation to build resource managers for any resource you might want to be part of a transaction.
For instance, if you need to perform certain manipulations on a file, or send an e-mail message as part
of the transaction, you could build your own resource manager using Visual C++ and the Transaction
Server SDK, which could be used to control and coordinate your non-database tasks as part of the
transactions being controlled by Transaction Server. This would enable you to commit or roll back your
file manipulations, and send or not send the e-mail message based on the success of the
transaction. ■

35.1389-6 CH28 596 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 597

Integrating Active Server Pages with Transaction

Server
Integrating Transaction Server into an Active Server Page application is very similar to inte-
grating custom components into an Active Server Page application, with a few differences.
First, any components that you will be calling through Transaction Server must be registered.
Next, the components are instantiated slightly differently from custom components. From this
point on, the differences are hard to spot.

There are some differences in how custom components need to be built in order to work
within Transaction Server, but I’ll get to that later in this chapter. First, I’ll look at using Trans-
action Server components from the perspective of Active Server Pages.

I’ll cover how to integrate Active Server Pages with Transaction Server by using a simple bank
account maintenance application included with Transaction Server. This example enables the
user to credit or debit small amounts to either of two accounts, as well as transfer funds be-
tween the two accounts. This example will require that you adapt the Visual Basic user inter-
face provided as the user interface for this sample application to work as an Active Server Page
application. You’ll also have to make some minor changes to the Visual Basic components so
that they work with Active Server Pages.

N O T E This example is limited to using two specific account numbers, 1 and 2, and small
amounts of money by some of the constraints coded into the sample application. The
application only performs updates to the database, except if the account table does not yet exist in the
database. If the table does not exist, then the sample application creates the table and inserts two
records for accounts 1 and 2. From this point on, all SQL statements issued to the database are either
selects or updates. For the sample application to enable other account numbers, you will need to
modify the sample code to recognize that the account specified doesn’t exist in the database, and to
insert a new record for the new account.

Because we are using a sample application that is provided with Transaction Server, the code for the
Transaction Server components used in this chapter is not included with the source code on the CD in
the back of the book. However, any modifications necessary to get these components to work with
Active Server Pages are documented in this chapter so that you should be able to make the appropri-
ate changes yourself with a copy of Visual Basic 5. ■

Registering Components with Transaction Server Part

Before you can register a custom component with Transaction Server, the component has to be VI
built as an in-process, ActiveX server DLL. After you have built your component as an ActiveX
Ch
DLL, you need to register it with Transaction Server before it can be used. You do this through
the Transaction Server Explorer. When you have started up the Transaction Server Explorer, 28
you need to make sure that the Distributed Transaction Coordinator (DTC) is running. You

35.1389-6 CH28 597 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 598 Chapter 28 Working with the Microsoft Transaction Server

can tell if the DTC is running by looking at the color of the screen in the computer icon for
your computer. When the DTC is running, the screen on the computer icon is green, and when
the DTC is not running, the screen on the computer icon is black. If the DTC is not running,
click the computer icon for your computer, and then select Tools | MS DTC | Start.

N O T E The DTC can be easily configured to start automatically when the NT system boots up. The
Transaction Server installation routine will default to configuring the DTC for starting
automatically. However, if you are installing Transaction Server on a development workstation, you might
want to configure the DTC for manual starting if you need to have control over whether the DTC is
running. ■

Creating Packages Before you can start registering components in Transaction Server, you
must have a package into which you are going to install the components. Packages are logical
groupings of objects that are generally used as a unit. As a general rule, you want to create one
package for every set of applications that will be making use of Transaction Server. A package
is an organizational grouping of components within Transaction Server. These are typically a
set of components that are to be used in a certain suite of applications (or a single application).
Because security and how long components are kept in memory once they become idle are
configured on a package level, these aspects will need to be taken into account when planning
packages for use with Transaction Server. You can create a package by following these steps:

1. Select the Packages Installed folder.

2. Choose File | New from the main menu.
3. On the first screen of the Package Wizard, choose Create An Empty Package (see
Figure 28.10).

FIG. 28.10
To register components
that you have built, you
need to create an empty
package into which the
components will be
installed.

4. Type a name for the package, as shown in Figure 28.11. For the catalog sales company
package, call it Bank. Click the Next button.

35.1389-6 CH28 598 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 599

FIG. 28.11
Provide the package
with a name that
reflects the functional-
ity, or family of
applications, that the
components in the
package will be
providing.

5. If the objects in the package need to run under a specific login account, select the This
User radio button and provide the user name and password. Otherwise, leave the default
radio button selected, as shown in Figure 28.12. The default choice runs all of the objects
in the package under the account of the user using the application that call these
objects (this could effect the availability of resources for the process components,
depending on how the access privileges are configured in the system security).

N O T E A package can be configured so that all of the components within it run using a specific
user account. This will enable you to configure the security settings of resources needed by
the components so that the resources will always be available. The default setting, however, is for the
components in the package to run using the user account of the application that calls them. This
means that if the calling application is running under a user account that has limited access to
resources, some of the components called by the user may not have full access to any resources
needed by the components. ■

6. Click the Finish button to complete the process.

FIG. 28.12
If you need the
components in the
package to execute as
a specific user login for
resource access Part
purposes, you need to
specify the user login VI
and password. Ch

28

35.1389-6 CH28 599 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 600 Chapter 28 Working with the Microsoft Transaction Server

Installing Components Once you have a package, you can install components into it. This is
where you register the ActiveX DLLs that you have and will be creating. You can register your
components by following these steps:

N O T E In the following section, I use the terms “install” and “register” interchangeably. Installing
components into Transaction Server and registering components with Transaction Server
are two ways of referring to the same process. ■

1. Select the Components folder in the package into which you want to install the compo-
nents (in this case, it’s the Bank package that we just created), as shown in Figure 28.13.
Then select File | New from the menu.

FIG. 28.13
Select the Components
folder in the package
that you have created to
install components into
the package.

2. Click the Install New Component(s) button, as seen in Figure 28.14.

FIG. 28.14
If you are installing
components that you
have built, you need to
select the Install new
components option.

3. Click the Add Files button, and then select the ActiveX DLL that you want to register, as
shown in Figure 28.15.

35.1389-6 CH28 600 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 601

FIG. 28.15
You need to select the
DLL containing the
components that you
are installing.

4. When you return to the Install Components screen, the upper list box should show the
DLL that you are installing, and the lower box should show all the classes within the DLL
(see Figure 28.16).

FIG. 28.16
The Install Components
dialog displays all the
components found in
the specified DLL.

5. Click the Finish button, and the components will be installed in Transaction Server, as
shown in Figure 28.17.
6. Select the components you just installed, one at a time, and right-click the mouse. Select Part
Properties from the pop-up menu. On the Component Properties editor, select the
Transaction tab, and select the transaction attribute desired for the currently selected
VI
component, as shown in Figure 28.18. Ch

I’ll take a closer look at the Banking example included with Transaction Server later in this 28
chapter, as you begin to build the functionality you need for your ASP application.

35.1389-6 CH28 601 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 602 Chapter 28 Working with the Microsoft Transaction Server

FIG. 28.17
After you install the
components, they will
show up in the
Transaction Server
Explorer.

FIG. 28.18
You need to open the
Properties dialog for the
installed component in
order to specify the
transaction attribute
setting.

N O T E Whenever you recompile any ActiveX DLL built in Visual Basic, you need to refresh the
component information in Transaction Server. This can be easily done by deleting the
component from the Transaction Server Explorer, and then reinstalling the component by following
the same steps as you followed to install the component originally. Another way of refreshing the
component information in Transaction Server is to choose Tools|Refresh All Components. The reason
that the component has to be re-registered with Transaction Server is because Visual Basic generates a
new CLSID (globally unique Class ID) for each component in the DLL every time the DLL is compiled.
This is not the case for most other development languages. ■

CAUTION
In working with Transaction Server 1.0, the Refresh All Components menu option often scrambled the
component names in the package that I was working with. I would end up with what would look like two or
three copies of the same component in the package, and some components would appear to be missing. I
did not notice any problems when attempting to run the applications that used these components, but I
usually ended up deleting all the components from the package and reinstalling them.

35.1389-6 CH28 602 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 603

In version 1.1, the refresh does work better, but Visual Basic components must still be dropped and
reregistered. However, once the components have been recompiled on the same system with Transaction
Server, attempting to delete the components from Transaction Server without refreshing them first will cause
you to be subjected to several confusing error messages, all due in part to the Class ID of the components
having changed with the recompile. The class is changing confuses Transaction Server, and the refresh helps
unconfuse it.

Creating Transaction Server Components

Creating Transaction Server components for use within an Active Server Page is similar to
creating any other custom component for use within an Active Server Page. The biggest differ-
ence is that you do not use the Server object when you create your objects. You still have to
call the CreateObject() function, but not as a member of the Server object. For instance, you
can create an instance of the Bank.Account component from the sample application by using
the following VBScript:
Dim objMTX ‘The Transaction Server Component

‘Create the Transaction Server Component

Set objMTX = CreateObject(“Bank.Account”)
If objMTX is nothing Then
‘Perform error handling here
Else
‘Call component methods here
End if

N O T E The code above contradicts all documentation from Microsoft on how to instantiate
Transaction Server components. According to all Microsoft documentation, the third line of
code should read:
Set objMTX = Server.CreateObject(“Bank.Account”)

However, I have not been able to successfully instantiate any Transaction Server components using the
Server object. You do need to be aware of this situation, as Microsoft could correct this functionality
in new versions of either ASP or Transaction Server, making the documented method for creating
Transaction Server objects in an ASP application work correctly. ■

Once you have successfully created the component, you will be able to see the component
begin spinning the Transaction Server Explorer, as you can in Figure 28.19. After you access a
component once or twice, the component may be activated and deactivated so quickly that the Part
Transaction Server Explorer doesn’t have time to update its display. As a result, you may not VI
see the components spin, even though they are active and working correctly.
Ch

28

35.1389-6 CH28 603 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 604 Chapter 28 Working with the Microsoft Transaction Server

FIG. 28.19
The components that
are visible in the
Transaction Server
Explorer spin when they
are active.

Spinning component

Calling Component Methods

After you create an instance of a Transaction Server component, you can call the component’s
methods much like you would call the methods of any other component. By referencing the
component methods via the component itself, you can call any of the methods that have been
exposed on the component that meet the requirements for use with Active Server Pages. For
instance, you can call the Post method of the Bank.Account component like this:
lRetnVal = objMTX.Post(lPrimeAcct, (lAmount * Multiplier), RtnMsg)

Something to keep in mind when designing your Active Server Page application is that well-
designed Transaction Server components do not maintain any state information. Because of
this, Transaction Server component methods should require that all information the method
needs must be passed as parameters to the method. There should be nothing for the compo-
nent to maintain between method calls. In a sense, working with Transaction Server compo-
nents is much like building a Web application, where the server is not maintaining any state
information between calls.

Returning Data from Components

The best way to return information from Transaction Server components is through the use of
parameters to the component method. The standard return value of the method will be a suc-
cess code to inform you whether the method was successful or not. This means that all data
that needs to be returned to the calling routine needs to be returned in parameters that were
passed to the method.

For instance, if you needed to call a component method called DoThis and you need to be re-
turned a textual message to display for the user, your calling code would look something like
this:
Dim lstrMsg As String

‘Call the DoThis method of the object

obj.DoThis(lstrMsg)
‘Display the message returned by the component
MsgBox(lstrMsg)

35.1389-6 CH28 604 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 605

Meanwhile, the component’s code for this same method would look like this:
Public Sub DoThis(astrMsg As String)
‘Set the message to the string passed in as a parameter
astrMsg = “This is what I’m doing”
End Sub

You can get your first experience integrating Active Server Pages with Transaction Server
components by using the Bank Account Maintenance application that is provided with Transac-
tion Server. This example has a simple Visual Basic front end, as shown in Figure 28.20. This
simple application provides the user with the ability to specify:

■ An account number
■ The amount to credit or debit
■ A second account number (if performing a transfer)
■ The transaction type
■ The number of iterations to perform (we’ll be leaving this option out of our ASP version)
■ Which component to call (we’ll only provide the first two of these)
■ Which language version to call: Visual Basic, Visual C++, or Visual J++ (we’ll limit our
ASP version to calling only the Visual Basic version of the components)

N O T E The sample application has three different versions of the component code, developed in
Visual Basic, Visual C++, and Visual J++. In the user interface for the sample application,
the user has a choice of which version to call. For our purposes here, we’ll only use the Visual Basic
version, although you can easily make the necessary modifications yourself to include all three
versions in the ASP adaptation of this sample application. ■

This application also displays the response time statistics and the result of each transaction for
the user.

FIG. 28.20
The Visual Basic
application that calls
the Bank Account
Maintenance compo-
nents will be fairly easy
to implement as an
Active Server Page
application. Part
VI
Ch
The Account Components 28
The Bank Account Maintenance application that is provided as an example with Transaction
Server consists of five classes. These classes, and their functionality, are listed in Table 28.1.
They can be used to credit and debit either of two accounts, or to transfer funds from one

35.1389-6 CH28 605 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 606 Chapter 28 Working with the Microsoft Transaction Server

account to the other. Along with this functionality, these classes also create the required tables
in SQL Server, if they don’t already exist, generate sequential receipt numbers, and maintain
blocks of receipt numbers in the database.

Table 28.1 Bank Account Example Classes

Class Description
Bank.Account This class performs the credit and debit actions on a specified
account, returning the resulting balance. It may be called directly
to perform simple credits and debits, or be called through the
Bank.MoveMoney class to perform more sophisticated actions.

Bank.MoveMoney This class acts somewhat like a traffic cop, calling all the other
classes to perform complex actions. This class calls the
Bank.Account class to perform credits and debits on a single
account by making two calls to transfer funds between two
accounts. This class also calls the Bank.GetReceipt class to
generate a receipt number that is returned to the calling applica-
tion.
Bank.GetReceipt This class generates sequential receipt numbers using the shared
resource facilities of Transaction Server. Whenever a block of
receipt numbers has been used up, it calls the
Bank.UpdateReceipt class to allocate the next block of receipt
numbers.
Bank.UpdateReceipt This class maintains the starting number of the next block of
receipt numbers in the database. By only updating the database
once for each block of receipt numbers generated, the number of
database transactions is minimized, thereby improving perfor-
mance.
Bank.CreateTable This class is called by the Bank.Account and Bank.UpdateReceipt
classes to create the necessary tables in SQL Server if the tables
don’t already exist. This class also inserts starting records for
accounts 1 and 2. No records are inserted for any other account
numbers, thus limiting the functionality of this example to those
two accounts.

Calling the Account Components from ASP

Before you can call the Bank Account Maintenance components through Transaction Server,
you need to re-create the front-end application as an HTML form. A simple version of this form
can be seen in Figure 28.21 (the HTML document is located on the CD accompanying this
book). This form contains the field names specified in Table 28.2.

35.1389-6 CH28 606 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 607

FIG. 28.21
You can easily build an
HTML form that closely
matches the Visual
Basic user interface in
functionality.

Table 28.2 The HTML Form Element Names

Element Prompt Element Name
Account Number PrimeAcct

Amount Amount

Transfer to Account SecondAcct

Transaction Type TransType

Component Component

The transaction type radio buttons have the following values:

■ Credit
■ Debit
■ Transfer

and the component radio buttons have these values:

■ Account Part

■ MoveMoney VI
Once you have converted the Visual Basic code to VBScript, handling the submitted data ele- Ch
ments and leaving out the code dealing with the unnecessary elements that we have eliminated 28
from our ASP version, you have the code listed in Listing 28.1.

35.1389-6 CH28 607 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 608 Chapter 28 Working with the Microsoft Transaction Server

Listing 28.1 ACCTRUN.ASP—Capturing the submitted data elements in

VBScript, we can create and call the Bank Account components through
Transaction Server.
Dim ProgID ‘Transaction Server Component to call
Dim RtnMsg ‘A Return Message to be displayed to the user
Dim TransType ‘The Transaction Type to run
Dim Multiplier ‘A multiplier to automatically credit or debit an account
Dim lPrimeAcct ‘The primary account
Dim lSecondAcct ‘The second account into which to transfer funds
Dim lAmount ‘The amount to be credited, debited, or transferred
Dim objMTX ‘The Transaction Server Component
Dim lRetnVal ‘The Return Value

RtnMsg = “”
‘Decide which component to use
If (Request.Form(“Component”) = “Account”) Then
ProgID = “Bank.Account”
Else
ProgID = “Bank.MoveMoney”
End if

‘Decide the transaction type

If (Request.Form(“TransType”) = “Debit”) Then
TransType = “Debit”
Multiplier = “-1”
Elseif (Request.Form(“TransType”) = “Credit”) Then
TransType = “Credit”
Multiplier = “1”
Else
If (ProgID = “Bank.Account”) then
RtnMsg = “Error. You must select the MoveMoney component to transfer
➥funds.”
Elseif (Request.Form(“SecondAcct”) = “”) Then
RtnMsg = “Error. You must provide a second account number to transfer
➥funds into.”
Else
TransType = “Transfer”
Multiplier = “1”
lSecondAcct = Request.Form(“SecondAcct”)
End if
End if

‘Check for values

If (Request.Form(“PrimeAcct”) = “”) Then
TransType = “”
RtnMsg = “Error. You must provide the account number to be credited or
➥debited.”
Else
lPrimeAcct = Request.Form(“PrimeAcct”)
End if

35.1389-6 CH28 608 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Integrating Active Server Pages with Transaction Server 609

If (Request.Form(“Amount”) = “”) Then

TransType = “”
RtnMsg = “Error. You must provide the amount to be credited or debited.”
Else
lAmount = Request.Form(“Amount”)
End if

If (TransType <> “”) Then

‘Create the Transaction Server Component
Set objMTX = CreateObject(ProgID)
If objMTX is nothing Then
TransType = “”
RtnMsg = “Error. Could not create Transaction Server Object.”
End if
End if

‘Call the Transaction Server Component

If (TransType <> “”) Then
If (Request.Form(“Component”) = “Account”) Then
lRetnVal = objMTX.Post(CLng(lPrimeAcct), (CLng(lAmount) *
➥CLng(Multiplier)), RtnMsg)
Else
lRetnVal = objMTX.Perform(CLng(lPrimeAcct), CLng(lSecondAcct),
➥CLng(lAmount), TransType, RtnMsg)
End if
End if

After you perform this script, you can build virtually the same HTML document, including the
response message returned from the Transaction Server components, and populate the HTML
elements with the values the user submitted, as shown in Figure 28.22. For the complete
HTML and ASP listings, you can find the full documents on the CD enclosed with this book.

FIG. 28.22
The same HTML form
can be used for the
response from the
Transaction Server
components by adding
the result message
above the form.

Part
VI
Ch

28

35.1389-6 CH28 609 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 610 Chapter 28 Working with the Microsoft Transaction Server

A Note from the Author

If you have been following along with this example, loading the Bank component into Transaction
Server and building the Active Server Page application using the above code (or by pulling the code
off the CD), you have probably found that it doesn’t work. Instead of working the way that this text
leads you to believe that it should, you receive an error message. The reason for this problem is that
you still need to make modifications to the Banking sample application that accompanies Transac-
tion Server in order to be able to use it with Active Server Pages. The VBScript in our ASP page is
passing all parameters to the Transaction Server components as Variant datatypes, while the
components are expecting string and long datatypes. You need to modify the VB components to
expect all method parameters to be Variant before these components will work with our ASP
version. The next section explores how components need to be designed for use with Transaction
Server and Active Server Pages.

Building Transaction Server Components for Use

with Active Server Pages
When designing and building custom components for use with Transaction Server and Active
Server Pages, there are a few design constraints that you need to keep in mind. These con-
straints include how the component methods need to be designed to work with the Transaction
Server threading, and how this will affect the way the component is used within ASP applica-
tions. Also affected are how parameters need to be declared for all public methods that might
be called from an ASP application.

Threading and Transaction Server Components

Normally, when you are using custom components with Active Server Page applications, the
component threading is controlled by the ASP executable so that a single ASP page has a dedi-
cated component thread for the duration of the ASP processing for that page. With components
that use an apartment threading model (such as those used with components built using Visual
Basic), this means that your ASP application can set variables within the component and make
subsequent calls to the component, knowing that the variables that you have set in the compo-
nent are still there. The state of the component is consistent until the page has completed all of
its processing and the component thread has been ended.

With components in Transaction Server, Transaction Server is controlling the threading of

components, not the ASP engine. Once Transaction Server has activated a component and has
it running in a thread, Transaction Server can share that single component thread among
multiple processes. This means that you cannot count on the state of the component to stay as
you have left it until the next call you make to one of the component’s methods.

Because of the way that Transaction Server controls the threading of components, each com-
ponent method call needs to be self-contained. The method should receive all necessary data
and variables through parameters to the method, and should return any results back to the
calling routine through other method parameters.

35.1389-6 CH28 610 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 Building Transaction Server Components for Use with Active Server Pages 611

N O T E It is possible to design and develop components for use with Transaction Server that do
maintain state between method calls. However, it is not advisable to do this, as they incur a
significant amount of additional overhead and don’t perform or scale as well as stateless components.
State-full components are also more prone to cause errors in processing because as soon as a
component signals to Transaction Server that it has successfully completed its actions, Transaction
Server commits the transaction and resets the component. This wipes out all state information that
was being maintained in the component.

An alternative to state-full components is the use of the shared objects within Transaction Server. This
is a facility within Transaction Server where it maintains a set of data values, and controls access to
them so that only one process thread may reference or update them at any point in time. These shared
objects are just what the name implies; they are shared between all process threads running within
Transaction Server, and may be altered by any of the running process threads. ■

Parameter Passing with Active Server Pages

Because all variables in VBScript are Variant, you need to make all of the parameters in the
public component methods to be Variant, also. Once you are within the component method,
the parameters can be converted into whatever datatype they need to be for the proper func-
tioning of the component itself.

For example, the Post method of the Account component in the Transaction Server banking
example was declared as follows:
Public Function Post(lngAccountNo As Long, lngAmount As Long, _
ByRef strResult As String) As Long

Because none of the parameters for this method are Variant, the method call fails when made
from an ASP application. In order to get this method to work with an ASP application, the first
line of the method was converted to the following code, where all variables are passed in as
Variant, and then converted to variables of the original datatypes.

Public Function Post(alngAccountNo As Variant, _

alngAmount As Variant,
ByRef astrResult As Variant) As Long

Dim lngAccountNo As Long

Dim lngAmount As Long
Dim strResult As String

lngAccountNo = CLng(alngAccountNo)
lngAmount = CLng(alngAmount) Part

Just before returning from this method, the result string is copied to the Variant version.
VI
Ch
astrResult = strResult

After these changes have been made to the Post method in the Account class and the equivalent 28
changes made to the Perform method in the MoveMoney class, you need to recompile the ActiveX
DLL in which these components are encapsulated. Next, you need to delete the components from

35.1389-6 CH28 611 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 612 Chapter 28 Working with the Microsoft Transaction Server

Transaction Server, and re-register the new version. After you have done all of this, you should be
able to get the example ASP application to work with Transaction Server.

From Here…
In this chapter, you have learned about Microsoft’s Transaction Server: what it is, how it works,
and how to use it with Active Server Pages. You’ve learned how Microsoft designed Transac-
tion Server to be a combination of a transaction monitor and an object request broker. You’ve
seen how using Transaction Server enables you to build your applications as a series of small,
single-purpose modules that use Transaction Server to glue them all together. Transaction
Server also enables you to build large, distributed applications that are scaled across several
servers, with relative ease.

The biggest shortcoming with Transaction Server is that all processing that you can perform
with it is synchronous in nature. This means that everything in the processing chain has to
wait until everything else in the chain is complete before processing control can be returned to
the calling application. If your application needs to perform some time-consuming actions that
the calling application does not need to wait for, you will have trouble creating this capability by
using Transaction Server alone. However, if you also incorporate Microsoft’s Message Queue
Server into your application architecture, you can use it to provide you with the capability to
spawn off asynchronous processes to perform these time-consuming tasks.
From here you might want to check out some of these other chapters on related subjects:

■ To gain an understanding of messaging between application processes and systems, and

how this can be used in an Active Server Page application, read Chapter 29, “Working
with the Message Queue Server.”
■ To learn what’s involved in building a corporate intranet or extranet application, check
out Chapter 30, “Application Development for the Corporate Intranet/Extranet.”
■ To see how you can integrate all of these technologies together to build a large-scale
Active Server Page application, see Chapter 32, “Putting It All Together: Creating an
N-Tier Online Catalog Application.”

35.1389-6 CH28 612 10/26/98, 10:09 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 Sring 11.13.97 LP#3

 29
613

C H A P T E R
Part
VI
Ch

29

Message Queuing in a
Working with the Distributed Environment
In most large, distributed enterprise

Message Queue Server environments, there are several

computer systems that need to pass
information back and forth. Most of
this information does not need to be
passed in an interactive manner, but
can make use of message queues
instead.

When you build enterprise Web applications, you often need Microsoft’s Message Queue
Server
to pass information to other applications within your orga-
nization. Usually, this information is either passed in a Microsoft’s new Message Queue
Server, part of its Windows NT
synchronous manner, where the Web application waits
Server, Enterprise Edition, enables
until the other application has the information, or in some
you to build large, distributed mes-
sort of asynchronous manner, often by writing the infor- saging applications.
mation to a file that is then picked up by another process.
Neither of these standard approaches are an ideal ap- Integrating Active Server
proach to the problem of passing this information along. A Pages with Message Queue
better solution would be to send the information directly Server
to the other application—whether it was located on the You can use Active Server Pages to
same computer as your Web server, or on another com- create and send messages to other
puter within your corporate network—without having to applications in your environment.
wait for the other application to receive the information This provides you with the capabil-
from you. It would be even better if the receiving applica- ity to spawn separate processes
tion doesn’t even have to be running or reachable when from your ASP application that do
you send the information. This would enable your Web not need to be located on the same
application to continue working even when the other computer as your Web server.
application was unavailable, your internal network was
having problems, and so on.

36.1389-6 CH29 613 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 614 Chapter 29 Working with the Message Queue Server

This is all possible through the use of what is known as message queuing systems (also knows as
message-oriented middleware, or MOM for short). These systems enable you to send a message
containing the information you need to pass to another application, in whatever form you need
to send the information in (text, binary, and so on), without waiting for the receiving applica-
tion to get the information. If the other application isn’t running or isn’t reachable, the message
is maintained in the queue until the other application is available and can retrieve the message.
One of the newest products in this category is Microsoft’s Message Queue Server.

Not all the functionality of Microsoft’s Message Queue Server is available or practical for use
in Active Server Page applications. There is, however, a core set of functions that are not only
available, but are very practical for use in your ASP applications. Anywhere you need to pass
information to another process and your ASP application does not need to receive any results
from the other process, Message Queue Server might be a potential, if not the most practical,
solution for use in your implementation. ■

What Is Message Queuing?

If you haven’t heard about message queuing systems, it is currently one of the quickest grow-
ing areas of enterprise software systems. It enables application developers to build systems
that easily pass information to other systems that are running on the same computer, on an-
other computer in the local area network (LAN), or on another computer that is far away over
the companies’ wide area network (WAN), or possibly even over the Internet.

In a sense, message queuing can be thought of as e-mail for applications. When you send some-
one an e-mail message, you don’t know (or care) if the message recipient is logged into the
e-mail system at the time. You don’t even care if the recipient’s computer is turned on. You do
know that, if you have addressed the message correctly, your message will eventually get to
the recipient. Now imagine that the sending and receiving people are applications instead. If
the sending application needs to pass information to another application, but doesn’t need to
wait for the receiving application to act on the information, then why should the sending appli-
cation need to make sure that the receiving application is up and running?

The message queuing system handles getting the message from the sending system to the
receiving system, regardless of what network protocols are being used, and performs whatever
translations are necessary to convert the message to the format needed by the receiving sys-
tem (ASCII to EBCDIC, Little Endian to Big Endian, and so on). The message queuing system
also maintains the message when one or more of the systems involved in the message delivery
are down or unreachable due to network problems.

In a sense, message queuing systems can be thought of as e-mail for applications. As you can
probably guess from the previous list of functions that the message queuing systems perform,
they provide much more than just mail delivery. Some of the functionality provided by message
queuing systems include:

■ Guaranteed message delivery.

■ Store and forward capabilities.

36.1389-6 CH29 614 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 What Is Message Queuing? 615

■ Queue browsing capabilities for the receiving application. Part

■ Message prioritization and aging for controlling the delivery and storage of messages. VI
■ Transactional messages.
Ch
■ Journal and dead-letter queues for auditing and administrative purposes.
29
Guaranteed Message Delivery
Message queuing systems provide guaranteed message delivery from the sending application
to the receiving application by maintaining the message queue on the disk drive of every com-
puter involved in the message delivery. If the sending application is on one computer and the
receiving application is on another, the message is first written to the queue on the disk of the
sending computer. The message is then sent to the queue on the disk of the receiving com-
puter. Only after the message has successfully made it into the disk queue on the second com-
puter is it removed from the disk queue on the first computer. The message is maintained in
the queue on the disk of the receiving computer until it is retrieved by the receiving applica-
tion, as shown in Figure 29.1.

FIG. 29.1
Message queuing
systems guarantee Message Sending Message Receiving
Application Application
message delivery by
maintaining the Message Queueing Server Message Queueing Server
message queues on the
disk drives of both the Message Queue Wide or Local Area Message Queue
sending and receiving Network

computers.
Queue Persistent Queue Persistent
Storage Storage

N O T E Although the message queuing systems do guarantee message delivery, there is still the
possibility of losing messages in the event of a catastrophic disk failure. In this case, if the
disk is not mirrored or some other means of providing a fault-tolerant disk drive system, then
messages in the queue could be lost. ■

With most message queuing systems, it is also possible to send messages that are maintained
in memory only for faster delivery to the destination system (often called express delivery).
When messages are sent in this way, if the computer holding the message crashes for some
reason before being able to pass the message on to the next process or computer in line, then
the message will be lost. This mode of message delivery is usually designated as an unrecover-
able method of message delivery.

For instance, if you are building a virtual store front, and you want to automate as much as
possible, you would want to pass every order placed by your customers to your order fulfill-
ment system. If the order fulfillment system were currently offline (performing maintenance

36.1389-6 CH29 615 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 616 Chapter 29 Working with the Message Queue Server

or daily batch processing), you would not want to risk losing the customer’s order in the event
that your Web server crashes before the order fulfillment system could come back online. For
this very reason, you would want to specify that the order message was recoverable when you
placed it into the queue to be passed to the order fulfillment system.

Store and Forward

Store and forward functionality enables message queues to be defined and configured so that
they take several hops between the sending and receiving systems. With each hop, the mes-
sage is maintained on the next intermediate computer until such time that the message can be
passed along to the next computer in the chain. This is a desirable configuration in situations
where the computers are very far apart geographically and communications between any two
points may be out of service at any point in time. It is also a useful configuration to use when
the means of communication is diverse, with a mixture of networking technologies in use, as
shown in Figure 29.2.

FIG. 29.2
Store and forward
functionality is desirable
in situations where
multiple communication
technologies are in use.
Message Sending Message Receiving
Application Intermediary Queue Application

Wide Area
Network

Queue Queue Queue

Storage Storage Storage

The primary benefit of the store and forward model is that each hop a message takes from one
computer to the next can be kept down to a reasonable length with a fairly high level of avail-
ability. By using this model over a world-wide network, an efficient message queuing system
could be designed, configured, and made operational with only a moderate amount of effort.

For instance, if your virtual store made use of numerous shipping outlets scattered around the
world, you would want to pass the order information to the shipping outlet closest to the cus-
tomer. Because of time differences, the servers at all of your shipping outlets are not likely to
be up and running at the same time. Another factor to take into consideration is the reliability
of the network. If there is a realistic chance that network connections between any two servers
may not be operational at all times, then you might want to design a series of queues through
which the order messages may pass when traveling from your Web server to the destination
shipping outlet. By configuring your network with numerous short hops that a message will
take between the origination and destination, the reliability that the message will reach its
destination is greatly increased. For this type of configuration, you would want to make exten-
sive use of store and forward functionality in your message queuing system.

36.1389-6 CH29 616 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 What Is Message Queuing? 617

Queue Browsing Part

Normally, when the receiving application retrieves a message from the queue, the message is VI
removed from the queue. However, there are times when you don’t want the message to be
Ch
removed from the queue. For these situations, most message queuing systems enable the
receiving application to browse the queue instead. By browsing the queue, the receiving appli- 29
cation can retrieve the contents of the messages in the queue without removing it from the
queue.

One way in which this capability can be used is if you have an application that needs to use the
data in the queue messages to perform some task that is dependent on another system. If that
other system isn’t available, you don’t want the message to be removed from the queue be-
cause your application needs to maintain the data until the other system is available for per-
forming the necessary task. In this case, the receiving application could browse the queue for
the incoming message, and then attempt to perform the task with the other system. If the
application is successful in performing its task, it can retrieve the message from the queue,
thereby removing it from the queue. If the application is unsuccessful in performing its task
because the other system is unavailable, the message can be left in the queue. Then, after a
suitable amount of time had passed (depending on the situation and the needs of the applica-
tion, this could be a short as 5-10 seconds to several hours), the application could browse the
queue and attempt its task again, repeating this pattern until the other system is available and
the application is able to successfully complete its tasks. If this application were taking data
updates from a message queue and making the appropriate changes to a database server, the
process would look something like Figure 29.3.

FIG. 29.3 1. Browse the queue for

Database Server
An application could data updates
2. Perform the data
browse a message updates against the
Database
queue for data updates database
Data Update Queue Updating
to be run against a Application
database server. If the
database is unavail- 3. If the database was
available, remove the
able, the messages can data update from the Database
be left in the queue. queue

Message Prioritizing and Aging

Message sending applications can assign a priority to a message to control the delivery and
where in the destination queue the message is placed. A higher priority message will be deliv-
ered to the destination queue in a quicker manner, and will likely be placed into the queue
closer to the front than messages with lower priority. The priority does not affect how the mes-
sages are retrieved from the queue, and cannot force the receiving application to retrieve the
message from the queue any faster than it would a message with a lower priority. The receiving
application can only retrieve the messages from the destination queue in the order in which
they were placed in the queue. The receiving application does not have the capability to bypass
lower priority messages in the queue to retrieve higher priority messages first.

36.1389-6 CH29 617 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 618 Chapter 29 Working with the Message Queue Server

N O T E The primary benefit of sending a message with a high priority is that the message is
delivered to the destination queue faster than messages with a lower priority. This will make
a difference in situations where a large number of messages are being placed into the queue with a
mix of priorities. If the situation has few messages being placed in the queue on an infrequent basis,
then there will be little benefit with setting a high priority for specific messages. ■

When sending a message, the sending application can specify a length of time for the message
to remain in the queue. If this time expires and the message has not been retrieved from the
queue by the receiving application, the message is automatically removed from the queue. The
message is removed from the queue even if it has not reached the destination queue, but is
somewhere in the process of being delivered. This is a feature that is useful when dealing with
data that is time sensitive and must be acted on within a certain amount of time. However, this
should only be used with data that, if not acted on within the specified time frame, is better off
not being acted on at all.

Transactional Messages
Most message queuing systems enable the sending and receiving of transactional messages.
These are messages that make use of a transaction coordinator, such as Microsoft’s Distrib-
uted Transaction Coordinator (DTC), to control the sending and receiving of the message. If
anything goes wrong in the entire process and the receiving application is not successful in
retrieving the message, the entire transaction is rolled back and the sending application is
notified of the failure along with what caused the transaction to fail.

Journal and Dead-Letter Queues

Most message queuing systems provide at least two administrative message queues, known as
the journal and dead-letter queues. These two queues perform two different functions for the
purpose of auditing and error recovery. The journal queues are used to record every message
that passes through the queue with which the journal queue is associated. These queues can
be used to perform audit-trail analysis of where and how a message passed on its way from the
sending application to the receiving application.

Dead-letter queues store messages that could not be delivered. If a message was being di-
rected from an origination queue on one machine to a nonexistent destination queue on an-
other machine, that message would be undeliverable. As a result, the message would be moved
into the dead-letter queue for later examination by the system administrator in order to deter-
mine what problem the message had. The dead-letter queue also often acts as an overflow
queue, where if a destination computer is unavailable and more messages are being sent to that
destination than can be held in the origination queue, the excess messages may be redirected
to the dead-letter queue. If this is the case, the messages in the dead-letter queue will not be
delivered once the destination computer becomes available and begins receiving messages.
Instead, an administrator will have to manually resend the messages to the appropriate destina-
tion queue.

36.1389-6 CH29 618 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Microsoft’s Message Queue Server 619

N O T E The sender will not automatically be notified if the message ends up in the dead-letter Part
queue unless the application is specifically designed and built to perform this type of VI
automatic notification. ■
Ch

29
Microsoft’s Message Queue Server
One of Microsoft’s newest products is Message Queue Server, Microsoft’s entry into the mes-
sage queuing system market. This is a very impressive product that is easy to use, both as a
developer and an administrator. In this section, you will examine Message Queue Server from
the perspective of understanding what is possible and realistic in Active Server Pages applica-
tions. There is a lot more functionality in Message Queue Server than will be covered in this
section, but most of it is beyond the scope of this chapter and this book.

One of the things that sets Message Queue Server apart from the competition is the capability
to automatically trigger events in the receiving application when a message has arrived at the
destination queue. This enables the receiving application to be completely event driven, while
other products require the receiving application to perform some sort of polling to check for
arriving messages. This can be accomplished through the use of call-back functions (where
Message Queue Server calls an exposed function in your application), or by passing a window
event to your application.

N O T E Some competing products, such as IBM’s MQSeries, provide a separate application that
can signal the receiving application that a message has arrived in the destination queue.
This can be done by triggering an event in the receiving application, or by starting the receiving
application. While accomplishing the same end result, this approach is not integral to the product and
results in additional processing overhead not present in Microsoft’s Message Queue Server. ■

Because Active Server Pages lend themselves more toward using the ActiveX components of
Message Queue Server than the API, the next section looks at Message Queue Server from
the perspective of the components. In examining these components, you’ll look at the proper-
ties with which you need to be familiar and how to use them. You’ll also look at the various
methods involved that enable you to find and open message queues, send and receive mes-
sages from the queues, and close and delete the queues, all from within an Active Server Page
application (although some of these functions are better left to other processes, I’ll still cover
how you can perform them using Active Server Pages). Because of the nature of Active Server
Page applications, I will not be looking at Message Queue Server events and how to take ad-
vantage of them in applications that lend themselves to using them.

ActiveX Components
Message Queue Server comes with a series of ActiveX components that can be used to per-
form various functions. These components are listed in Table 29.1.

36.1389-6 CH29 619 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 620 Chapter 29 Working with the Message Queue Server

Table 29.1 Message Queue Server ActiveX Components

Component Description
MSMQQuery This component can be used to locate a collection of
existing queues.
MSMQQueueInfos This is a collection of queues located using the
MSMQQuery component.
MSMQQueueInfo This component contains the information needed to
create or open a queue.
MSMQQueue This component represents an actual message queue,
and provides functionality to receive and send mes-
sages.
MSMQMessage This component represents a message that is to be sent
or received.
MSMQEvent This component contains all of the event handling for
Message Queue Server. It can be used to trigger
application events when messages arrive in a destination
queue. We will not be looking at this component in this
chapter.
MSMQTransactionDispenser This component is used to create a transaction object.
MSMQTransaction This component is used to commit and abort transac-
tions.

I’ll examine each of the components in detail as you look at each of the basic functions that you
will need to perform in working with Message Queue Server. These functions are:

■ Creating a new queue.

■ Locating an existing queue.
■ Opening a queue for sending or receiving.
■ Sending messages to a queue.
■ Receiving messages from a queue.
■ Closing an open queue.
■ Deleting an existing queue.

Creating Queues
Creating a queue is done through the use of the MSMQQueueInfo component. This component
has several properties, which are listed in Table 29.2. Only two of them have to be provided
with any data.

36.1389-6 CH29 620 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Microsoft’s Message Queue Server 621

Table 29.2 Properties of the MSMQQueueInfo Component Part

Property Description VI
dateCreateTime Indicates when the queue was created. This is a read-only property Ch
and cannot be set.
29
dateModifyTime Indicates when the queue’s properties were last modified. This is a
read-only property and cannot be set.
guidQueue This is a unique ID for the queue that is created by Message Queue
Server when the queue is created. This is a read-only property and
cannot be set.
guidServiceType This is a GUID that identifies the queue type. A queue type ID can
be generated by using the uuidgen.exe application. This is an
optional property that can be used to locate the queue. This property
must be specified before creating the queue.
isTransactional This is a boolean value that specifies whether the queue is transac-
tional or not. This is a read-only property and cannot be set.
lAuthenticate This property specifies whether the queue will accept
nonauthenticated messages.
lBaseProperty This property defines the default priority of all messages submitted
to this queue.
lJournal This property defines whether messages retrieved from the queue
should have a copy written to the journal queue.
lJournalQuota This property specifies the maximum size of the journal queue.
lPrivLevel This property specifies the privacy level of the queue, and whether
the queue will accept unencrypted messages.
lQuota This property specifies the maximum size of the queue.
strFormatName This property specifies the format name of the queue. This property
must be filled in before the queue can be opened. However, this does
not have to be provided to create the queue, in which case Message
Queue Server creates a value from the strPathName property.
strLabel This is a string label that describes the queue. This property should
be filled in before creating a new queue, as this is one of the most
common properties that you will use in searching for an existing
queue. The maximum length of this string is 124 characters.
strPathname This is the only property that is required for creating a new queue.
This specifies the computer name, whether the queue is public or
private, and the name of the queue. The maximum length of this
property is 124 characters.

36.1389-6 CH29 621 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 622 Chapter 29 Working with the Message Queue Server

To create a new queue, all you really have to specify is the queue path, which consists of the
computer name, whether the queue is private, and the queue name separated by slashes.
The computer name may have a single period substituted for it to specify the local computer.
If a queue is to be a private queue, which means that it will only be accessible to applications
running on the same computer as the queue, the middle portion of the path name should be
Private$. Queue path names are formatted as follows:

<Computer name>/[Private$/]<Queue name>

So, to create a queue on a computer named Bullfrog, with a queue name of MyQueue, you
would format the queue path name as follows:
Bullfrog/MyQueue

To create that same queue as a private queue, it would be:

Bullfrog/Private$/MyQueue

And to create the queue on the local computer, you could use this form instead:
./MyQueue

After you have specified the path name and the label for your queue, you can create the queue
using the Create method of the MSMQQueueInfo component. You can create a new queue using
the following VBScript code:
Dim qinfoMyQueue
Set qinfoMyQueue = Server.CreateObject(MSMQ.MSMQQueueInfo”)
qinfoMyQueue.strPathName = “./MyQueue”
qinfoMyQueue.strLabel = “My Queue”
qinfoMyQueue.Create

At this point, you have a queue that you can open and begin sending and receiving messages
through.

CAUTION
In the beta version of Message Queue Server that I am working with to write this chapter, I cannot advise
that you create queues within an Active Server Page application. The queues that I create in an ASP
application cannot be opened for receiving messages in any other application or programming language. I
could open the queue for sending messages in anything else, but I could not receive messages. I could open
the queue in any mode and perform any action from within an ASP application. If I create the queue using
another application, I still have full functionality within my ASP application, as well as within any other
application or programming tool.

Locating Queues
To locate an existing queue, you use a combination of the MSMQQuery and the
MSMQQueueInfos components. First, you create an instance of the MSMQQuery component
and run its LookupQueue method, which returns a MSMQQueueInfos component instance. The
LookupQueue method takes a number of parameters, all of which are optional. The parameters
are listed, in the order in which they are passed to the LookupQueue method, in Table 29.3.

36.1389-6 CH29 622 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Microsoft’s Message Queue Server 623

Table 29.3 Parameters for the MSMQQuery.LookupQueue Method Part

Parameter Description VI
guidQueue The ID of the queue to be located. Ch

guidServiceType The service type of the queue to be located. 29

strLabel The label of the queue to be located.
dateCreateTime The creation time of the queue to be located.
dateModifyTime The modification time of the queue to be located.
relServiceType Relationship parameter for the service type parameter.
relLabel Relationship parameter for the label parameter.
relCreateTime Relationship parameter for the creation time parameter.
relModifyTime Relationship parameter for the modification time parameter.

The relationship parameters default to a test for equality on the values supplied for the corre-
sponding parameters. Any of the parameters that are left out are not used in locating matching
queues. The relationship parameters can be any of the values listed in Table 29.4.

Table 29.4 LookupQueue Relationship Parameter Values

Label Value Description
REL_EQ 1 Equal
REL_NEQ 2 Not equal
REL_LT 3 Less than
REL_GT 4 Greater than
REL_LE 5 Less than or equal
REL_GE 6 Greater than or equal
REL_NOP 0 Ignore this parameter

The MSMQQuery.LookupQueue method returns a MSMQQueueInfos component, which is a collec-

tion of queue information components (MSMQQueueInfo). This component has two methods,
both of which you use in navigating through the collection of queues. The first method, Reset,
places you at the beginning of the collection of queues. You use the Next method to navigate to
the first queue in the collection, as well as to the remaining queues in the collection. The Next
method returns an instance of the MSMQQueueInfo component.

36.1389-6 CH29 623 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 624 Chapter 29 Working with the Message Queue Server

By using these two components, you can locate the queue that you created earlier by using the
following VBScript code:
Dim query
Dim qinfos
Dim qinfoMyQueue
Set query = Server.CreateObject(“MSMQ.MSMQQuery”)
Set qinfos = query.LookupQueue(,,“My Queue”)
qinfos.Reset
Set qinfoMyQueue = qinfos.Next

Notice in the LookupQueue method that you must supply two commas to indicate that you are
supplying the third parameter, and to use the default for all of the rest of the parameters. If you
had been using Visual Basic, you could have just specified the parameter name and let VB
determine which parameter you were supplying.

At this point, you should have a queue description object, assuming that the queue was located.
You can check to make sure that the lookup was successful by checking to see if the queue
descriptor is “nothing.” If the queue is “nothing,” then the queue was not located and may need
to be created. Otherwise, you can continue by opening the queue.

Opening Queues
To open the queue, you have to use the Open method of the MSMQQueueInfo component. This
method takes two parameters, the access mode and the sharing mode. The available access
modes are listed in Table 29.5.

Table 29.5 Queue Open Access Mode Parameter Values

Label Value Description
MQ_PEEK_ACCESS 32 The application will only be able to peek at messages
in the queue without removing any messages.
MQ_RECEIVE_ACCESS 1 The application will be able to retrieve messages,
thereby removing them from the queue. If the queue
is opened using this access mode, the application will
also be able to peek at the messages without
removing them from the queue.
MQ_SEND_ACCESS 2 The application will be able to send messages to the
queue.

The sharing mode of the queue must be one of the values listed in Table 29.6.

36.1389-6 CH29 624 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Microsoft’s Message Queue Server 625

Table 29.6 Queue Sharing Mode Parameter Values Part

Label Value Description VI

MQ_DENY_NONE 0 This is the default value and must be used if Ch
opening the queue to peek at or send mes-
sages. This sharing mode enables other
29
applications to open the same queue.
MQ_DENY_RECEIVE_SHARE 1 This value should only be used if opening the
queue for receiving messages. This value will
prevent any other applications from opening
the same queue for receiving messages.

This method returns a MSMQQueue component instance. After you have either located a queue
or created a new queue, you can open the queue for sending messages with the following code.
Dim qMyQueue
Set qMyQueue = qinfoMyQueue.Open(2, 0)

At this point, you can begin sending messages to the queue.

N O T E In VBScript, you must use the values for the open access mode and sharing mode, as the
ASP application does not understand the textual representations of these values. ■

Sending Messages
In order to send a message, you have to use an instance of the MSMQMessage component. After
you have the message object, you need to set several of its properties, including the message
label and body properties. The properties of the message component are listed in Table 29.7.

Table 29.7 MSMQMessage Component Properties

Property Description
body This is the actual message that is being sent. This can be any
datatype, structure, binary data, or even an ActiveX object.
binSenderCert The digital security certificate of the sender. This is for use in
sending and receiving encrypted messages.
binSenderID This is an identifier of the application that sent the message. This
is a read-only property that is set by Message Queue Server.
dateArrivedTime This property contains the time that the message arrived in the
destination queue. This is a read-only property that is automati-
cally set by Message Queue Server.

continues

36.1389-6 CH29 625 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 626 Chapter 29 Working with the Message Queue Server

Table 29.7 Continued

Property Description
dateSentTime This property contains the time that the message was sent. This
is a read-only property that is automatically set by Message
Queue Server.
quidSrcMachine This is the GUID (Globally Unique Identifier) of the machine that
sent the message. This is a read-only property that is automati-
cally set by Message Queue Server.
id This is a unique message identifier that is generated by Message
Queue Server. This is a read-only property.
idCorrelation This is the ID of an original message, if this message is a
response or acknowledgment of another message.
isAuthenticated This is a Boolean value that specifies whether the message has
been authenticated by Message Queue Server. This is a read-only
property that is automatically set by Message Queue Server.
lAck This property specifies what sort of acknowledgment should be
placed in to the journal queue when the message is received from
each queue.
lAppSpecific This property can be used for an application-specific index that
can be used for sorting messages.
lAuthLevel This property specifies whether the message needs to be
authenticated when it reaches the target queue.
lClass This property specifies the message type. This can be a normal
message, a positive or negative acknowledgment, or a report
message. This is a read-only property that is automatically set
by Message Queue Server.
lDelivery This property specifies the method of delivery. The default value
for this property is MQMSG_DELIVERY_EXPRESS, which keeps the
message in memory where it is not recoverable in case of a
system crash. The other possible setting for this property is
MQMSG_DELIVERY_RECOVERABLE, which keeps the message stored
on the disk drive of every computer in the delivery chain, so that
the message can be recovered and delivered in the event of a
system crash.
lEncryptAlg This property specifies the encryption algorithm to be used for
encrypting the message.
lHashAlg This property specifies the hash algorithm to be used for
authenticating the message.
lJournal This property specifies whether to log this message in the journal
queue when it is retrieved, or to store it in the dead-letter queue if
the delivery time expires.

36.1389-6 CH29 626 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Microsoft’s Message Queue Server 627

Property Description Part

lPriority This property specifies the priority of the message. This is a value
from 0 (lowest) to 7 (highest), with a default value of 3. This value
VI
is used to determine how the message is routed across the Ch
network to its destination queue.
29
lPrivLevel This property specifies whether the message is private and
should be encrypted.
lMaxTimeToReachQueue This property specifies in seconds the maximum time for the
message to reach the destination queue. If the time specified
expires before the message has reached its destination queue,
the message is deleted or stored in the dead-letter queue. The
default value for this is LONG_LIVED, which is an administratively
controlled time that has a default value of 90 days.
lMaxTimeToReceive This property specifies in seconds the maximum time for the
message to be retrieved from the destination queue. If the time
specified expires before the message has been retrieved, the
message is deleted or stored in the dead-letter queue. The default
value for this is LONG_LIVED, which is an administratively con-
trolled time that has a default value of 90 days. This value should
always be equal to or greater than the lMaxTimeToReachQueue
property.
lSecurityContext This property specifies the security information needed to
authenticate messages.
lSenderIDType This property specifies the type of sender sending the message.
lTrace This property specifies whether the message should produce a
tracing report, which specifies the queue manager, message ID,
time to next hop, and so on.
lenBody This property specifies the length of the message body. This is a
read-only property that is automatically set by Message Queue
Server.
queueinfoAdmin This property specifies the queue to be used to send acknowledg-
ment messages.
queueinfoDest This property specifies the destination queue for the message.
This is a read-only property that is automatically set by Message
Queue Server.
queueinfoResponse This property specifies a response queue for sending response
messages.
strLabel This property specifies a descriptive label for the message.

36.1389-6 CH29 627 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 628 Chapter 29 Working with the Message Queue Server

N O T E The types of encryption that are available with Message Queue Server are RC2 and RC4,
two standard encryption algorithms that fall under the RSA banner using a Public/Private
Key methodology. Message Queue Server does provide the hooks so that you can plug in different
encryption algorithms if you so desire. ■

After you create an instance of the MSMQMessage component, you can set various properties,
and then send the message using the Send method, specifying the open queue to be used. The
properties that you will probably want to set include the lAck, lDelivery, strLabel, and body.
Other common properties that you might want to set are lMaxTimeToReceive and lPriority.
Using the queue that you opened in the previous section, you could send a message using the
following VBScript:
Dim MyMsg
Set MyMsg - Server.CreateObject(“MSMQ.MSMQMessage”)
MyMsg.strLabel = “My message”
MyMsg.body = “This is my message, which I am sending through the queue.”
MyMsg.lAck = MQMSG_ACKNOWLEDGMENT_NONE
MyMsg.lDelivery = MQMSG_DELIVERY_RECOVERABLE
MyMsg.Send qMyQueue

Receiving Messages
Receiving messages is even simpler than sending messages. After you have a queue open for
receiving, you can create the MSMQMessage object by calling the MSMQQueue.Receive method as
in the following code:
Dim RecvdMsg
Set RecvdMsg = qMyQueue.Receive

Once you have a message object, you can look at the message properties to read the message
label and body, as well as the various other message properties.

Closing Queues
When you finish sending all the messages you need to send or receiving the messages you
need to receive, you can close the queue by using the MSMQQueue component’s Close method,
as follows:
qMyQueue.Close

Once you have closed the queue, you cannot send or receive any more messages without re-
opening the queue.

Deleting Queues
When you have completely finished using a queue, you can always delete it using the
MSMQQueueInfo component’s Delete method, as in the following VBScript:

qinfoMyQueue.Delete

36.1389-6 CH29 628 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Message Queue Server and Active Server Pages 629

Deleting the queue does not delete the queue information object that you used to delete the Part
queue. It can be reused to create a new queue with the appropriate properties reset as neces-
sary. Once a queue has been deleted, it cannot be opened for sending or receiving and all mes-
VI
sages currently in the queue are lost. Ch

29
Message Queue Server and Active Server Pages
Because the logical use of an Active Server Page application with Message Queue Server is in
the generation of messages that are to be received by another application, that is where you
will be putting your focus in building a working example application. You’ll make use of the API
test application that is included with Message Queue Server as one of the example applications
to perform all the tasks that you don’t want to include in your ASP application. You could per-
form all of these tasks in the ASP application, and I encourage you to try it using the example
code provided above with each of the examples.

In your application, you will perform the following tasks:

1. Use the MQ API test application to create your message queue.

2. Locate, open, and send a message to the queue using ASP and VBScript.
3. Receive the message from the queue using the MQ API test application.
4. Delete the queue using the MQ API test application.
Once you have this system up and running, you might want to repeat steps 2 and 3 several
times, performing step 2 several times before performing step 3 for each step 2 you performed.

Creating a Message Queue

If you pull up the Message Queue Server Explorer, you should see all the existing queues on
your system in either the Sites or the Enterprise Server folders. If you haven’t created any
queues, Explorer should look like Figure 29.4.

If you open the MQ API test example application (located in the \Msmq\Sdk\Samples\Bin direc-
tory, and may also be installed on the Start menu in the samples folder within the Message
Queue Server folder), which is included with Message Queue Server, you can create a queue
for use in this example (I’ll include the code to create the queue if it is not found in the ASP
application, but everything will work better if you create the queue here). If you choose Api |
MQCreateQueue from the application menu, you will be presented with a dialog box asking for
the path name and the queue label. This first part of the path name should already be filled in
with the computer name on which you are running. Fill in the remaining portion with the path

36.1389-6 CH29 629 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 630 Chapter 29 Working with the Message Queue Server

name for the queue you are creating. Next, fill in a label for the queue, which you will be using
to locate the queue in you ASP application. In this case, you can fill in the path name as
ASPExampQueue and the label as ASP Example Queue, as shown in Figure 29.5. When
you click the OK button, you’ll be presented with a confirmation in the main API test window,
as shown in Figure 29.6.

FIG. 29.4
Upon opening the
Message Queue Server
Explorer, you will be
presented with all of the
existing queues on the
computers in your
enterprise configuration.

FIG. 29.5
Using the API test
application, you can
create a new queue by
providing just the path
name and label.

FIG. 29.6
The API test application
displays a message
confirming that the
queue was created
successfully.

After you create the queue, go back to the Message Queue Server Explorer, choose
View | Refresh from the main menu, and see the queue you just created appear in your server
(see Figure 29.7).

36.1389-6 CH29 630 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Message Queue Server and Active Server Pages 631

FIG. 29.7 Part

After creating a new
queue, you can see the VI
new queue in the Ch
Message Queue Server
Explorer by refreshing 29
the display.

Sending a Message to the Queue

To be able to submit messages to the queue that you just created, you first need to have an
HTML form that you can use for sending messages to the Web server. For this example, use a
simple form like the one shown in Figure 29.8, which prompts the user for the message label
and textual contents. (The HTML form used in this example, MSMQExample.htm, is available on
the CD that accompanies this book.)

FIG. 29.8
You can use a simple
HTML form to submit a
message label and text
for submitting to your
message queue.

On the receiving end of your Web server, you need to write a little VBScript in your ASP docu-
ment that is receiving the data submitted by your HTML form, and pass the data along to the
message queue. In this ASP page, you will perform the following steps:

1. Try to locate the queue that you created.

2. If you are unable to locate the queue, create a new queue.

36.1389-6 CH29 631 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 632 Chapter 29 Working with the Message Queue Server

3. Open the queue for sending the message.

4. Create a MSMQMessage object, and populate it with the data values submitted by the form.
5. Send the message to your queue.
6. Close the queue.

You can perform all of these steps using the code in Listing 29.1.

Listing 29.1 SUBMSG.ASP—You can capture the data submitted by the

user and place it into the queue as a message.
Dim query
Dim qinfos
Dim qinfoDest
Dim qDest
Dim msgSent
Dim msgBody
‘Find or create the ASP Example Queue
Set query = Server.CreateObject(“MSMQ.MSMQQuery”)
Set qinfos = query.LookupQueue(,,”ASP Example Queue”)
qinfos.Reset
Set qinfoDest = qinfos.Next
If qinfoDest Is Nothing Then
Set qinfoDest = Server.CreateObject(“MSMQ.MSMQQueueInfo”)
qinfoDest.strPathName = “.\ASPExampQueue”
qinfoDest.strLabel = “ASP Example Queue”
qinfoDest.Create
End If
‘Open the Queue
Set qDest = qinfoDest.Open(2, 0)
‘Send the Message
Set msgSent = Server.CreateObject(“MSMQ.MSMQMessage”)
msgSent.strLabel = Request.Form(“MsgLabel”)
msgBody = Request.Form(“MsgBody”)
msgSent.body = msgBody
msgSent.lAck = MQMSG_ACKNOWLEDGMENT_NONE
msgSent.lDelivery = MQMSG_DELIVERY_RECOVERABLE
msgSent.lMaxTimeToReceive = -1
msgSent.Send qDest
qDest.Close

N O T E Notice in Listing 29.1 that you did not set the message body directly to the value
submitted by the user. It would seem like it would be easier to just do the following:
msgSent.body = Request.Form(“MsgBody”)

However, when you attempt this, you will receive an error message stating that the object cannot be
automated. Instead, you need to place the value of the message body into a temporary holding
variable, and then set the message body to the variable. I can’t explain why this object cannot be
automated, and it may be fixed in a future version of ASP and/or MSMQ. ■

36.1389-6 CH29 632 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Message Queue Server and Active Server Pages 633

After you submit the message to the queue using the ASP page containing the code listed Part
previously (the entire ASP document can be found on the CD accompanying this book), you
can pull up the Message Queue Server Explorer to see whether the message was submitted
VI
into the queue. If you select the queue you created, right-click the mouse, select Properties, Ch
and then select the Status tab, you can see how many messages are currently in the message
queue (see Figure 29.9). You can also choose View | Refresh from the main menu with your
29
queue selected to see the messages in the queue in the main Explorer window, as shown in
Figure 29.10. Double-clicking the message brings up a Message Properties dialog box, which
you can use to examine the contents of the message by choosing the Body tab, as shown in
Figure 29.11.

FIG. 29.9
You can browse the
current status of your
queue through the
Message Queue Server
Explorer.

FIG. 29.10
You can visually display
the message in the
queue by refreshing the
Explorer view of the
queue.

36.1389-6 CH29 633 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 634 Chapter 29 Working with the Message Queue Server

FIG. 29.11
From the Explorer
window, you can
examine the contents
of messages in the
queue.

Receiving the Message from the Queue

Now that you have submitted a message to the queue, you can try and retrieve the message
from the queue. You can build the code to do this into an ASP application, although that is not a
very practical means of retrieving the message as it is unlikely that you would do this in a real
application situation. Your other options are to build a custom application, or use the API test
example application. For this example, use the API test application.

Before you can retrieve any messages, you need to open the message queue for receiving
messages. You can do this by choosing Api | MQOpenQueue from the main menu of the API
test application. This will bring up a dialog box that you can use to choose which queue to
open. (If you closed the API test application since you created the queue, you need to locate the
queue first.) Select the check box to indicate that you are opening the queue for receiving
messages, as shown in Figure 29.12, and then click OK. If you are able to open the queue, the
main window of the API test application will show that the queue was opened successfully (see
Figure 29.13).

FIG. 29.12
When opening the
message queue for
receiving messages, you
need to specify the
access mode for
receiving.

36.1389-6 CH29 634 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 Message Queue Server and Active Server Pages 635

FIG. 29.13 Part

The API test application
will inform you whether VI
the queue was opened Ch
successfully.
29

When you have the queue open, you can choose Api | MQReceiveMessage to retrieve the
message from the queue. Once again you will be presented with a dialog box for you to specify
the open queue and the timeout value (how long to wait before returning if there is no waiting
message already in the queue), as shown in Figure 29.14. If you click OK, the message will be
retrieved and its contents will be displayed in the main window of the API test application,
along with information about the message, such as the sending user, the message priority, and
whether the message was encrypted or authenticated (see Figure 29.15).

FIG. 29.14
The API test application
will ask you to verify the
queue and the timeout
period before receiving
a message.

FIG. 29.15
The received message
contents will be added
to the display in the API
test main window.

36.1389-6 CH29 635 10/26/98, 10:10 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 636 Chapter 29 Working with the Message Queue Server

N O T E In the beta version of Message Queue Server that I am using to write this, the API test
application is adding additional garbage to the end of the message text in the display. This
is being added by the API test application, and is not actually part of the message. If you check the
contents of the message through the Explorer, you will find that this garbage is not included in the
message. ■

Closing and Deleting the Queue

When you finish sending and receiving all the messages from this application, you need to
close the queue in the API test application. You can do this by choosing Api | MQCloseQueue
from the main menu. You will then see a dialog box to verify which open queue you want to
close, as shown in Figure 29.16. Clicking OK will close the queue and add a message to the API
test application main window stating that the queue has been closed (see Figure 29.17).

FIG. 29.16
The API test application
will ask you to select
which open queue to
close.

FIG. 29.17
The API test application
will inform you if there is
any problem closing the
queue.

Once you have completely finished with this queue, you can delete it, thereby removing it from
your Message Queue Server configuration. The queue must be closed before you can delete it
in the API test application. Choose Api | MQDeleteQueue from the main menu. You will be
presented, once again, with a dialog box asking you to select which queue is to be deleted (see
Figure 29.18). Clicking the OK button will delete the queue, with the confirmation message
being added to the main window display of the API test application (see Figure 29.19). If you
pull up the Message Queue Server Explorer and refresh the computer, you will see that the
queue is gone.

36.1389-6 CH29 636 10/26/98, 10:11 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 From Here… 637

FIG. 29.18 Part

When you delete a
queue, the API test VI
application will ask you Ch
to specify which queue
to delete. 29

FIG. 29.19
Once you have chosen
the queue to be
deleted, the API test
application will display
whether the queue was
successfully deleted.

At this point, if you submit a new message using the ASP application, the ASP application will
create a new version of the queue, which you may have to modify the ASP code to receive all
messages from and to delete once you are done.

N O T E To gain a better understanding of Message Queue Server, you might want to build your own
version of the API Test sample application using Visual Basic and the Message Queue
Server ActiveX components (the same ActiveX components we used in our ASP application.) You can
also build the MSMQ API Test application using Visual C++ or Delphi by using either the ActiveX
components or writing directly to the MSMQ API. The API Test sample application is supplied with
MSMQ complete with source code for Visual C++, which you can examine and modify yourself. ■

From Here…
In this chapter, you have learned what message queuing systems are and how they work.
You’ve also taken a look at Microsoft’s new Message Queue Server, and seen how it can be
used with an Active Server Page application to submit messages to another process that would
need to act on the information provided. This could be used in any context where you need to
trigger another process from input provided to an ASP application, in which you do not need to
provide the user with the immediate results of the process.

36.1389-6 CH29 637 10/26/98, 10:11 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 638 Chapter 29 Working with the Message Queue Server

From here, you might want to perform some experiments with Message Queue Server to see
what kind of custom applications you can build to receive the messages, and what you can have
those applications do with the information you have passed to them. You can also dig into the
administration and configuration of Message Queue Server, configuring queues that take infor-
mation from one server to another, where the sending ASP application is on one computer and
the receiving application is located on another. Unfortunately, the administration and configura-
tion issues involved in performing this are beyond the scope of the chapter and this book.

You might want to check out some of these other chapters on related subjects:

■ To learn how to use Microsoft’s Transaction Server with Active Server Pages to off-load
synchronous processing to other servers in your enterprise environment, read Chapter
28, “Working with the Microsoft Transaction Server.”
■ To learn what’s involved in building a corporate intranet or extranet application, check
out Chapter 30, “Application Development for the Corporate Intranet/Extranet.”
■ To see how you can integrate all of these technologies together to build a large-scale
Active Server Page application, see Chapter 32, “Putting It All Together: Creating an
N-Tier Online Catalog Application.”

36.1389-6 CH29 638 10/26/98, 10:11 AM

Brands3/stone3 SEUsing ActiveServerPages #1389-6 11.13.97 sring CH29 LP3

 30
639

C H A P T E R

Part
Developing Applications for an VI
Application Intranet Ch
See how to increase productivity
30
Development for the right away with little investment.

Building an Extranet

Corporate Intranet/ Understanding this marriage of the

intranet and Internet can mean
creating cutting-edge applications.
Extranet

When most people think of Web-based applications, the

examples that spring to mind are of the interactive sites
seen at www.intel.com or www.microsoft.com. But, in
reality, most of the application development you will prob-
ably end up doing with ASP will be on an intranet server
somewhere.

Many recent surveys have found that nearly 100 percent

of all companies will be implementing intranets in the
coming months and years. As an ASP developer, this is
great news. There are literally thousands of stand-alone
applications for everything from bookkeeping to cus-
tomer support that will be ported to an internal Web
server, if they aren’t already.

Extranets are Web-based applications that contain both

an intranet and Internet part. By burrowing through the
firewall to query a database that is populated internally,
you can securely provide information to those who
need it. ■

37.1389-6 CH30 639 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 640 Chapter 30 Application Development for the Corporate Intranet/Extranet

Developing Intranet Applications

If you counted all the Internet sites in the world, you would get a pretty big number. But if you
counted all the intranet sites in the world, you would have a huge number. Many companies
have an intranet site for every department. Others have taken a more structured approach.
Still, there will always be far more intranet Web sites than those on the Internet.

As a Web developer, this means that you will probably spend most of your time developing
applications for one of these intranet sites. In a recent poll, nearly every company asked stated
their intention to start porting their existing non-Web based applications to a version that will
run on a browser.

This means that most of the time, you won’t have to worry about having the greatest looking
graphics and the most eloquent prose accompanying your application. Instead, you can concen-
trate on functionality.

When building an intranet application, there are a couple big differences to be aware of: larger
bandwidth and standardized browsers. In reality, these differences enable you to develop more
complex applications more quickly.

Leveraging Bandwidth
The vast majority of Internet surfers are browsing the Web on a modem. And while modems
have grown faster and faster in speed, they come nowhere near the speed of a modern internal
network.

This means that you can expect the users of your intranet application to be surfing at a mini-
mum of 10MB a second. Generally, when developing for the Internet, you have to keep page
size small for fear of bogging down someone’s browser with a long download.

So what do you do with this larger than usual bandwidth? The figures below illustrate the trade
offs involved in creating any client/server application. Figure 30.1 shows an example of a thin
client, and Figure 30.2 shows fat-client browsers.
When developing an Internet-based application, you want to send as little information as pos-
sible, while simultaneously putting as much of the processing as possible onto the browser.

When building an intranet application, you don’t need to worry nearly as much about the
amount of information being sent. And, except for special cases, you also don’t have to worry
about excessive processor load on your server.

FIG. 30.1
Web browsers can be
thin clients...
(lots of bandwidth)

Client Server
(little processing) (lots of processing)

37.1389-6 CH30 640 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 Developing Intranet Applications 641

FIG. 30.2
...or fat clients.

(less bandwidth)

Client Server
(lots of processing) (lots of processing)
Part
VI
Ch
CAUTION
Don’t take this to mean that you don’t ever need to worry about excessive processor load. If you are running 30
a heavy-duty app that is going to be hit by thousands of employees every day (or even every hour), you will
indeed need to start thinking about the load on the server.
As a matter of fact, when it comes to very large corporate intranets, it is almost better to think of the site as
an Internet site.
Another point to consider is the amount of telecommuters that will be using your applications. These users
will be surfing to you on modem lines. If you expect a large number of these mobile users, it is also good to
think in terms of an Internet site.

A lot of development time is spent working around the limitations of the Internet and World
Wide Web. With an intranet, most of these problems go away. Instead, as a developer, you can
focus on creating a useful application.

Another nice advantage that you receive with increased bandwith is the capability to incorpo-
rate multimedia into your Web pages. On the Internet, graphics, sounds, and movies have to be
carefully manipulated to make sure they don’t require the user to wait for them to download.
With your intranet, you can incorporate a large AVI and not have to sweat about it being too
large.

So, imagine you are creating a training Web page. This Web page may be based on an existing
CD-ROM. There will probably be many movie-type tutorials on the CD-ROM. You can convert
these to some browser-enabled format and go. The bandwidth you will receive on your corpo-
rate intranet gives you this freedom.

When You Know What Browser They Have...

…then write for that browser.

If all of your users are going to be using Netscape Navigator, then it doesn’t make a lot of sense
to utilize client-side ActiveX controls. Conversely, if your client base is Internet Explorer, you
can include ActiveX controls in your layout.

When building Web-based applications, many developers prefer to use client-side scripting for
their data validation. JavaScript and VBScript can both be used to catch a multitude of errors.
The following list shows some examples of validation that client scripting works great for.

37.1389-6 CH30 641 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 642 Chapter 30 Application Development for the Corporate Intranet/Extranet

■ Multiple spaces
■ Null values
■ Illegal characters
■ Characters that must be present (such as the @ in e-mail)
■ Proper length

Of course, some types of validation are either very cumbersome or impossible with client-side
validation. The following is a list of validation that should probably take place on the server:

■ Duplicates in database
■ Validating address fields like ZIP code and country
■ Data integrity

On the Internet, if you want to use client-side scripting, you have several choices. You can
detect which type of browser the user has. Once you know that, you can then use either
VBScript for Internet Explorer and JavaScript for Navigator clients. However, taking this kind
of approach requires a lot more work because you have to develop the validation in both script-
ing languages.

Another alternative is to write your client-side scripting using the subset of JavaScript that both
browsers support. (Both Internet Explorer and Navigator support JavaScript to one degree or
another.) However, you have to be careful to only use those features of JavaScript that you are
sure the various browsers that will be visiting your site will be using.

When developing for the corporate intranet, these worries evaporate to a very large degree.
Most companies have defined a standard browser for their users. If you know which browser
your users will be using, you can save development time when developing your client scripting.

Below is a list of some of the features you can expect to be able to utilize if your standard
browser is Internet Explorer.

■ VBScript
■ JScript (Microsoft’s version of JavaScript)
■ ActiveX controls
■ ActiveX Documents
■ Java
■ Dynamic HTML (version 4)

Here are the major features you can use if your user’s browser is Netscape Navigator:

■ Java
■ JavaScript
■ Internet Foundation Classes

37.1389-6 CH30 642 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 Developing Intranet Applications 643

■ LiveWire
■ Dynamic HTML (version 4)

As you can see from the preceding lists, there are some similarities, but also some vast differ-
ences in the two major browsers. So, knowing which browser you are developing for can
greatly enhance your development effort.

Part
Sample Applications for the Intranet VI
On the Web, certain applications are seen over and over again: user registration, online order-
ing, and so forth. With intranets, other types of applications can be found on almost every site. Ch
In this section, you will examine some of the major applications that you will (or already have) 30
developed and build a simple framework for them.

Project Management Tool

One of the major uses of an intranet is as a project management tool. This is an application that
enables project teams to record tasks for various team members, record meeting notes, and
other activities related to tracking a project from start to finish.

There will be four major components to this application: 1) a page to create a new project,
2) a page to add, delete, or modify a member of the project team, 3) a page to record, view, and
modify tasks that are assigned to team members, and 4) a page to view and record meeting
notes.

In this section, you will learn how to create pages that enable you to create a project, add mem-
bers to a project, and assign tasks to team members. The other features, such as modifying
members and tasks, will be left as an exercise for the reader.

One of the first design features you need to decide on is which technologies to leverage. To a
very large degree, this will be based on what type of browser is on the network. In this case,
let’s assume it is Internet Explorer 3.02. So, you have a choice between VBScript and JScript
for browser validation. You know what the browser is, so you don’t have to worry about Naviga-
tor browsers. In this case, let’s use VBScript for the client-side data validation.

Next, you need to decide what scripting language to use for your ASP pages. Again, the choice
is between VBScript and Jscript. For consistency sake, let’s again go with VBScript. One of the
most compelling reasons for using VBScript in ASP pages is the wealth of documentation and
examples in VBScript for ASP, although there are more and more JScript examples coming out.

Finally, you can leverage the increased bandwidth of your intranet by including some gratu-
itous graphics and sounds for users to use to personalize the application. This feature doesn’t
really benefit application performance, but it is easy to implement, and makes using the appli-
cation more fun and easy for the user.

Listing 30.1 shows the ASP code to create a new project.

37.1389-6 CH30 643 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 644 Chapter 30 Application Development for the Corporate Intranet/Extranet

Listing 30.1 Adding a new project.

<html>
<head>
<meta http-equiv=”Content-Type”
content=”text/html; charset=iso-8859-1">
<meta name=”GENERATOR”
content=”Microsoft FrontPage (Visual InterDev Edition) 2.0">
<title>Add New Project</title>
</head>

<body bgcolor=”#FFFFFF”>
<form name=”form1" method=”POST”>
<p align=”center”><font size=”6">Add New Project</font></p>

<p align=”center”>Project Name

<input type=”text” size=”20" name=”projectName”></p>
<p>&nbsp;</p>

<p align=”center”>Project Lead ID

<input type=”text” size=”20" name=”projectLead”
<% if Session(“projectLead”) <> “” then
Response.Write(“value=” & Session(“projectLead”))
end if%>>
<A HREF=”empLookup.asp”>Click here</A> to find your ID number</p>

<p align=”center”>Project Description

<input type=”text” size=”20" name=”projectDescription”></p>
<p>&nbsp;</p>

<p align=”center”>
<input type=”submit” name=”Submit” value=”Submit”></p>
</form>
</body>
</html>

Notice how the code asks for the ID of the employee in charge of the project. The user creating
this project might not have any idea what that ID is. So, you are proving them a simple lookup
page where they can find their ID.

This is another example of the great power of the intranet. On the Internet, you would probably
not allow this type of functionality. Imagine you write an Internet-based widget ordering sys-
tem. You might give people a page they can visit to look up the status of their order. However,
you will probably not give them the capability to look up their order.

First of all, this would be a security hole, enabling intruders to find out the status of orders that
are not theirs. The second reason to not enable this is the strain on your server. With Internet
sites, you must assume that the server load could grow exponentially. Allowing a status order
lookup might bring your server over the edge. Listing 30.2 shows the code to add a new em-
ployee.

37.1389-6 CH30 644 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 Developing Intranet Applications 645

Listing 30.2 Adding a new employee.

<html>
<head>
<meta http-equiv=”Content-Type”
content=”text/html; charset=iso-8859-1">
<meta name=”GENERATOR”
content=”Microsoft FrontPage (Visual InterDev Edition) 2.0">
<title>Add New Employee</title> Part
</head>
VI
<body bgcolor=”#FFFFFF”> Ch
<% if Request.Form(“name”) = “” then %>
<form name=”form1" method=”POST”>
30
<p align=”center”><font size=”6">Add Employee</font></p>

<p align=”center”>Employee Name

<input type=”text” size=”20" name=”name”></p>
<p>&nbsp;</p>

<p align=”center”>Phone
<input type=”text” size=”20" name=”phone”></p>
<p>&nbsp;</p>

<p align=”center”>
<input type=”submit” name=”Submit” value=”Submit”></p>
</form>
<% else
Dim name
Dim phone
name = request.form(“name”)
phone = request.form(“phone”)

set objConn = Server.CreateObject(“ADODB.Connection”)

objConn.Open “DSN=ProjectTool;DriverId=25;FIL=MS
Access;MaxBufferSize=512;PageTimeout=5;”

set objRst = Server.CreateObject(“ADODB.Recordset”)

objRst.ActiveConnection = objConn
objRst.LockType = adLockOptimistic
objRst.Source = “tblEmp”
objRst.CursorType = adOpenKeyset
objRst.Open

objRst.AddNew
objRst(“empName”) = name
objRst(“empPhone”) = phone
objRst.Update

‘set the empID in the session not a variable

Session(“empID”) = objRst(“empID”)
objRst.Close
%>

continues

37.1389-6 CH30 645 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 646 Chapter 30 Application Development for the Corporate Intranet/Extranet

Listing 30.2 Continued

The following employee was added:
<pre>
<%=name%>
<%=phone%>
ID #: <%=Session(“empID”)%>
<%end if%>
</body>
</html>

In Listing 30.2, you are adding new employees to the database. Because this system resides on
an intranet, you can also add some interesting functionality. If there exists in the images direc-
tory of the Web server a JPEG called firstNameLastName.jpg, you can display that employee’s
photo whenever you look at his information.

On the Internet, giving each employee a photo might become a huge bandwidth burden. But
on an intranet, it can give you huge benefits. Imagine you are a new employee assigned to a
project. You surf to the page that contains a list of the other employees in the project. By going
to that page, you can see what the other employees you will be working with look like. Listing
30.3 shows the code to add a new task for an employee.

Listing 30.3 Adding a new task.

<html>
<head>
<meta http-equiv=”Content-Type”
content=”text/html; charset=iso-8859-1">
<meta name=”GENERATOR”
content=”Microsoft FrontPage (Visual InterDev Edition) 2.0">
<title>Add New Task</title>
</head>

<body bgcolor=”#FFFFFF”>

<% if Request.Form(“task”) = “” then %>

<form name=”form1" method=”POST”>
<p align=”center”><font size=”6">Add Task</font></p>

<p align=”center”>Task
<input type=”text” size=”20" name=”name”></p>
<p>&nbsp;</p>

<p align=”center”>Status
<input type=”text” size=”20" name=”phone”></p>
<p>&nbsp;</p>

<p align=”center”>Project ID - <A href=”projectlookup.asp”>click

here</A> to look up
<input type=”text” size=”20" name=”phone”></p>
<p>&nbsp;</p>

37.1389-6 CH30 646 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 Developing Intranet Applications 647

<p align=”center”>Employee ID - <A href=”emplookup.asp”>click

here</A> to look up
<input type=”text” size=”20" name=”phone”></p>
<p>&nbsp;</p>

<p align=”center”>
<input type=”submit” name=”Submit” value=”Submit”></p>
</form> Part
<% else
Dim task VI
Dim status
Dim empID Ch
Dim projectID
task = request.form(“task”)
30
status = request.form(“status”)
empID = request.form(“empID”)
projectID = request.form(“projectID”)

set objConn = Server.CreateObject(“ADODB.Connection”)

objConn.Open “DSN=ProjectTool;DriverId=25;FIL=MS
Access;MaxBufferSize=512;PageTimeout=5;”

set objRst = Server.CreateObject(“ADODB.Recordset”)

objRst.ActiveConnection = objConn
objRst.LockType = adLockOptimistic
objRst.Source = “tbltasks”
objRst.CursorType = adOpenKeyset
objRst.Open

objRst.AddNew
objRst(“task”) = task
objRst(“status”) = status
objRst(“empID”) = empID
objRst(“projectID”) = projectID
objRst.Update
objRst.Close
%>

The following task was added:

<pre>
<%=task%><BR>
<%=status%>
<%end if%>
</body>
</html>

In the figures below, you can see what these three pages look like.

In Figure 30.3 you can see the HTML page as it exists to add a new project. Notice that al-
though the layout is simple, there is quite a bit that will need to happen in the background.

37.1389-6 CH30 647 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 648 Chapter 30 Application Development for the Corporate Intranet/Extranet

FIG. 30.3
Adding a new project.

In Figure 30.4, you can see the page that adds a new employee.

Next, check out Figure 30.5 to see what the page looks like to add a new task.

FIG. 30.4
Adding a new
employee.

37.1389-6 CH30 648 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 Building an Extranet 649

FIG. 30.5
Adding a new task.

Part
VI
Ch

30

Another interesting idea you can implement with an intranet-based project tool is to enable the
project leader to create an AVI discussing the project. Management, the project team, and
other employees could browse this AVI. It would give them a nice multimedia presentation of
what that project is all about.

Building an Extranet
An extranet is an application that contains both an intranet component and an external Internet
component. The reasons for building this type of application are obvious; most of them revolve
around communicating to customers’ information that usually requires more expensive medi-
ums, such as mail and freight, or selling materials online, such as wine, books, or what have
you.

However, extranets can be very difficult to develop because of the added level of security that
is necessary. If your extranet is compromised by unscrupulous sorts, you can open yourself up
to lawsuits and lost customers.

Securing Your Extranet from Intruders

There are a number of ways in which to secure your site against intruders. The most popular
are firewalls and dual-homed gateways. Each of these two devices can help to prevent intruders
from breaking into your site by carefully managing the traffic that is allowed in and out at a
very low level.

37.1389-6 CH30 649 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 650 Chapter 30 Application Development for the Corporate Intranet/Extranet

Firewalls are devices that sit in between your Web server and the outside world. They can be
configured to enable only network traffic that is harmless to communicate with your Web
server, and also to enable and deny access based on criteria such as destination address and
protocol. See Figure 30.6 for an example of firewalls preventing unwanted access.

FIG. 30.6 firewall

denying access
Firewalls can prevent
unwanted access.
Your
network

The
Internet

Some hacker

While this device can be a very powerful mechanism for preventing unwanted access, it is not
fool-proof. By using techniques such as spoofing, intruders can still find ways to break past
firewalls and access your data. However, if used correctly, they can provide powerful protection
for your extranet application.

Dual-homed gateways are computers that contain two network cards and sit on two different
networks. Because of the increased intelligence present with a dual-homed host, you can com-
pletely shut down traffic from the outside world except for communication from specifically
designed machines.

Dual-homed hosts give you a huge amount of flexibility, while securing you better than any
other form of protection (next to taking out the network cable). However, they can cause the
traffic between networks to crawl. As with all things, it is a trade off of performance/development
time versus security.

Confidential Information
The other major aspect of securing your extranet is to make sure that the communication link
between the browser and the server is free from prying eyes while confidential information is
being passed.

The primary technique to secure communications is through encryption. Encryption is an

old-fashioned technology. There are several different ways in which you can encrypt the

37.1389-6 CH30 650 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 From Here… 651

communication between your Web server and your user’s browser. The most popular is Secure
Sockets, an encryption method that Netscape developed.

Secure Sockets are based on public key encryption, in which you can encrypt your data from
prying eyes. The way it works is fairly simple.

You have what is called a key pair. With this pair, there is a private key and a public key. Every-
one has access to your public key. With this key, they can encrypt a message they are sending Part
to you. However, only you have your private key, which is necessary to decrypt the message.
The public key is one way—prying eyes cannot use it to decrypt messages sent to you. VI
Ch
Conversely, you can encrypt a message with your private key, and the only way to decrypt it is
to use your public key. At first, this may not seem useful, but in reality, it is a really great way 30
for someone to authenticate that the data is truly from you, otherwise, it would decrypt to
garbage.

The technology to generate these key pairs can vary, but most key pairs are generated using
the RSA public key cryptography. This cryptography is available from RSA Data Security, Inc.,
and Netscape has licensed this cryptography specifically for encryption.

Secure Sockets are easy to set up and they don’t require any additional programming on the
developer’s part. The server and browser take care of everything.

It is simple to implement Secure Sockets on your Web server. The following are the steps
involved:

1. Generate a key pair and a request file.

2. Get a certificate from a certificate authority.
3. Install your certificate on your server.
4. Turn on SSL security on those folders.

From Here…
In this chapter you have examined some of the concepts involved in creating applications for
corporate intranets. You have also seen what an extranet is, and how it can be used to create
powerful and lucrative online solutions.

You have learned how to best take advantage of the increased bandwidth of the intranet. You
have also seen how knowing what browser your users will be using can lower development
time.

Further, you have learned the issues of building a secure extranet application. Understanding
security can mean the prevention of huge headaches later. I encourage you to delve further
into the topic of security as a developer. Too much information has never hurt a programmer,
and knowing too little about security can bite you in the end.

For related information, please see Chapter 31, “Using Internet Information Server 4.0 to Man-
age Your Web Applications,” which will explore this great new Web server from Microsoft and
show you how it will ease your Web application development time and costs! ■

37.1389-6 CH30 651 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 652 Chapter 30 Application Development for the Corporate Intranet/Extranet

37.1389-6 CH30 652 10/26/98, 10:12 AM

Brands03/Stone#3 SEU Active Server Pages #1389-6 SRing 11/13/97 CH30 lp#3
 31
653

C H A P T E R

The Microsoft Management

Using Internet Console
Learn how to use the Microsoft

Information Server 4.0 Management Console to administer

your ASP applications and other
Part
network services.

to Manage Your Web Increase Active Server Page

VI
Ch
Functionality
Applications Review how IIS 4.0 expands the
functionality of Active Server Pages
31
with transactional-based scripts,
process isolation, and new ASP
components.

I nternet Information Server 4.0 consolidates the Microsoft

Transaction Server, Active Server Pages, HTTP, FTP, and
Gopher servers, the Certificate server, the News and
Mails servers, and advanced Web site management into
one manageable interface. Of course, with no bias, one of
the most exciting aspects of IIS 4.0 is the increased real-
world functionality needed for Active Server Pages. IIS
includes transactional-based scripts, the capability to run
ASP applications in their own memory space, and the
capability to activate client- and server-side script debug-
ging capabilities, which finally provides a well-rounded
framework needed to build mission-critical applications. ■

38.1389-6 CH31 653 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 654 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

The Microsoft Management Console

IIS 4.0 provides a centralized network management tool called the Microsoft Management Con-
sole (MMC). From the Management Console, you can manage and configure the HTTP, FTP,
and Gopher servers, the Transaction Server, the Certificate server, and the News and Mails
servers.

Network Service Management

The Microsoft Management Console is a new administration interface that is designed to act as
the common denominator for various network administration needs. The MMC interface is
used to manage network components through “snap-in” technology. A snap-in is an in-process
OLE server that is responsible for managing a specific network interface. At the time of this
writing, the MMC supports a centralized control mechanism over the HTTP, FTP, Gopher,
News, and SMTP Mail servers and the Transaction Server, as shown in Figure 31.1.

FIG. 31.1
The Microsoft Manage-
ment Console uses
snap-in technology to
manage various network
interfaces.

The MMC is divided into two panes. The left pane displays a hierarchical tree-view of snap-in
items, or namespaces. The namespace items can act as objects, containers, or tasks represent-
ing the highest level of the snap-in component. These namespaces enable you to add objects
such as URL links and ActiveX monitoring tools directly into the MMC. The individual items
within the namespaces are called nodes. The detailed information of each node is displayed in
the right pane, or the result pane.

The real power of MMC technology is that you are able to create your own MMCs. These
customized MMCs can be tailored to your own specific administration needs, without risking

38.1389-6 CH31 654 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 655

the stability of an entire mission-critical system. You can use MMC to disperse administration
tasks throughout your organization without giving complete control or full administrator rights
to “network–impaired” users. For example, you can create customized MMCs for different
levels of administration experience for a staff help desk application. As the level of experience
grows, the staff can be given access to advanced MMCs with more powerful administration
and security features. Less experienced users would use MMCs that minimize the potential of
a catastrophic failure from a simple mistake. Figure 31.2 shows a selection of snap-ins you can
choose from.

FIG. 31.2
Use snap-in technology
to build and customize
MMCs to disperse
Part
network administration.
VI
Ch

31

The portability of the MMC is twofold. The MMC files themselves can be transferred to any
machine on which the MMC is installed. By moving the MMC files between machines, admin-
istration tasks can be easily distributed throughout the organization. Second, the MMC can
connect to any other machine where the MMC has been installed to access, control, and man-
age remote namespaces. Of course, the capability to remotely manage an MMC is dependent
on NT security settings defined for your domains.

Besides providing a direct interface to network components, MMCs also provide direct access
to the native NT administration tools, such as the Key Manager, Performance Monitor, Event
Viewer, Server Manager, and User Manager. These tools can be readily accessed when you
need them for the desired service, as shown in Figure 31.3.

The MMC uses property sheets to configure the various services. By right-clicking the service
and selecting the Property item, you can access the settings for the service. For example,
Figure 31.4 demonstrates accessing News Server property settings via the MMC. You can use
these configuration settings in conjunction with the various network management tools, such
as the Performance Monitor, to help troubleshoot and optimize your services. A practical situa-
tion that combines these tools might be if users are complaining that they cannot connect to
the News server. You can use the MMC to activate Performance Monitor to see whether the
maximum number of users for that News server is being reached. From the same MMC inter-
face, you can use the Property tag of the News namespace item to increase the amount of
connections allowed.

38.1389-6 CH31 655 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 656 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

FIG. 31.3
The MMC provides quick
access to networking
administrative tools
such as the Perfor-
mance Monitor.

FIG. 31.4
The MMC provides easy
access to the Add-ins
administration tools and
properties.

Web Site Management

The direct control over the networking components can also be applied to creating and admin-
istering your Web sites. Unlike previous versions of the Internet Information Server, IIS 4.0
enables multiple sites within the same HTTP server. With the previous version of IIS, you were
limited to hosting only one site per IP address. Now you can create multiple sites bound to the
same IP address, as you can see in Figure 31.5.

For example, assume that you create two Web sites on one IIS server bound to the same IP ad-
dress and name the sites engineering and customer support. You can access the engineering
site by using HTTP://engineering as the URL. To access the customer support site, use HTTP:
//customer support.

38.1389-6 CH31 656 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 657

FIG. 31.5
IIS 4.0 enables
multiple sites bound to
a single IP address.

Part
VI
Ch

31
In addition to the capability to create multiple virtual sites within one HTTP server, each indi-
vidual site can have its own unique settings. These settings are independent from other sites
hosted by the same HTTP server. To configure the settings for a particular site, right-click the
Web or FTP site. When you select the Property tag, a tabbed dialog box appears displaying the
current configuration settings for the site, as shown in Figure 31.6.

FIG. 31.6
IIS 4.0 enables each
site on the server to
have its own configura-
tion settings.

The Web Site Property dialog box uses nine tabs to set the configurations of that site. These
settings enable you to control the Web site, security accounts, performance, ISAPI filter, home
directory, documents, HTTP Headers, and custom error settings, as you can see in Table 31.1.

38.1389-6 CH31 657 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 658 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

Table 31.1 The Web Site Configuration Tabs

Tab Name Description
Web Site Sets the IP address and port connections and login capabilities.
Security Accounts Sets the user permissions and remote operator permission.
Performance Establishes performance tuning, bandwidth, and connection
configurations.
ISAPI Filters Manages ISAPI filters for the sites.
Home Directory Sets access permissions and application configurations.
Documents Sets default document and footer information.
Directory Security Sets password authentication, secure communication, and
TCP/IP restrictions.
Custom Errors Defines and sets HTTP error messages.

Of these settings, three tabs pertain to configuring the application settings for your Web-based
applications: the Home Directory, Directory Security, and the ISAPI Filter tabs.

Home Directory The Home Directory tab, as shown in Figure 31.7, is used to set the applica-
tion settings of the site. From this location, you can activate the site to run in its own process
space, set the application permission for the site, and create, configure, or remove and unload
the application from memory.

FIG. 31.7
The Home Directory tab
controls the Web
application settings for
a specific Web site.

If you want to configure your site or virtual directory to run as an application, click the Create
button. When you transform your site or virtual directory to an application that runs within its
own memory space, IIS treats your site as a component and automatically registers the site
object within the Transaction Server, as shown in Figure 31.8.

38.1389-6 CH31 658 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 659

FIG. 31.8
IIS 4.0 treats your site
or virtual directory as a
component and uses
the Microsoft Transac-
tion Server to manage
the application.

Part
VI
Ch

31

Directory Security The Directory Security tab, as shown in Figure 31.9, is used to control
password authentication, secure communication, and TCP/IP access control.

FIG. 31.9
The Directory Security
tab is used to set or
view authentication
methods, access
control permissions,
and secure data
transfer methods for
your applications.

Authentication methods specify how the users will be identified when they access the site. The
Authentication Methods dialog box, as shown in Figure 31.10, enables four different authenti-
cation methods to access the NT file system (NTFS):

■ Allow Anonymous Access

■ Basic Authentication

38.1389-6 CH31 659 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 660 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

■ Windows NT Challenge/Response
■ Enable SSL Client Authentication

FIG. 31.10
IIS 4.0 supports four
different authentication
methods to help identify
users of HTTP connec-
tions.

In order for any HTTP request to be processed, NT checks to see whether the request is being
made by a valid user account with appropriate permissions. To enable anonymous access to
your site, IIS creates a user account named IUSER_computername with Log On Locally user
rights. In most situations, anonymous access is used for general information-sharing situations.

Basic authentication is used when information is limited only to a specific user community
based on NT user accounts. Basic authentication will prompt the user for a valid username and
password. However, keep in mind that the username and password are being passed in clear
text or unencrypted form. This information could be extracted from a variety of network packet
analyzers and cause possible security infractions.

To protect against this possible security violation, Microsoft developed an authentication

method named Windows NT Challenge/Response. The authentication scheme provides an
encrypted data exchange between the Internet Explorer browser and the IIS Web server to
valid NT user accounts.

However, because the Windows NT Challenge/Response is limited to the Internet Explorer

browser, using Secure Socket Layer provides the capability to send username and password
from any browser to the hosting Web server.

In Enable SSL Client Authentication, you have the two options to control permissions to your
site. The SSL Client Certificate Required To Access Resource option requires that the request-
ing browser has a digital certificate. Digital certificates are issued to validate the authenticity of
users. These certificates can be issued from a trusted third party or from your own certificate
server from within your Intranet. The option Mapping a Certificate to an NT User Account will
take the information from the client certificate to find a valid NT user account. This enables
you to control different levels of permissions within the site through NT without having to
change or reissue user certificates.

38.1389-6 CH31 660 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 661

From the Directory Security tab, you can also choose the Secure Communication option to
create a secure channel between the Web server and SSL-complaint browsers. You can further
control access to your sites based on the requesting browser’s IP address. Although limiting
the IP address is not generally used on public Web sites, it is particularly useful for intranet and
extranet applications where business groups are organized by their IP address. For example,
you can regulate permissions to the accounting Web site by accepting only requests from the
account IP sub-mask.

ISAPI Filters The ISAPI Filters tab, as shown in Figure 31.11, is used to add, edit, and re-
move ISAPI filters for your site. Remember that the filters can be applied to a specific site
within the HTTP server. In previous versions of IIS, ISAPI filters could only be applied to the
HTTP service level and not targeted to a specific site.

Part
FIG. 31.11
The ISAPI Filters tab VI
enables ISAPI
Ch
applications to be
applied to a specific 31
Web site.

Configuring Active Server Application Settings

To drill down into the specific configurations of your Active Server Page application, use the
Create or Configuration buttons that were shown in Figure 31.7. From this interface, you can
set the application settings for the site or virtual directory. In particular, you can control the
application mappings, Active Server Pages, Active Server Page debugging, and CGI script
timeout values, as shown in Figure 31.12.

The Application Mappings tab shown in Figure 31.12 enables you to set what application will be
used to process files according to its file extension. For example, files ending with .ASP exten-
sions (Active Server Pages) will be processed by the asp.dll located in the
c:\winnt\system32\inetsrv directory. The Cache ISAPI Application option enables you to
cache the API DLLs. Therefore, when additional .ASPs are called, this path mapping to the
DLL is not necessary. By default, this option is selected.

38.1389-6 CH31 661 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 662 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

FIG. 31.12
IIS 4.0 enables you to
configure and manage
the settings of your
application.

The Active Server Pages tab is used to set the default properties of Active Server Pages when
the server-side scripts are processed. The Active Server Page tab enables you to manage appli-
cation configurations and script file caching (see Figure 31.13).

FIG. 31.13
The default settings for
Active Server Pages can
be configured in this
tab.

The Application configuration section enables you to set the default properties of Active Server
Pages, as shown in Table 31.2.

Table 31.2 Configuring Active Server Page Properties

Page-Level Override
Property Description Variable
Enable Session State Creates a new EnableSessionState
Session object to
track user variables.

38.1389-6 CH31 662 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 663

Page-Level Override
Property Description Variable
Session timeout Sets the timeout Session.TimeOut
value that destroys
the user Session
object.
Enable buffering Determines whether the Response.Buffer
entire page is
completely generated
on the server before
sending the page
to the browser.
Enable Parent Path Sets whether the N/A Part
browser can
navigate up the
VI
directory structure Ch
using the parent
directory (..).
31
Write unsuccessful Used to write N/A
requests to log unsuccessful HTTP
page requests to the
NT event log.
Number of script Sets the maximum N/A
engines cached number of Active
scripting languages
to be cached in
memory.
Default ASP language Sets the default Language
Active Server Page
scripting language.
Script Timeout Sets the amount of Server.ScriptTimeout
time in seconds a
script will process
before terminating.

You can also use this tab to set ASP caching options. When an ASP file is cached, the .ASP
source file is cached, not the results of a processed ASP. Also, you can see performance in-
creases by specifying more memory for caching. To cache all .ASP files, select the Cache All
Requested ASP radio button. To limit the size of cached ASPs in memory, set the desired
amount of memory in megabytes to be assigned for caching. The optimal cache setting de-
pends on the nature of your ASP application. Use the NT Performance Monitor to monitor your
Web server resources to find the optimal settings for your ASP applications.

38.1389-6 CH31 663 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 664 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

N O T E Keep in mind that most of these variables can be overwritten at the page level when the
script is processed. ■

You can override these default ASP settings at the page level to ensure that your scripts are
properly executed if the default settings are changed for the site, or if your application is moved
to another server.

The ASP Debugging tab of the Application Configuration dialog box, as shown in Figure 31.14,
enables you to manage the debugging and script errors messaging of the Internet Information
Server. IIS 4.0 can utilize client and server debugging capabilities using Microsoft’s Script
Debugger. To enable ASP server or client debugging, select the appropriate check box. This
tab also enables you to control the error messages sent to the client. The error messages can
consist of either displaying the detailed message generated by the server or can be a generic
error message that you specify. In most situations, displaying the detailed ASP error message
is recommended as this information can be helpful in identifying the source of the scripting
error.

FIG. 31.14
Use the MMC to activate
client- and server-side
ASP debugging.

The fourth tab of the Application Configuration dialog box, shown in Figure 31.15, enables you
to set the timeout value for Common Gateway Interface (CGI) scripts to expire. The CGI
timeout value is expressed in seconds.

Setting Security Accounts

Another great aspect of the MMC and IIS 4.0 are the remote administration features. As more
and more sites are integrating scripts and application logic, the centralized administration tasks
of network administrators and webmasters have grown exponentially. Because of this growth,
meeting the various configuration needs is often the bottleneck to delivering applications. To
prevent the bottleneck from occurring, the Security Accounts tab, as shown in Figure 31.16,
provides the capability to disperse administration features to other remote operators.

38.1389-6 CH31 664 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 665

FIG. 31.15
Common Gateway
Interface timeout values
are also set in the
Application Configura-
tion dialog box.

Part
VI
Ch
FIG. 31.16
IIS 4.0 offloads administra- 31
tion tasks to remote users.

Now the specified remote users, as designated by NT user accounts, can control the properties
of a site—without having to contact the webmaster—via HTML. This process still enables
network administrators and webmasters to manage sites at a higher level, but it allocates the
administration specifics of a minor site to localized developers or project managers. Figure
31.17 illustrates a remote user using the HTML interface to manage his site.

These remote management capacities are not only exciting for application developers, but can
also be directly applied to Internet service providers. Now Internet service providers can ini-
tially set up a site for a user and enable the user to maintain and manage the specifics of the
site. For example, Figure 31.18 illustrates a remote user applying security settings to his site.

38.1389-6 CH31 665 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 666 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

FIG. 31.17
Remote users can
completely manage
their site via HTML.

FIG. 31.18
Webmasters can deploy
the HTML interface to
manage all the sites on
a server at a higher
level, leaving the
detailed site manage-
ment up to localized site
users.

Creating New Sites and Virtual Directories

The MMC provides wizards to help set up new Web and FTP sites and virtual directories. The
Setup Wizard asks a series of questions that create a site or directory based on information you
provide. To create a new Web or FTP site, right-click the machine you want to create the new

38.1389-6 CH31 666 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 The Microsoft Management Console 667

site on and select Create New. Alternatively, you can choose the Action and Create New
options from the MMC menu. After choosing a new Web or FTP site, the Setup Wizard will
guide you through the site setup, as shown in Figure 31.19.

FIG. 31.19
Creation of Web sites,
FTP sites, and virtual
directories is facilitated
by the use of wizards.

Part
VI
Ch
The Web Site Creation Wizard will ask for the IP address and port number of the Web server,
the path of the home directory, permit anonymous access, and select script permission levels. 31
When you create a new virtual directory, the Configuration dialog window will be slightly dif-
ferent from the Web site configuration window. Because a virtual directory does not need
HTTP service specifications, such as performance configuration, IP address, ISAPI filter, and
remote configuration issues, these options are eliminated from the virtual directory dialog box.
The remaining tags essentially remain the same as the Web site configuration window. This
difference is illustrated in Figure 31.20.

FIG. 31.20
Setting Virtual
Directories directory
options.

Changing the Default Site Values

The use of wizards may help you to create individual sites, but they do not necessarily enable
you to set all the properties needed to run a site. The purpose of any wizard is to provide a
simple interface to gather information that varies from user to user. The wizard does not need

38.1389-6 CH31 667 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 668 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

to prompt for values that are considered constant. These assumed constants are stored in a
master template. You can also change the default settings for the master template for the spe-
cific machine. To change the default settings, select the machine name hosting the HTTP or
FTP service in the MMC. Then select the Edit button under the Master Properties section, as
shown in Figure 31.21.

FIG. 31.21
The default settings for
Web sites and FTP sites
are stored in a master
template file for each
hosting machine.

New Active Server Pages Functionality

With the new release of the Internet Information Server 4.0 comes the advanced functionality
of Active Server Pages. This extended functionality is based on advances in the programming
functionality, development, and debugging environments and new components. Because the
information covered in this section has already been covered in detail in various chapters, this
section will provide a summary of the new features with Active Server Pages.

The Script Debugger

Before IIS 4.0, debugging your Active Server Page code was a very tedious and frustrating job.
With the release of IIS 4.0, Microsoft Script Debugger provides a debugging environment for
both client- and server-side VBScript and JScript. The Script Debugger replicates most of the
functionality found in Visual Basic. This environment, as shown in Figure 31.22, enables you to
set breakpoints, retrieve and set property and object values, and provide different stepping
mechanisms to step through the scripting code. For a closer view at the Script Debugger, see
Chapter 9, “Using the Script Debugger.”

Transactional Scripts, Microsoft Transaction Server, and Isolated

ASP Applications
One of the most significant advances for developing Internet-based applications made
possible with the release of the Internet Information Ser ver 4.0 is the capability to create

38.1389-6 CH31 668 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 New Active Server Pages Functionality 669

transaction-based scripts. Because processing ASP scripts is what really controls the applica-
tion workflow and component’s functionality, the capability to wrap transaction processing
around components at the ASP script level is critical.

FIG. 31.22
IIS 4.0 integrates a
Script Debugger to set,
analyze, and step
through client- and
server-side code.

Part
VI
Ch

31

By separating your applications into components, you benefit from code reuse, a centralized
location for software updates, and the capability to distribute your applications across machine
boundaries to address scalability issues that might plague your applications. Now you can
directly use these components within transactional-based scripts. These components are then
managed by the Transaction Server using the ObjectContext object. The ObjectContext ob-
ject enables you to either commit or abort a processed transaction.
The Transaction Server enables you to isolate individual components, as well as entire ASP
applications, to be executed in their own memory space. The creation of individual runtime
environments protects other Web applications from failing as a result of another component’s
failure. This is particularly valuable in preventing the entire Web server from shutting down as
a result of a single individual Web application suffering catastrophic failure. Furthermore,
process isolation enables you to stop an application and unload its components by stopping and
restarting the Web server.

The Transaction Server is now implemented as a Microsoft Management Console (MMC)

snap-in. The MMC provides a centralized administration point to various network interfaces,
such as the HTTP, FTP, News, Mail, and Transaction Servers. Through the Transaction Server
Explorer, you can completely manage component interaction by viewing the status of various
packages and components, managing individual and groups of objects, setting the startup

38.1389-6 CH31 669 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 670 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

options, and shutting down individual server processes. For more information on using trans-
actions in your ASP scripts, refer to Chapter 12, “Controlling the Server with the Server Object
and Using Transaction Processing with the ObjectContext Object.”

New ASP Components

ASP provides new base components that add to the native scripting functional of the Internet
Information Server. Table 31.3 provides a summary of the installable components.

Table 31.3 The Active Server Page Components

Component Name Description
Browser Capabilities Component Determines the requesting browser’s
characteristics.
Advertisement Rotator Component Generates scrolling advertisement
bars.
Counter Component Creates incremental variables.
Page Counter Component Keeps track the amount of times a page
is requested.
Content Rotator Dynamically changes the content on a
page every time the page is requested
or refreshed.
Content Linking Component Creates a systematic Web page
navigation system.
Active Messaging/SMTP Component Creates an interface into Microsoft
Exchange and other e-mail–based
systems.
Tools Component Provides additional programming
functionality for server-side processing.
Permission Checker Component Checks the user access permissions
for files.
Database Access Component (Active Data Object) Creates a database connection to an
ODBC-compliant database.

For more information on these components, see Chapter 17, “Implementing Existing Server-
Side Components.”

Utilizing New HTTP 1.1 Support

The Internet Information Server 4.0 benefits from four new features of the HTTP 1.1 specifica-
tion. These new features focus on connection functionality and HTTP file management. The
four features can be summarized as shown in Table 31.4.

38.1389-6 CH31 670 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 From Here… 671

Table 31.4 HTTP 1.1 Features

Feature Description
Persistent Connection Keeps the TCP connection between the client and server alive.
Pipelining Enables multiple requests to be made on a connection before
receiving any response from the host.
HTTP PUT Provides update functionality on the Web server.
HTTP DELETE Provides delete functionality on the Web server.

The persistent connection and pipelining features transform the stateless nature of the Web
into a connection-oriented system. Traditionally, HTTP dictated that after a Web server was
finished servicing your request, the connection between the browser and the server was re- Part
leased. The connection is then established with another user and then dropped after the Web VI
server finishes delivered information. However, the constant process of establishing and re- Ch
leasing TPC connections requires a significant amount of resources.

HTTP 1.1 helps to solve the need to establish multiple connections by establishing and main-
31
taining a connection between the browser and the Web server. The capability to maintain a
constant connection is refered to as persistent connection.
Second, HTTP 1.1 also supports pipelining. Pipelining enables multiple requests to be transmit-
ted over the same connection before receiving any responses from the host.

The remaining two features of HTTP 1.1 deliver value to Web server by enabling publishing
and managing files via HTTP. HTTP 1.1 supports HTTP PUT and HTTP DELETE commands. The
HTTP PUT command enables information to be uploaded to a server. The HTTP DELETE com-
mand provides the capability to remove content from the Web server.

The HTTP 1.1 specification was designed to be backward compatible, but in order to use the
new features of HTTP 1.1, both the browser client and Web server must support HTTP 1.1. If
the requesting browser is not HTTP 1.1 compatible, the IIS 4.0 Web server will generate HTTP
1.0 responses. The IIS Web server accomplishes this by scanning the header information sent
from the requesting browser. The Header information identifies the HTTP level of the request-
ing client. If the browser is HTTP 1.1-compliant, the IIS server responds by sending HTTP 1.1
responses. If the browser is 1.0-compliant, the Web server uses HTTP 1.0 to respond to the
client.

From Here…
In this chapter, you can see that the Internet Information Server and Active Server Pages work
in conjunction with the Transaction Server to provide a transactional-based application deploy-
ment mechanism over the Web. The Internet Information Server 4.0 is managed as a snap-in
interface to the Microsoft Management Console. The MMC provides a centralized administra-
tion tool to various network interfaces, such as IIS, FTP News, Mail, and Transaction Servers.

38.1389-6 CH31 671 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 672 Chapter 31 Using Internet Information Server 4.0 to Manage Your Web Applications

The consolidated interface provides property sheets to quickly administer and configure the
network services. The MMC enables you to easily configure your Web site to run as its own
application within its own memory space. The capability to capitalize on process isolation pre-
vents the site from being influenced by other applications and prevents your components from
affecting other applications. Furthermore, the MMC also provides a centralized location to
manage the access permission of your site and well as enable you to tailor the Active Server
Pages runtime environments.

■ Chapter 12, “Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object.”
■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of
tracking a user through an ASP application.
■ Chapter 15, “Retrieving User Information: The Request Object,” demonstrates how to
collect information from the user using forms, hidden fields, and cookies.
■ Chapter 18, “Creating Your Own Active Server Components,” will teach how to create
your own ActiveX components to meet the needs of your applications.

■ Chapter 28, “Working with the Microsoft Transaction Server,” illustrates the use of the
Transaction Server to manage components within your ASP applications.
■ Chapter 32, “Putting It All Together: Creating an N-Tier Online Catalog Application,” will
demonstrate how to use the various features of Active Server Pages to build a multi-tier
Web-based application.

38.1389-6 CH31 672 10/26/98, 10:13 AM

Brands03/Stone03 SEU Active Server Pages #1389-6 SRing 11/13/97 CH31 lp#3
 32
673

C H A P T E R

Building Database
Putting It All Together: Components
Learn how to create ActiveX data-

Creating an N-Tier base components to take advantage

of transaction processing for your
database resources from Active

Online Catalog Server Pages.

Creating Business Rule

Application Servers
Create business rules and ActiveX
components in Visual Basic that can Part
be distributed across your organiza- VI
tion to form the foundation of your
applications. Ch

Register and Distribute Your

32
This chapter represents the beginning steps in transforming Components
your Web-based application into a distributed, component- Learn how to register your compo-
based Web system. This chapter demonstrates this ap- nents using the Microsoft Transac-
proach for the Amazing DataWidget WareHouse online tion Server to manage and distribute
catalog. Specifically, this chapter will demonstrate how to your COM components on local and
convert the database application logic and embedded busi- remote machines.
ness rules demonstrated in Chapter 26 and convert them
into ActiveX components managed by the Microsoft Trans- Deploying ASP and Win32
action Server. These ActiveX components can then be Clients Using the New
processed on the Web server or can be distributed onto Components
remote application servers. In addition, the middle-tier Learn how Active Server Pages and
components can be accessed by Active Server Pages and Win32 clients can both access the
by any COM-compliant Windows application. same functionality from the middle
tier components and how Visual
The new components will be created to contain the data- Basic can be used to test and debug
base access logic that was once embedded in existing Active components.
Active Server scripts. These scripts will perform the
database access and compile into ActiveX Dynamic Link
Libraries (DLLs).

39.1389-6 CH32 673 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 674 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

These database components will be responsible for loading the initial products into the applica-
tion, determining the current inventory of an item, and adding and deleting items from inven-
tory. Furthermore, you will take embedded Active Server application logic to create another
business rules server to validate account numbers before placing an order. After building these
various layers, the components will be installed or registered to use their functionality regard-
less of machine boundaries and network protocol. In addition, this chapter will demonstrate
how these middle-tier business servers can be accessed from any Web browser or any Win32
client. ■

Create the Application Initialization Component

In the previous “Putting It All Together” chapters, the Amazing DataWidgets WareHouse Web
site has demonstrated various coding methods to present the online catalog material to the
potential customers. Chapter 16, “Putting It All Together: Building an Online Catalog,” demon-
strated loading the catalog product listing from values entered directly in ASP, and Chapter 26,
“Putting It All Together: Creating a Dynamic Database Application,” demonstrated extracting
the catalog information from a database. In both situations, you have loaded the variables that
are used throughout the site into application-level variables when the Web application is first
started. This one-time data-load enables all users of the application to have access to the same
data without wasting resources on establishing and maintaining connections to external sys-
tems throughout the application. This loading sequence used the Application_OnStart event
of the GLOBAL.ASA file to load variables and variable arrays into application-level variables.

The fundamental reason for using the one-time data load is to maximize performance by mini-
mizing input/output from the IIS server to other systems, such as database or messaging
systems. This approach provides three advantages in delivering and deploying applications.
First, minimizing I/O through the application reduces system resources that can be better
used for your application. Second, by designing closed looped systems, your applications won’t
be associated with or dependent on poor performance from external systems. Minimizing data
access points provides a more centralized data filter or control mechanism at both design time
and runtime. Providing defined locations in the application where external information can
access data helps maintain strict data-flow, minimize data corruption from incomplete transac-
tion processes, and enables you to better scale software concerns.

Reducing the I/O is considerably more important in a Web-based architecture. Not only are
you sharing the same resources to deliver HTML documents to the browser, but you are also
sharing the same I/O resources the Web server has access to. Therefore, at any one time, your
Internet applications could possibly have tens of thousands concurrent users, all of which are
trying to access a variety of back-end systems. This is why an optimized Input/Output ap-
proach, as shown in Figure 32.1, can have significant effects on your Web-based applications.

With this in mind, in this chapter you will create two ActiveX components using Visual Basic
5.0. These components will be compiled as DLLs to run as in-process, non-visual application
servers that will perform specific tasks. The first custom-built ActiveX component, the
Datawidget.dll, will be used to house several database functions needed for the order pro-
cessing section of the Web site. This component is used to provide insert, update, and delete

39.1389-6 CH32 674 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Creating the Customized Database Access Component 675

methods to the Amazing DataWidgets inventory database. The second ActiveX component, the
DataWidgetProcessing.dll, is responsible for processing and verifying correct account num-
bers. This component does not include any database functionality and is simply used to process
account validation rules.

FIG. 32.1 Load Load Process Workflow Batch Updates

Optimized I/O systems Application Session and and
Variables Variables Application Logic User Output
can have significant
performance effects on
your Web-based
systems.
User User
Data
Interface Interface
Data Store Store

(An optimized system)

Creating the Customized Database Access

Part
Component VI
When a user first enters an application, the application-level variables are loaded into memory
Ch
using the Application_OnStart event. In the original catalog application, the variables are
loaded via VBScript using Active Server Pages. To take advantage of database connection 32
pooling, the initial load sequence will be transferred into the Database Object Component,
which will be created using Visual Basic 5. The database component will contain a variety of
database functionality needed for this application. A specific unit of functionality can be imple-
mented from external objects using a method. Methods are used to trigger embedded function-
ality of an object.

N O T E Keep in mind that you can build the components with any COM-compliant application
development tool, such as Visual Basic, Visual C, Delphi, or PowerBuilder. ■

To begin creating the ActiveX database component, open Visual Basic 5.0. Choose ActiveX
DLL from the New tabbed dialog box when Visual Basic is opened, as shown in Figure 32.2.

If this menu has been disabled, you can create a new ActiveX component by selecting New
from the File menu to display the available Visual Basic templates, as shown in Figure 32.3.

39.1389-6 CH32 675 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 676 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

FIG. 32.2
Visual Basic 5.0 can be
used to create in-
process Dynamic Link
Libraries to house your
application logic.

FIG. 32.3
Use Visual Basic project
templates to create
ActiveX DLLs.

After the class module has been created, Visual Basic uses different windows to organize and
manage projects. Use the Properties window, or press F4, to name the new class module
clsDBinfo. Likewise, name the project DataWidgets. These two names are used to identify the
Application.Class object reference needed by Active Server Pages to use
Server.CreateObject(Application.Class) method.

Visual Basic uses functions and subroutines to create ActiveX DLL methods, which methods
expose an object’s internal functionality. Other COM-compliant objects can then call this func-
tionality. To create a method that returns the information in the Products table, create a public
function called LoadProductsTable. This function will return an array that extracts information
from the Products table. The Products table contains the currently available product informa-
tion from the Amazing DataWidgets WareHouse.

To create the new function, either directly type the syntax on the page, or select Add Proce-
dure from the Tools menu to display the Add Procedure dialog box (see Figure 32.4).

This dialog box will create the following Visual Basic code:
Public Function LoadProductsTable()

End Function

39.1389-6 CH32 676 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Creating the Customized Database Access Component 677

FIG. 32.4
Create your
component’s methods
using functions and
subroutines in Visual
Basic.

Now, you want to expand the functionality of the procedure by adding an input variable to
identify an ODBC name. Furthermore, because you know that you will be returning the results
to Active Server Pages, the function results should be returned as a Variant. These accommo-
dations can be met by changing the function as shown in the following:
Function LoadProductsTable (strODBC) As Variant

End Function

Now you can add the desired application logic to the function. In this situation, you are going
to mimic the Active Server script in the previous version of the Amazing DataWidgets
WareHouse site. Listing 32.1 selects all records in the Products table and transfers selected
members of the recordset into an array that is passed back to the calling page. Part
VI
Listing 32.1 Loading the available product listings with the Ch
LoadProductTable method. 32
Function LoadProductsTable(strODBC) As Variant
‘=== Load the catalog information from the specified data source
Set dataconn = CreateObject(“ADODB.Connection”)
dataconn.Open strODBC
Set cmdRS = CreateObject(“ADODB.RecordSet”)
sql = “SELECT ProductName, Description, UnitPrice, UnitsInStock FROM Products”
cmdRS.Open sql, dataconn, adOpenKeyset, adLockBatchOptimistic
If cmdRS.EOF And cmdRS.BOF Then
‘=== no records
Else
Dim icount As Integer
cmdRS.MoveFirst
Do While Not cmdRS.EOF
icount = icount + 1
cmdRS.MoveNext
Loop
cmdRS.MoveFirst
ReDim aProductArray(icount, 4)
Do While Not cmdRS.EOF
i = i + 1
aProductArray(i, 1) = cmdRS(“ProductName”)
aProductArray(i, 2) = cmdRS(“Description”)
aProductArray(i, 3) = cmdRS(“UnitPrice”)
aProductArray(i, 4) = 0
cmdRS.MoveNext

continues

39.1389-6 CH32 677 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 678 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

Listing 32.1 Continued

Loop
End If
cmdRS.Close
dataconn.Close
LoadProductsTable = aProductArray
End Function

After the function is finished, the class module needs to be compiled as a DLL. But before you
do so, the Data Object Library must be referenced in the project to create the database connec-
tion. The reference can be added to the project by selecting Reference from the Project menu.
A dialog box displaying the currently installed object libraries will be displayed. Select the
Microsoft ActiveX Data Objects 1.5 Library. Now that component has been coded, it is ready to
be compiled. To do this, select Make Datawidget.dll from the File menu and save the new DLL.

Registering the Component

In order for other COM-based programs to use the new DLL, the component DLL must be
registered with the machine that will call the DLL. The DLL itself can be physically located on
the same machine or located on a remote machine. These components can be installed and
registered it two ways. The first and recommended method is to install the components in a
transaction monitor, such as the Microsoft Transaction Server (MTS). The Microsoft Transac-
tion Server provides, as covered in Chapter 28, “Working with the Microsoft Transaction
Server,” straightforward component registration, administration, and configuration features
with runtime resource management features. The second method to register your component
is to use the Regsrv32 and DCOMCNF utilities. The Regsrv32 utility gives you the capability to
register the component on a local or remote machine. The DCOMCNF utility provides the
administration utility to configure the installed components. Note that using the Regsrv32 and
DCOMCNF methods does not supply the runtime resource management of the components
that is supplied when using the Transaction Server. The Regsrv32 and DCOMCNF utilities
rely on the individual components to manage their own runtime resources.

Installing Components on the Microsoft Transaction Server

To optimize the performance and scalability of the Amazing DataWidget WareHouse online
ordering system, the validation and database processing components will be installed in the
Microsoft Transaction Server. The Transaction Server automatically registers the component
when the components are installed on the Transaction Server.

To install the components in MTS, use the Microsoft Management Console to open the Trans-
action Explorer view. The Transaction Explorer provides a hierarchical navigation formation to
manage local and remote components, as shown in Figure 32.5.

39.1389-6 CH32 678 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Registering the Component 679

FIG. 32.5
The Transaction Explorer
provides a hierarchical
administration utility to
manage components.

Part
To install your components, a transaction package must be created first. A package is used to VI
group common components together to provide a central administration utility for the installed
Ch
components. To create a package for your component, highlight the Packages Installed folder
and either select New from the Action menu or right-click the folder and select New. The 32
Transaction Server will prompt you to install a pre-built package or create a blank package.
After selecting the Create Blank Package option, name the package “Amazing DataWidget
WareHouse.” Next, the Package wizard will prompt you for information about what NT user
account to run the packages under. For demonstration purposes, assume you have the proper
rights to run the components, select the Interactive User – The Current Logged On User op-
tion. Click the Finish button, and the Package Wizard will create the new package and display
the package in the Transaction Explorer view.

After the new package is created, two folders are created under the Package folder. The Com-
ponents folder is used to house the individual components, and the Roles folder is used to set
permission and security settings. To install the two components, right-click the Components
folder and select New to activate the Component Wizard, as shown in Figure 32.6.

The Component Wizard will prompt you to install components that are already registered on
the system or to install a new component. After selecting the Install New Component(s) option,
click the Add Files button to select the location of the DLLs, as shown in Figure 32.7.

39.1389-6 CH32 679 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 680 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

FIG. 32.6
Installing the compo-
nents in the Transaction
Server with the
Component Wizard.

FIG. 32.7
Multiple components
can be simultaneously
installed with the
Component Wizard.

After you click the Finish button, the Component Wizard installs the components, returns you
to the Transaction Explorer view, and displays the newly installed components, as shown in
Figure 32.8.

After the components are installed, the Explorer view provides different display formats to
access the properties of the components and packages installed in the MTS. For example, you
can expose a component’s properties or methods by right-clicking the item, as shown in Fig-
ure 32.9.

Now that the components have been installed on the Transaction Server, the Transaction
Server will handle all resource management for the components. The Transaction Server
combines the installation and registration of components with the capability to manage the
runtime component resources. However, you do not have to use the Transaction Server to
remotely install and configure your components. The following section demonstrates how to
register your components without installing the components in the Transaction Server.

39.1389-6 CH32 680 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Registering the Component 681

FIG. 32.8
The installed compo-
nents can now be
administered within the
Transaction Server.

Part
FIG. 32.9 VI
Viewing and adminis-
Ch
tering the package,
components, and
methods properties is
32
accomplished by right-
clicking the item.

39.1389-6 CH32 681 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 682 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

Using the Regsrv32 and DCOMCFG Utilities for Component

Configuration
An alternative to installing and configuring the component with the Transaction Server is to
use the Regsrv32 and DCOMCFG utilities. The Regsvr32.exe utility, found in your system
directory, is used to register the local or remote component. The DCOMCFG utility provides
the administration interface to set and configure the component settings.

To use the Regsrv32 utility, you can use the command prompt or the Run window, as shown in
Figure 32.10.

FIG. 32.10
Use the Regsrv32
command to register
components on local or
remote machines.

For example, the following code registers the new datawidget.dll on the current machine:
C:\WINNT\system32\regsvr32.exe C:\WINNT\system32\datawidget.dll

To register a component located on a remote machine, specify the path to the host machine.
For example, the following code registers a DLL located on a machine named
CompaqAppServer:

C:\WINNT\system32\regsvr32.exe \\CompaqAppServer\system32\datawidget.dll

N O T E In order for components to be shared on remote machines, DCOM must be installed,

enabled, and configured with the appropriate NT security settings. ■

Notice that to register a DLL that has a space in the path, you must wrap the DLL path in quo-
tations, as shown below:
C:\WINNT\system32\regsvr32.exe “C:\Program
Files\DevStudio\VB\datawidgets\datawidget.dll”

Alternatively, if you need to remove the registered component, use the –u flag. For example, to
remove the previous registered datawidget.dll, execute the following line of code:
C:\WINNT\system32\regsvr32.exe –u “C:\Program
Files\DevStudio\VB\datawidgets\datawidget.dll”

This capability to activate remote components is made possible through Distributed COM
(DCOM). Of course, you just can’t start distributing your process loads onto other machines
without proper security and configuration rights. To enable and configure this distributed
processing capability, use the DCOMCNFG utility, as shown in Figure 32.11.

39.1389-6 CH32 682 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Registering the Component 683

FIG. 32.11
Use DCOMCNFG to
administer distributed
component properties.

To start the DCOMCNFG utility, type DCOMCNFG from any command window or select Run
from the Start menu to initiate the Run dialog box. Use the Default Properties tab of the
DCOM Configuration dialog box to set whether DCOM is enabled on the current machine, as
shown in Figure 32.12. Make sure that the option to Enable Distributed COM on This Com-
Part
puter is checked.
VI
FIG. 32.12 Ch
Use the Default
Properties tab of 32
DCOMCNFG to enable
DCOM.

Finally, DCOMCFG enables you to administer DCOM permissions using the Default Security
tab. From here, you can set the general access, launch, and configuration DCOM settings
shown in Figure 32.13.

39.1389-6 CH32 683 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 684 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

FIG. 32.13
The Default Security tab
is used to administer
access, launch, and
configuration permis-
sions of DCOM.

Integrating the New ActiveX Components into

the ASPs
Once the new component has been created and registered, the Active Server scripts have to be
modified to use the component’s methods. After the components have been integrated into the
ASP, you will add error trapping to ensure that the components did establish their proper con-
nections.

Calling the Component from the Active Server Page

Now that the new ActiveX component is properly mapped, the Active Server Pages have to be
modified to call the methods of the new components. You could simply replace the code mod-
ules that used to exist in the GLOBAL.ASA file with references to the new Data Widget’s database
component. However, the Amazing DataWidget WareHouse Web developers want the flexibil-
ity to update the application variables whenever new product information changes in the data-
base without having to restart the Web server or Web site.

To accomplish real-time updates to application variables, a new initialization Active Server Page
is created to update the Products information stored in the application variables. This page
gives the Web developer the flexibility to update the application level without relying on the
Application_OnStart event. The requesting users can be automatically routed to this page
when they enter the Web site. Web developers can simply include this non-visual initialization
page directly in the user workflow to automatically update the application-level variables. For
example, the default site can use redirection methods to call this page automatically when the
root directory is accessed. Alternatively, you can place security settings on this file to only give
the administrators access to this page. Therefore, when the data needs to be refreshed, all the
administrator has to do is browse to the page to load the new application-level variables.

39.1389-6 CH32 684 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Integrating the New ActiveX Components into the ASPs 685

Regardless of how or where the page is called, the new component is called the same way. You
can create an instance by naming the application and class name. Once the object is instanti-
ated, you can reference any of its methods. For example, the following ASP code returns the
Product table information into a local array called aProductArray by calling the
LoadProductsTable method of the DataWidgets datasource.

<%
Application(“strODBC”) = “DataWidgets”
Dim aProductArray
Set myDBObject = Server.CreateObject(“datawidget.clsDBinfo”)
aProductArray = myDBObject.LoadProductsTable(Application(“strODBC”))
%>

Notice that the ODBC name has been added as an application variable.

Creating the Generic Error Form

Now that your custom-created component can be called from Active Server Pages, you want to
be able to trap invalid user input as well as possible component failure. This user and compo-
nent failure is integrated into the Amazing DataWidget WareHouse Web site with the addition
of server-side error trapping. This error screen, as shown in Figure 32.14, is generated by
sending values to an Active Server Page that displays different error types depending on the Part
values passed to it.
VI
Ch
FIG. 32.14
Create generic error 32
screens to accept a
variety of processing
errors.

39.1389-6 CH32 685 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 686 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

Now you will implement some of the error trapping logic that was covered in Chapter 10, “The
Compostition of an Active Server Application.” Error trapping can be enabled by using the On
Error statement. The On Error statement enables you to control how the script will react to an
error. Listing 32.2 demonstrates the use of the Resume Next statement to trap for possible
errors.

Listing 32.2 Using ASP error trapping to ensure the component has been
properly referenced.
<%
Application.Lock
dim aProductArray
Set myDBObject = server.CreateObject(“datawidget.clsDBinfo”)
aProductArray = myDBObject.LoadProductsTable(Application(“strODBC”))

On Error Resume Next

If Err Then
page = Request.ServerVariables(“SCRIPT_NAME”)
errormsg = “A Critical Error has occurred on page “ &page &”. <BR>Please
contact the System Administrator”
urlstr = “error.asp?errortype=” &Server.URLEncode(errortype) & “&errormsg=”
&Server.URLEncode(errormsg)
Response.Redirect urlstr
Else
Application(“aSoftware”) = aProductArray
set atemp = nothing
End If
Application.UnLock

‘=== initial user’s session to be an empty basket

session(“basket”)=Application(“aSoftware”)

Response.Redirect “main.asp”
%>

In Listing 32.2, the script checks to see whether the Err object exists. When an error occurs,
Visual Basic script creates an Err object. If an error does exist, the user is automatically routed
to an error template Active Server Page. Depending on the input, the error box presents itself
in different ways. In this case, the call to the DataWidget component failed and presents a
Critical error message to the user.

If an error does not occur, the application variable aSoftware is transferred to a user’s session
variable named basket. This transfer is used to enable individuals to modify their own orders
without affecting application variables.

Creating the Validation Server

In order to help distribute the processing load away from the Web server, a validation ActiveX
Server is going to be created to process account information. The Amazing DataWidget

39.1389-6 CH32 686 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Creating the Validation Server 687

WareHouse has pre-defined algorithms to help identify and validate accounts. In this situation,
the validation server will validate any nine-digit number beginning with 200 and ending
with 16.

Because this validation server is going to be distributed onto another machine that has been
designated as an application server, a completely separate DLL needs to be created. Create a
new ActiveX DLL project following the same steps as described in the preceding example.
Name the project DataWidgetProcessing and the new class module clsValidate. Next, create
a function named ValidateNumber that accepts an account number to process, as shown in
Figure 32.15.

FIG. 32.15
Use Visual Basic to
create an account
validation server to
transfer processing
power from the Web
server onto remote
machines.
Part
VI
Ch

32

Next, create the code to match the required processing logic. Listing 32.3 demonstrates the
simple validation routine that accepts only nine-digit numbers beginning with 200 and ending
with 16.

Listing 32.3 Creating an account validation method.

Public Function ValidateNumber(num) As Variant
‘=== Performs a validation process on the account number
idNumber = Trim(num)
Dim ret
If IsNumeric(idNumber) Then
If Len(idNumber) = 9 Then
stateID = Left(idNumber, 3)
If stateID = 200 Then

continues

39.1389-6 CH32 687 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 688 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

Listing 32.3 Continued

LocID = Right(idNumber, 2)
Select Case LocID
Case 16
‘=== The pre-approved mailing IDs
ret = True
Case Else
ret = False
End Select
End If
End If
End If
ValidateNumber = ret
End Function

After successfully saving the new DLL, place the new DLL on the application server. To regis-
ter the remote component, use the machine name and path with the regsrv32 command, as
shown below:
C:\WINNT\system32\regsvr32.exe
\\CompaqAppServer\system32\DataWidgetProcessing.dll

Now, this method can be called by referencing an instance of the DataWidgetProcessing com-
ponent. The ValidateNumber method returns True if the account number is valid and returns
False if the account number is invalid. The following code example demonstrates using the
ValidateNumber method.

<%
Set myAccountObject = Server.CreateObject(“datawidgetprocessing.clsvalidate”)
If myAccountObject.validateNumber(Request.Form(“txtaccountnumber”)) Then
‘=== Account is Valid
Else
‘=== Account failed
End If
%>

Creating the Order Processing and Inventory Control

Database Component
The next part of the Amazing DataWidget WareHouse application demonstrates the use of the
database access components created earlier in this chapter to place an order. The program
uses three methods to retrieve, add, and delete items from the inventory database.

Creating the GetProductInventory Method

The GetProductInventory method is used to determine how many items of a specific type are
available in inventory. The amount of inventory in stock is needed to trigger an automatic
warehouse ordering process if the inventory gets below a specific threshold. To create this

39.1389-6 CH32 688 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Creating the Order Processing and Inventory Control Database Component 689

method, reopen the DataWidget Visual Basic project. To add additional methods the existing
component, simply create additional functions. For example, to create the
GetProductInventory function, add the following function to the class module as shown in
Listing 32.4.

Listing 32.4 Creating the GetProductInventory method.

Public Function GetProductInventory(strODBC, strProductName) As Variant
‘=== Retrieves the amount of items in the inventory for a specific item
Set dataconn = CreateObject(“ADODB.Connection”)
dataconn.Open strODBC
Set cmdTmp = CreateObject(“ADODB.Command”)
sql = “SELECT UnitsInStock FROM Products where ProductName= ‘“ &
strProductName & “‘“
Set cmdRS = CreateObject(“ADODB.RecordSet”)
Set cmdTmp.ActiveConnection = dataconn
cmdRS.Open sql, dataconn, adOpenKeyset, adLockBatchOptimistic
If cmdRS.EOF And cmdRS.BOF Then
ret = 0
Else
ret = cmdRS(“UnitsInStock”)
End If
Part
cmdRS.Close
dataconn.Close
GetProductInventory = ret
VI
End Function Ch

32
This listing retrieves how many of the specific product is in stock. To implement the
GetProductInventory method from a client, reference the GetProductInventory method after
an instance of the component has been created:
Set myObject = CreateObject(“datawidget.clsDBinfo”)
txtInventory.Text = myObject.GetProductInventory(“datawidgets”, “Mind-Bender
Control”)

Creating the AddProductInventory Method

The AddProductInventory method is used to add an amount of a specific product to the inven-
tory table (see Listing 32.5).

Listing 32.5 Creating the AddProductInventory method.

Public Function AddProductInventory(strODBC, strProductName, iAmount) As Variant
‘=== Adds an amount of a specific inventory item
Set dataconn = CreateObject(“ADODB.Connection”)
dataconn.Open strODBC
Set cmdTmp = CreateObject(“ADODB.Command”)
sql = “UPDATE Products Set UnitsInStock=UnitsInStock+”
sql = sql & iAmount & “ WHERE ProductName = ‘“ & strProductName & “‘“

continues

39.1389-6 CH32 689 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 690 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

Listing 32.5 Continued

Set cmdRS = CreateObject(“ADODB.RecordSet”)
Set cmdTmp.ActiveConnection = dataconn
cmdRS.Open sql, dataconn, adOpenKeyset, adLockBatchOptimistic
dataconn.Close
If Not Err Then
ret = True
End If
AddProductInventory = ret
End Function

The AddProductInventory method is used to add more inventory for a specific product when
items in the warehouse are restocked. The AddProductInventory method can be called by
using the following code:
Set myObject = CreateObject(“datawidget.clsDBinfo”)
ret = myObject.AddProductInventory(“datawidgets”, “Mind-Bender Control”, 5)

Creating the DeleteProductInventory Method

The DeleteProductInventory method is called when a user places an order and the amount of
items ordered is removed from inventory. This method can be implemented using Listing 32.6.

Listing 32.6 Creating the DeleteProductInventory method.

Public Function DeleteProductInventory(strODBC, strProductName, iAmount) As
Variant
‘=== Remove a specific amount of product from inventory
Set dataconn = CreateObject(“ADODB.Connection”)
dataconn.Open strODBC
Set cmdTmp = CreateObject(“ADODB.Command”)
sql = “UPDATE Products Set UnitsInStock=UnitsInStock-”
sql = sql & iAmount & “ WHERE ProductName = ‘“ & strProductName & “‘“
Set cmdRS = CreateObject(“ADODB.RecordSet”)
Set cmdTmp.ActiveConnection = dataconn
cmdRS.Open sql, dataconn, 0, 1
dataconn.Close
If Not Err Then
ret = True
End If
DeleteProductInventory = ret
End Function

This method can be accessed after creating a reference to the DataWidget component using
the following code:
Set myObject = CreateObject(“datawidget.clsDBinfo”)
ret = myObject.DeleteProductInventory(“datawidgets”, “Mind-Bender Control”, 3)

39.1389-6 CH32 690 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Creating the Order Processing and Inventory Control Database Component 691

Building and Testing Components

After you create the new class module, one of the best ways to test your new DLL is through
another Visual Basic application. To accomplish this, add a new project to the current Visual
Basic workspace by selecting File, Add Project, and then Standard Exe. Visual Basic 5.0 en-
ables you to open multiple projects within one instance of Visual Basic, as shown in Figure
32.16. Notice the Project Group Window in the upper-right corner of Figure 32.16 displays
having a class module project and a forms project open within the same environment.

FIG. 32.16
Visual Basic enables
you to host multiple
projects within one
development environ-
ment.

Part
VI
Ch

32

At first, this capability to house multiple projects may just seem like a convenience. But the real
power of this feature is that it enables you to provide inline-debugging capabilities to your new
component. Now you can place breakpoints within the DataWidget class module to provide in-
line component debugging after the client calls the specific method. This debugging capability
is illustrated in Figure 32.17.

To start testing the new component, add two command buttons and a text box to the form.
These command buttons are going to be used to test the AddProductInventory and
DeleteProductInventory methods. After naming the command buttons, create an instance of
the DataWidget component using the CreateObject syntax. Now any of the instantiated
object’s methods can be referenced. Listing 32.7 demonstrates using different form events to
trigger the three new methods.

39.1389-6 CH32 691 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 692 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

FIG. 32.17
Visual Basic provides
inline-debugging
capabilities for your
components.

Listing 32.7 Creating method calls to the new component in a native Visual
Basic form.

Private Sub btnAdd_Click()

Dim myObject As Object
Set myObject = CreateObject(“datawidget.clsDBinfo”)
ret = myObject.AddProductInventory(“datawidgets”, “Mind-Bender Control”, 5)
txtInventory.Text = myObject.GetProductInventory(“datawidgets”, “Mind-
Bender Control”)
Set myObject = Nothing
End Sub

Private Sub btnDelete_Click()

Dim myObject As Object
Set myObject = CreateObject(“datawidget.clsDBinfo”)
ret = myObject.DeleteProductInventory(“datawidgets”, “Mind-Bender Control”,
3)
txtInventory.Text = myObject.GetProductInventory(“datawidgets”, “Mind-
Bender Control”)
ret = LoadProductsTable(“datawidgets”)
Set myObject = Nothing
End Sub

Private Sub Form_Load()

Dim myObject As Object
Set myObject = CreateObject(“datawidget.clsDBinfo”)
txtInventory.Text = myObject.GetProductInventory(“datawidgets”, “Mind-
Bender Control”)
Set myObject = Nothing
End Sub

39.1389-6 CH32 692 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 Putting It All Together: Filling the Order 693

To run the form application to test your new component, right-click the new project and choose
Select As Startup. When you run the application, the standard EXE project will execute to
display the form you created to test the DataWidget DLL, as shown in Figure 32.18.

FIG. 32.18
Use native VB forms to
test your middle-tier
components.

This Visual Basic application can now access the Datawidget’s methods that you created. This
Visual Basic form accesses the same non-visual, middle-tier component that the ASP and the
IIS server will hit against.

Putting It All Together: Filling the Order Part

Now that you have created the various components to perform account validation and database
manipulation, you can integrate the components together into a logical unit of work. When the
VI
order is placed, the user’s account number is the first item to be processed. If the account Ch
number is not valid, the error page instructs the user to enter a correct account number, as
shown in Figure 32.19.
32

FIG. 32.19
The error page template
displays the invalid field
name and prompts the
user to retry.

The second stage starts to iterate through the user’s basket looking for any items that have
been ordered. When an item is found, the database checks the current amount of that item in
inventory. If the inventory amount is greater than 15, the requested order amount is deducted
from inventory. If the inventory amount is below 15, more items are added to the inventory and
then the requested amount to the product item is deducted from inventory. After all the

39.1389-6 CH32 693 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 694 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

ordered items are processed, the user is routed to the summary page where information about
their order is displayed. Listing 32.8 demonstrates this capability.

Listing 32.8 Placing the order: Integrating the database and validation
components together into one ASP.
<%
Set myAccountObject = Server.CreateObject(“datawidgetprocessing.clsvalidate”)
If myAccountObject.validateNumber(Request.Form(“txtaccountnumber”)) Then
Set myOrderingObject = server.CreateObject(“datawidget.clsDBinfo”)
aUsersBasket=Session(“basket”)
iProductsOrdered = UBound(aUsersBasket,1)
For iCounter = 1 to iProductsOrdered
iUnitQuantity=clng(aUsersBasket(iCounter,4))
If iUnitQuantity > 0 Then
strProductName = aUsersBasket(iCounter,1)
‘=== reserve items
iCurrentProductInventory=myOrderingObject.GetProductInventory(Application(“strODBC”),
strProductName)
If iCurrentProductInventory > 15 Then
‘=== Everything is fine
ret = myOrderingObject.DeleteProductInventory
(Application(“strODBC”), strProductName, iUnitQuantity)
Else
‘=== Add more items
ret = myOrderingObject.AddProductInventory(Application(“strODBC”),
strProductName, 25)
ret = myOrderingObject.DeleteProductInventory (Application(“strODBC”),
strProductName, iUnitQuantity)
End If
End If
Next
Response.Redirect “summary.asp”
Else
dim errortype, returnpage, fieldname, urlstr
errortype=”Retry”
returnpage =”customer.asp”
fieldname =”Account Number”
urlstr=”error.asp?errortype=” &Server.URLEncode(errortype) & “&returnpage=”
&Server.URLEncode(returnpage) & “&fieldname=” &Server.URLEncode(fieldname)
Response.Redirect urlstr
End If
%>

Listing 32.8 first checks to ensure that the account number passes verification using the
DataWidgetProcessing component. If the account number is valid, the script determines the
amount of items in the user’s shopping basket. Next, the amount of existing inventory for that
product is determined to ensure that enough items are in stock to fill the order using the
DataWidget component. If the in-stock inventory is greater than a minimum re-order value, the
products are removed from the inventory using DeleteProductInventory. If not enough inven-
tory is available, the AddProductInventory method restocks the inventory and then the items
are processed.

39.1389-6 CH32 694 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 From Here… 695

Finally, if everything is successful, the summary page is displayed to the user (see Figure
32.20).

FIG. 32.20
A successful order
generates a virtual
receipt.

Part
VI
Ch

If the user enters an invalid account number, the error function is used to prompt the user for 32
valid information, as shown in Figure 32.15.

From Here…
This chapter demonstrated how to create a database access component and an account number
verification ActiveX server. The ActiveX servers are compiled as in-process DLLs and can run
within the same application process space of the Web site or processed on remote machines.
This chapter also illustrated that the same business logic embedded in the ActiveX components
can be accessed by any HTML browser client via Active Server Pages as well as native Win32
applications. The dual role of component architecture enables a diverse user community to
access your pages while allowing advanced functionality of native GUI development tools avail-
able to in-house staff to access the same application logic.

The goal of this chapter is to demonstrate the techniques so that you can build and implement
your own Web-based solutions. At this point, you should be able to look back on the evolution
of the Amazing DataWidget’s WareHouse online catalog system and apply the same model to
fit your specific application and development needs. With the IIS and Active Server Pages, the
potential for developing, deploying, and managing applications over the Web is an exciting
reality.

39.1389-6 CH32 695 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 696 Chapter 32 Putting It All Together: Creating an N-Tier Online Catalog Application

■ Chapter 12, “Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object.”
■ Chapter 13, “Managing the User Session: The Session Object,” illustrates the nuances of
tracking a user through an ASP application.
■ Chapter 15, “Retrieving User Information: The Request Object,” demonstrates how to
collect information from the user using forms, hidden fields, and cookies.
■ Chapter 18, “Creating Your Own Active Server Components,” will teach you how to
create your own ActiveX components to meet the needs of your applications.

■ Chapter 28, “Working with the Microsoft Transaction Server,” illustrates the use of the
Transaction Server to manage components within your ASP applications.

39.1389-6 CH32 696 10/26/98, 10:14 AM

Brands3/stone3 SEUsingActiveServerPgs 1389-6 11.14.97 SRing CH32 LP#3

 APPENDIXES

Appendixes
A VBScript 2.0 Reference 699

B Active Server Object Reference 735

C ADO Reference 761

D Active Server Component Reference 775

40.1389-6 Part VII 697 10/26/98, 10:15 AM

Brands SEU Active Server Pages #1389-6 Part AP SRing 10/28/97 lp#1
40.1389-6 Part VII 698 10/26/98, 10:15 AM

Brands SEU Active Server Pages #1389-6 Part AP SRing 10/28/97 lp#1
 A
699

APPENDIX

App

VBScript 2.0 Reference

VBScript 2.0 is a light-weight scripting language that pro-

vides both server-side and client-side scripting functionality.
VBScript is the default scripting language of the Internet
Information Server, and is used to provide advanced
server-side processing that extends the power and reach
of your Web server. VBScript can be run on the Web
server to process application logic and is used to generate
the standard HTML. As a result of only delivering stan-
dard HTML to the clients, no browsers, user groups, or
operating systems are restricted from receiving informa-
tion from your Web site or Web-enabled applications. ■

41.1389-6 AppA 699 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 700 Appendix A VBScript 2.0 Reference

Scripting Active Server Pages

Dynamically generating HTML based on server-side scripts is made possible by the Active
Server scripting engine. The Internet Information Server provides native scripting support for
VBScript and JScript. To ensure that the proper scripting engine interprets your script, specify
the scripting language for the Active Server Page using the LANGUAGE directive. The LANGUAGE
tag must be made on the first line of code on the Active Server Page, as shown below:
<%@ LANGUAGE=”VBSCRIPT” %>

This example sets VBScript as the page scripting language responsible for interpreting the
script nested within the inline ASP delimiters <% %>. VBScript can be embedded within these
delimiters and directly integrated within the HTML on a page, as shown in the following code:
<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<BODY>
<% If session(“NewUser”) Then %>
<P>Hello, Welcome to the online community news letter</P>
<% Else %>
<P>Welcome back, <% Session(“UserName”) %>! </P>
<% End If%>
</BODY>
</HTML>

VBScript procedures can be implemented in two ways on the Internet Information Server:
either within the ASP delimiters, <% and %>, or by explicitly using the <SCRIPT> tag. To embed
procedures within your ASP delimiters, define the subroutine or procedure as shown below:
<%
Sub CalcNewBalanceRef(ByRef Balance)
Balance = (FormatCurrency(Balance+Balance*22.5/100/12))
Response.Write Balance
End sub
%>

The previous code example creates a subroutine called CalcNewBalance and accepts an argu-
ment named Balance. The procedure can then be accessed from the script by using the Call
statement, as shown in the following example:
<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<BODY>
<% ‘=== Begin Sub ===
Sub CalcNewBalanceRef(Balance)
Balance = (FormatCurrency(Balance+Balance*22.5/100/12))
Response.Write Balance
End Sub
%>
‘==== Begin Body ===
Dim myAcctBalance
myAcctBalance = 3250
Call CalcNewBalanceRef(myAcctBalance)
%>
</BODY>
</HTML>

41.1389-6 AppA 700 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Scripting Active Server Pages 701

You can also create individual procedure calls with the <SCRIPT> tag. However, if this script is
going to run on the server, you need to include the RUNAT attribute to specify where the script
should be executed.
App
<SCRIPT RUNAT = Server LANGUAGE = “VBScript”>
<!--
Function TrapNull(vDBText)
A
Dim Ret
If ISNULL(vDBText) Then
Ret = ””
Else
Ret = Trim(vDBText)
End if
TrapNull = Ret
End Function
- ->
</SCRIPT>

In this example, a function called TrapNull removes null characters by using the IsNull func-
tion. If the data is null, a zero-length string is returned to the calling function. If the value is not
null, the Trim function will eliminate any leading or trailing spaces from the string and return
the value to the calling parent code. For example, the previous example could be called as
shown in the following code:
<TR><TD Align = ”Left”><% Response.Write TrapNull(tbl_user_name) %></TD></TR>

You can also use multiple scripting languages within ASP, as shown below:
<%@ LANGUAGE=”VBSCRIPT” %>
<HTML>
<SCRIPT LANGUAGE = “JScript” RUNAT = SERVER>
function WriteJavaScript(str){
Response.Write(str);
}
</SCRIPT>
<SCRIPT LANGUAGE=”VBScript” RUNAT=SERVER>
Function WriteVBScript(str)
Response.Write(str)
End Function
</SCRIPT>

<BODY>

<%
str = “Hello, Welcome to the online community news letter<BR>”
WriteJavaScript(str)
str = “Today’s top news stories are:<BR>”
WriteVBScript(str)
%>

</BODY>
</HTML>

This previous example demonstrated using the LANGUAGE tag to assign the default scripting
languages that translate the script embedded between the <% %> delimeters and also

41.1389-6 AppA 701 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 702 Appendix A VBScript 2.0 Reference

demonstrates how to add VBScript and JavaScript functions to the same ASP page. This script
will produce the following results:
Hello, Welcome to the online community news letter
Today’s top news stories are:

VBScript can also be implemented as a client-side scripting vehicle, but only for the Internet
Explorer browser. To use VBScript on the client, use the <SCRIPT> tag:
<SCRIPT LANGUAGE=”VBScript”>
<!--
Sub loadAbortRetryIgnor()
dim strMsg, iButton, strTitle, indexid
strMsg = “Could Not Access Database” ‘=== Set the
➥message string
iButton = vbAbortRetryIgnore + vbQuestion + vbDefaultButton2
strTitle = “VBScript Abort, Ignore and Retry Message Box” ‘=== Set the
➥title bar text
indexid = MsgBox(strMsg, iButton, strTitle) ‘=== Trap the
➥results
End Sub
-->
</SCRIPT>

VBScript only has one datatype: the Variant datatype. The Variant is used because of its flex-
ibility with all datatypes. The Variant datatype is unique in the sense that the Variant actually
changes behavior depending on the type of data it is storing. The Variant does use subtypes to
help provide some optimization and datatype-specific functionality. The subtype information is
shown in Table A.1.

Table A.1 The Variant Subtypes

Subtype Description
Empty Variant is un-initialized.
NULL Variant intentionally contains no valid data.
Boolean Contains either True or False.
Byte Contains integer in the range 0 to 255.
Integer Contains integer in the range -32,768 to 32,767.
Currency -922,337,203,685,477.5808 to 922,337,203,685,477.5807.
Long Contains integer in the range -2,147,483,648 to 2,147,483,647.
Single Contains a single-precision, floating-point number in the range
-3.402823E38 to -1.401298E-45 for negative values; 1.401298E-45
to 3.402823E38 for positive values.
Double Contains a double-precision, floating-point number in the range
-1.79769313486232E308 to -4.94065645841247E-324 for negative
values; 4.94065645841247E-324 to 1.79769313486232E308 for
positive values.

41.1389-6 AppA 702 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Operators 703

Subtype Description
Date (Time) Contains a number that represents a date between January 1,
100, and December 31, 9999.
App
String Contains a variable-length string that can be up to approximately
2 billion characters in length. A
Object Contains an object.
Error Contains an error number.

If the variant subtype is identified as Empty, the variable’s value is 0 for numeric variables or a
zero-length string (“”) for string variables.

Operators
VBScript has a wide range of built-in operators that are used to manipulate, control, and com-
pare variables. The VBScript operators are listed in Table A.2.

Table A.2 Operators

Name Symbol Description
Addition Operator + Sums two numbers.
And Operator & Performs a logical conjunction on two
expressions.
Concatenation Operator & Forces string concatenation of two
expressions.
Division Operator / Divides two numbers.
Eqv Operator EQV Performs a logical equivalence on two
expressions.
Exponentiation Operator ^ Raises a number to the power of an
exponent.
Imp Operator IMP Performs a logical implication on two
expressions.
Integer Division Operator \ Divides two numbers and returns an
integer result.
Is Operator IS Compares two object reference variables.
Mod Operator MOD Divides two numbers and returns only
the remainder.
Multiplication Operator * Multiplies two numbers.

continues

41.1389-6 AppA 703 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 704 Appendix A VBScript 2.0 Reference

Table A.2 Continued

Name Symbol Description
Negation Operator - Finds the difference between two
numbers or sets the negative value of a
number.
Not Operator NOT Performs logical negation on an
expression.
Or Operator OR Performs a logical disjunction on two
expressions.
Subtraction Operator - Finds the difference between two
numbers or sets the negative value of a
number.
Xor Operator XOR Used to perform a logical exclusion on
two expressions.

Statements and Conditional Statements

VBScript uses statements to define and help organize coding structures. Conditional state-
ments are used to control application flow and program execution. Table A.3 displays the decla-
ration statements and conditional statements found in VBScript 2.0.

Table A.3 Statements

Name Description
Call Transfers program execution to a subroutine or function.
Const Declares a constant’s value to use in place of literal values.
Dim Declares variables and allocates storage space.
Do...Loop Repeats a block of statements while a condition returns True.
Erase De-allocates dynamic array storage space.
Exit Exits a procedure or conditional statement.
For...Next Repeats a group of code statements a specified number of
times.
For Each...Next Repeats a group of code statements for each element in an
array or collection.
Function Declares a function procedure.
If...Then...Else Conditionally executes a group of statements.

41.1389-6 AppA 704 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 705

Name Description
On Error Enables error-handling.
Option Explicit Forces explicit declaration of all variables. App
Private Used to declare and allocate storage space for private A
variables.
Public Used to declare and allocate storage space for public variables.
Randomize Initializes the random-number generator.
ReDim Declares dynamic array variables and allocates or reallocates
storage space.
Rem Used to add comments or remarks within the code.
Select Case Executes one of several groups of statements, depending on
the value of an expression.
Set Assigns an object reference to a variable or property.
Sub Declares subroutine procedure.
While...Wend Executes a series of statements as long as a given condition is
True.

Intrinsic Functions
VBScript provides access to compiled application logic built into the VBScript programming
language by the use of functions. These functions usually accept information, process the infor-
mation, and return the results of the function to the calling line of code. Table A.4 can be used
to help identify a function name depending on its functionality.

Table A.4 VBScript 2.0 Function Categories

Category or Topic Function Name
Array handling Array, IsArray, Erase, LBound, UBound

Conversions Abs, Asc, Chr, CBool, Cbyte, CCur, Cdate, CDbl, Cint, CLng,
Sng, CStr, DateSerial, DateValue, Hex, Oct, Fix, Int, Sgn,
TimeSerial, TimeValue

Dates/times Date, Time, DateAdd, DateDiff, DatePart, DateSerial,

DateValue, Day, Month, MonthName, Weekday, WeekdayName,
Year, Hour, Minute, Second, Now, TimeSerial, TimeValue

Formatting strings FormatCurrency, FormatDateTime, FormatNumber,

FormatPercent

continues

41.1389-6 AppA 705 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 706 Appendix A VBScript 2.0 Reference

Table A.4 Continued

Category or Topic Function Name
Input/output InputBox, LoadPicture, MsgBox

Math Atn, Cos, Sin, Tan, Exp, Log, Sqr, Rnd

Script engine ID ScriptEngine, ScriptEngineBuildVersion,

ScriptEngineMajorVersion, ScriptEngineMinorVersion

Strings processing Asc, Chr, Filter, Instr, InstrRev, Join, Len, LCase, Ucase,
Left, Mid, Right, Replace, Space, Split, StrComp, String,
StrReverse, LTrim, RTrim, Trim

Variants processing IsArray, IsDate, IsEmpty, IsNull, IsNumeric, IsObject,

TypeName, VarType

For a complete alphabetical listing of VBScript functions and their descriptions, see Table A.5.

Table A.5 Alphabetical Listing of VBScript 2.0 Functions

Function Name Description
Abs Returns the absolute value of a number.
Array Returns a Variant containing an array.
Asc Returns the ANSI character code of a string.
Atn Returns the arctangent of a number.
CBool Returns an expression that has been converted to a Boolean
subtype.
CByte Returns an expression that has been converted to a Byte
subtype.
CCur Returns an expression that has been converted to a Currency
subtype.
CDate Returns an expression that has been converted to a Date
subtype.
CDbl Returns an expression that has been converted to a Double
subtype.
Chr Returns the character associated with the specified ANSI
character code.
CInt Returns an expression that has been converted to an Integer
subtype.
CLng Returns an expression that has been converted to a Long
subtype.

41.1389-6 AppA 706 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 707

Function Name Description

Cos Returns the cosine of an angle.
CreateObject Creates and returns a reference to an Automation object. App
CSng Returns an expression that has been converted to a Single A
subtype.
CStr Returns an expression that has been converted to a String
subtype.
Date Returns the current system date.
DateAdd Returns a date to which a specified time interval has been
added.
DateDiff Returns the number of intervals between two dates.
DatePart Returns the specified part of a given date.
DateSerial Returns a Variant of subtype Date for a specified year, month,
and day.
DateValue Returns a Variant of subtype Date.
Day Returns a whole number between 1 and 31, inclusive, represent-
ing the day of the month.
Exp Returns e (the base of natural logarithms) raised to a power.
Filter Returns a zero-based array containing a subset of a string array.
Fix Returns the integer portion of a number.
FormatCurrency Returns an expression formatted as a currency value.
FormatDateTime Returns an expression formatted as a date or time.
FormatNumber Returns an expression formatted as a number.
FormatPercent Returns an expression formatted as a percent value with a
trailing percent (%) symbol.
Hex Returns a string representing the hexadecimal value of a
number.
Hour Returns a whole number representing the hour of the day
between 0 and 23.
InputBox Displays a prompt for an input dialog box.
InStr Returns the numeric position of the first occurrence of one
string within another.
InStrRev Returns the numeric position of one string within another string
from the end of string.
continues

41.1389-6 AppA 707 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 708 Appendix A VBScript 2.0 Reference

Table A.5 Continued

Function Name Description
Int Returns the integer portion of a number.
IsArray Returns a Boolean value indicating if a variable is an array.
IsDate Returns a Boolean value indicating if an expression can be
converted to a date.
IsEmpty Returns a Boolean value indicating whether a variable has been
initialized.
IsNull Returns a Boolean value that indicates whether an expression
contains no valid datatypes.
IsNumeric Returns a Boolean value indicating whether an expression can
be evaluated as a number.
IsObject Returns a Boolean value indicating whether an expression
refers to an Automation object.
Join Returns a string created by joining strings stored in an array.
LBound Returns the base index value for a dimension of an array.
LCase Returns a string that has been converted to lowercase.
Left Returns the specified number of characters from the left side of
a string.
Len Returns the number of characters in a string.
LoadPicture Returns a reference to a picture object.
Log Returns the natural logarithm of a number.
LTrim Returns a copy of a string without leading spaces.
Mid Returns a specified number of characters from a string.
Minute Returns the amount of the minutes in the current system’s time.
Month Returns the month of the year.
MonthName Returns a string identifying the specified month.
MsgBox Displays a message in a dialog box.
Now Returns the current system date and time.
Oct Returns a string representing the octal value of a number.
Replace Returns a string in which a specified sub-string has been
replaced with another sub-string a specified number of times.
Right Returns a specified number of characters from the right side of
a string.

41.1389-6 AppA 708 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 709

Function Name Description

Rnd Returns a random number.
Round Returns a number rounded to a specified number of decimal App
places.
A
RTrim Returns a copy of a string without trailing spaces.
ScriptEngine Returns the scripting language in use.
ScriptEngine Returns the build version number of the script engine in use.
BuildVersion

ScriptEngine Returns the major version number of the script engine in use.
MajorVersion

ScriptEngine Returns the minor version number of the scripting language

MinorVersion in use.
Second Returns the seconds value of the current system time.
Sgn Returns an integer indicating whether a number is positive,
negative, or zero.
Sin Returns the sine of an angle.
Space Returns a string consisting of the specified number of spaces.
Split Returns a zero-based, one-dimensional array containing a
specified number of sub-strings.
Sqr Returns the square root of a number.
StrComp Returns a value indicating the result of a string comparison.
StrReverse Returns a string in which the character order of a specified
string is reversed.
String Returns a repeating character string of the length specified.
Tan Returns the tangent of an angle.
Time Returns the current system time.
TimeSerial Returns the time for a specific hour, minute, and second.
TimeValue Returns the current time.
Trim Returns a copy of a string without leading and trailing spaces.
TypeName Returns subtype information about a variable.
UBound Returns the largest available subscript for a dimension of an
array.
UCase Returns a string that has been converted to uppercase.
VarType Returns the subtype of a variable.

continues

41.1389-6 AppA 709 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 710 Appendix A VBScript 2.0 Reference

Table A.5 Continued

Function Name Description
Weekday Returns a whole number representing the day of the week.
WeekdayName Returns the specified day of the week.
Year Returns the current year.

The following section demonstrates how to use each function by exposing the syntax of the
VBScript 2.0 intrinsic functions.

The Abs() Function

Description: Returns the absolute value of a number.

Syntax: Abs(number)

Arguments: number is any numeric expression.

The Array() Function

Description: Returns a Variant containing an array.

Syntax: Array(clist)

Arguments: clist is a comma-delimited list of values to add to the array.

The Asc() Function

Description: Returns the ANSI character code of a string.

Syntax: Asc(string)

Arguments: string is any valid string expression.

The Atn() Function

Description: Returns the arctangent of a number.

Syntax: Atn(number)

Arguments: number is any valid numeric expression.

The CBool() Function

Description: Returns an expression that has been converted to a Boolean subtype.

Syntax: CBool(expression)

Arguments: expression is any valid expression.

41.1389-6 AppA 710 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 711

The CByte() Function

Description: Returns an expression that has been converted to a Byte subtype.

Syntax: CByte(expression) App

Arguments: expression is any valid expression. A

The CCur() Function
Description: Returns an expression that has been converted to a Currency subtype.

Syntax: CCur(expression)

Arguments: expression is any valid expression.

The CDate() Function

Description: Returns an expression that has been converted to a Date subtype.

Syntax: Cdate(expression)

Arguments: expression is any valid date expression.

The CDbl() Function

Description: Returns an expression that has been converted to a Double subtype.

Syntax: CDbl(expression)

Arguments: expression is any valid expression.

The Chr() Function

Description: Returns the character specified by the ANSI character code.

Syntax: Chr(ansicode)

Arguments: ansicode is the ANSI character code.

The CInt() Function

Description: Returns an expression that has been converted to an Integer subtype.

Syntax: CInt(expression)

Arguments: expression is any valid expression.

The CLng() Function

Description: Returns an expression that has been converted to a Long subtype.

Syntax: CLng(expression)

Arguments: expression is any valid expression.

41.1389-6 AppA 711 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 712 Appendix A VBScript 2.0 Reference

The Cos() Function

Description: Returns the cosine of an angle.

Syntax: Cos(number)

Arguments: number is any valid numeric expression.

The CreateObject() Function

Description: Creates and returns a reference to ActiveX Automation object.

Syntax: CreateObject(objname)

Arguments: objname is any valid ActiveX Automation object.

The CSng() Function

Description: Returns an expression that has been converted to a Variant of subtype Single.

Syntax: CSng(expression)

Arguments: expression is any valid expression.

The CStr() Function

Description: Returns an expression that has been converted to a Variant of subtype String.

Syntax: CSng(expression)

Arguments: expression is any valid expression.

The Date() Function

Description: Returns the current system date.

Syntax: Date()

Arguments: none

The DateAdd() Function

Description: Returns a date to which a specified time interval has been added.

Syntax: DateAdd(timeinterval, number, date)

Arguments: timeinterval is the time interval to add (as specified in Table A.6); number is
amount of time intervals to add; and date is the starting date.

41.1389-6 AppA 712 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 713

Table A.6 Time Intervals

Setting Description
YYYY Year App

Q Quarter A
M Month
D Day
W Weekday
WW Week
H Hour
M Minute
S Second

The DateDiff() Function

Description: Returns the number of intervals between two dates.

Syntax: DateDiff(timeinterval, date1, date2 [,firstdayofweek[, firstweekofyear]])

Arguments: timeinterval is the time interval to add (as specified in Table A.6); date1 and
date2 are valid date expressions to be evaluated; and firstdayofweek and firstweekofyear
are optional values to specify the first day of the week and first week of the year.

The DatePart() Function

Description: Returns the specified part of a given date.

Syntax: DatePart(timeinterval, date[, firstdayofweek[, firstweekofyear]])

Arguments: timeinterval is the time interval to add (as specified in Table A.6); date is a valid
date expression; firstdayofweek and firstweekofyear are optional values to specify the first
day of the week and first week of the year.

The DateSerial() Function

Description: Returns a Date subtype for a specified year, month, and day.

Syntax: DateSerial(year, month, day)

Arguments: year is the numeric expression representing the year; month is the numeric ex-
pression representing the month; day is the numeric expression representing the day.

41.1389-6 AppA 713 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 714 Appendix A VBScript 2.0 Reference

The DateValue() Function

Description: Returns a Date subtype.

Syntax: DateValue(date)

Arguments: date is any valid date expression.

The Day() Function

Description: Returns a whole number representing the day of the month.

Syntax: Day(date)

Arguments: date is any valid date expression.

The Exp() Function

Description: Returns e, the base of natural logarithms, raised to a power.

Syntax: Exp(number)

Arguments: number is any valid numeric expression.

The Filter() Function

Description: Returns a zero-based array containing subset of a string array.

Syntax: Filter(InputStrings, Value[, bInclude[, Compare]])

Arguments: InputStrings is a one-dimension string array; Value is the value being searched;
bInclude (optional) can be True or False. If set to True, the return array contains values found.
False returns an array containing elements not matching the search. Compare is an optional
value indicating a comparison constant.

The Fix() Function

Description: Returns the first negative integer greater than or equal to number.

Syntax: Fix(number)

Arguments: number is any valid numeric expression.

The FormatCurrency() Function

Description: Returns an expression formatted as a currency value.

Syntax: FormatCurrency(Expression[,iDigits [,bLeadingDigit [,bParen

[,bGroupDigits]]]])

Arguments: Expression is a valid numeric expression; iDigits is an optional numeric value

used to indicate number of digits to the right of the decimal point; bLeadingDigit is an optional

41.1389-6 AppA 714 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 715

tristate value used to display a leading zero; bParen is an optional tristate value used to display
parentheses around negative values; and bGroupDigits is an optional tristate value used to
display a number as specified in the group delimiter settings of the Control Panel’s regional
settings. The Tristate values are displayed in Table A.16 App

A
The FormatDateTime() Function
Description: Returns an expression formatted as a date or time.

Syntax: FormatDateTime(Date[,NamedFormat])

Arguments: Date is any valid date expression, and NamedFormat is an optional date/time con-
stant in Table A.10

The FormatNumber() Function

Description: Returns an expression formatted as a number.

Syntax: FormatNumber(Expression[,iDigits [,bLeadingDigit [,bParen

[,bGroupDigits]]]])

Arguments: Expression is a valid numeric expression; iDigits is an optional numeric value

used to indicate the number of digits to the right of the decimal point; bLeadingDigit is an
optional tristate value used to display a leading zero; bParen is an optional tristate value used to
display parenthesis around negative values; and bGroupDigits is an optional tristate value used
to display a number as specified in the regional settings of the Control Panel.

The FormatPercent() Function

Description: Returns an expression formatted as a percent value with a trailing percent (%)
symbol.

Syntax: FormatPercent(Expression[,iDigits [,bLeadingDigit [,bParen

[,bGroupDigits]]]])

Arguments: Expression is a valid numeric expression; iDigits is an optional numeric value

used to indicate the number of digits to the right of the decimal point; bLeadingDigit is an
optional tristate value used to display a leading zero; bParen is an optional tristate value used to
display parentheses around negative values; and bGroupDigits is an optional tristate value
used to display numbers as specified in the group delimiter settings of the Control Panel’s
regional settings.

The Hex() Function

Description: Returns a string representing the hexadecimal value of a number.

Syntax: Hex(number)

Arguments: number is any valid numeric expression.

41.1389-6 AppA 715 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 716 Appendix A VBScript 2.0 Reference

The Hour() Function

Description: Returns a whole number representing the hour of the day between 0 and 23.

Syntax: Hour(time)

Arguments: time is any valid date/time expression.

The InputBox() Function

Description: Displays a prompt for an input dialog box.

Syntax: InputBox(promptmsg[, title][, default][, xpos][, ypos][, helpfile,

context])

Arguments: promptmsg is the message string to be displayed in the dialog box; title (op-
tional) is the string expression displayed in the title of the dialog box; default (optional) is the
string expression displayed in the text box of the dialog box; xpos (optional) is number of twips
from the left side of the dialog box to the left of the screen; ypos (optional) is the number of
twips from the top of the dialog box to the top of the screen; helpfile (optional) specifies the
location of a help file; and context (optional) specifies the help context number in helpfile.
Twips is a graphical constant used to set the ScaleMode property of an object

The InStr() Function

Description: Returns the numeric position of the first instance of one string within another.

Syntax: InStr([start, ]strSearchme, strsearchfor[, compare])

Arguments: start (optional) is the numeric position to start the string search; strSearchme is
the string expression to be searched; strsearchfor is the string expression search value; and
compare (optional) is the value indicating the comparison constant, as described in Table A.8.

The InStrRev() Function

Description: Returns the numeric position of one string within another starting from the end
of the string.

Syntax: InStrRev(strSearchme, strsearchfor [, start[, compare]])

Arguments: strSearchme is the string expression to be searched; strsearchfor is the string

expression search value; start (optional) is the numeric position to start the string search; and
compare (optional) is the value indicating the comparison constant, as described in Table A.8.

The Int() Function

Description: Returns the integer portion of a number.

Syntax: Int(number)

Arguments: number is any valid numeric expression.

41.1389-6 AppA 716 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 717

The IsArray() Function

Description: Returns a Boolean value indicating whether a variable is an array.

Syntax: IsArray(vName) App

Arguments: vName is the name of the variable to be determined. A

The IsDate() Function
Description: Returns a Boolean value indicating whether the expression can be converted to a
date.

Syntax: IsDate(expression)

Arguments: expression is any valid expression.

The IsEmpty() Function

Description: Returns a Boolean value indicating whether a variable has been initialized.

Syntax: IsEmpty(expression)

Arguments: expression is any valid expression.

The IsNull() Function

Description: Returns a Boolean value that indicates whether an expression contains no valid
datatype.

Syntax: IsNull(expression)

Arguments: expression is any valid expression.

The IsNumeric() Function

Description: Returns a Boolean value indicating whether an expression can be evaluated as a
number.

Syntax: IsNumeric(expression)

Arguments: expression is any valid expression.

The IsObject() Function

Description: Returns a Boolean value indicating whether an expression refers to an
Automation object.

Syntax: IsObject(expression)

Arguments: expression is any valid expression.

41.1389-6 AppA 717 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 718 Appendix A VBScript 2.0 Reference

The Join() Function

Description: Returns a string created by joining strings stored in an array.

Syntax: Join(alist[, delimiter])

Arguments: aList is the name of a one-dimensional array, and delimiter (optional) is the
string delimiters used to separate the elements within the string. The default value is the space
(““) character.

The LBound() Function

Description: Returns the base index value for a dimension of an array.

Syntax: LBound(arrayname[, dimension])

Arguments: arrayname is the name of any array, and dimension (optional) is a number indicat-
ing the dimension to find the lower bound.

The LCase() Function

Description: Returns a string that has been converted into lowercase characters.

Syntax: LCase(string)

Arguments: string is any valid string expression.

The Left() Function

Description: Returns the number of characters from the left side of a string.

Syntax: Left(string, length)

Arguments: string is any valid string expression, and length is the number of characters to
return.

The Len() Function

Description: Returns the number of characters in a string or the number of bytes required to
store a variable.

Syntax: Len(string | varname)

Arguments: string is any valid string expression; varname is any valid variable name.

The LoadPicture() Function

Description: Returns a reference to a picture object.

Syntax: LoadPicture(picturename)

Arguments: picturename is the name of the picture filename.

41.1389-6 AppA 718 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 719

The Log() Function

Description: Returns the natural logarithm of a number.

Syntax: Log(number) App

Arguments: number is any valid numeric expression greater than 0. A

The LTrim() Function
Description: Returns a string without leading spaces.

Syntax: LTrim(string)

Arguments: string is any valid string expression.

The Mid() Function

Description: Returns a specified number of characters from a string.

Syntax: Mid(string, start[, length])

Arguments: string is any valid string expression; start is a numeric character position to
begin extraction from; and length (optional) is the number of characters to return.

The Minute() Function

Description: Returns the number of the minutes in current system time.

Syntax: Minute(time)

Arguments: time is any valid time expression.

The Month() Function

Description: Returns the number of the month of the year.

Syntax: Month(date)

Arguments: date is any valid date expression.

The MonthName() Function

Description: Returns a string identifying the specified month.

Syntax: MonthName(month[, bAbbreviate])

Arguments: month is the numeric representation for a given month; bAbbreviate (optional) is
a Boolean value used to display month abbreviations. True will display the abbreviated month
name, and False (default) will not show the abbreviation.

41.1389-6 AppA 719 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 720 Appendix A VBScript 2.0 Reference

The MsgBox() Function

Description: Displays a message in a dialog box.

Syntax: MsgBox(prompt[, buttons][, title][, helpfile, context])

Arguments: buttons is an optional numeric expression indicating button style to display (see
Table A.13 for messagebox constants); title (optional) is a string expression that is displayed
in the title bar of the message box; helpfile (optional) specifies the location of the help file;
and context (optional) specifies the help context number in the help file.

The Now() Function

Description: Returns the current system date and time.

Syntax: Now()

Arguments: None

The Oct() Function

Description: Returns a string representing the octal value of a number.

Syntax: Oct(number)

Arguments: number is any valid numeric expression.

The Replace() Function

Description: Returns a string in which a specified sub-string has been replaced with another
substring a specified number of times.

Syntax: Replace(strSearchMe, strSearchFor, strReplaceWith [, start[, count[,

compare]]])

Arguments: strSearchMe is a string expression containing a sub-string to be replaced;

StrSearchFor is the string expression to search for within strSearchMe; strReplaceWith is the
string expression to replace sub-string strSearchFor; start (optional) is the numeric charac-
ter position to begin search; count (optional) is the numeric amount of time to replace values;
and compare (optional) is a value indicating the comparison constant.

The Right() Function

Description: Returns a specified number of characters from the right side of a string.

Syntax: Right(string, length)

Arguments: string is any valid string expression, and length is any valid numeric expression
representing the number of characters to return.

41.1389-6 AppA 720 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 721

The Rnd() Function

Description: Returns a random number.

Syntax: Rnd[(number)] App

Arguments: number is any valid numeric expression. A

The Round() Function
Description: Returns a number rounded to a specified number of decimal places.

Syntax: Round(expression[, numRight])

Arguments: expression is any valid numeric expression to be rounded, and NumRight (op-
tional) is any numeric expression use to indicate the number of digits to the right of the deci-
mal point.

The RTrim() Function

Description: Returns a copy of a string without trailing spaces.

Syntax: RTrim(string)

Arguments: string is any valid string expression.

The ScriptEngine() Function

Description: Returns the scripting language in use.

Syntax: ScriptEngine()

Arguments: None.

The ScriptEngineBuildVersion() Function

Description: Returns the build version number of the script engine in use.

Syntax: ScriptEngineBuildVersion()

Arguments: None.

The ScriptEngineMajorVersion() Function

Description: Returns the major version number of the script engine in use.

Syntax: ScriptEngineMajorVersion()

Arguments: None.

41.1389-6 AppA 721 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 722 Appendix A VBScript 2.0 Reference

The ScriptEngineMinorVersion() Function

Description: Returns the minor version number of the scripting language in use.

Syntax: ScriptEngineMinorVersion()

Arguments: None.

The Second() Function

Description: Returns the current seconds value of the current system time.

Syntax: Second(time)

Arguments: time is any valid time expression.

The Sgn() Function

Description: Returns an integer indicating the sign of a number.

Syntax: Sgn(number)

Arguments: number is any valid numeric expression.

The Sin() Function

Description: Returns the sine of an angle.

Syntax: Sin(number)

Arguments: number is any valid numeric expression that can express an angle in radians.

The Space() Function

Description: Returns a string expression consisting of the number of defined spaces.

Syntax: Space(number)

Arguments: number is any valid numeric expression.

The Split() Function

Description: Returns a zero-based, one-dimensional array containing elements constructed
from a string expression.

Syntax: Split(expression[, delimiter[, count[, compare]]])

Arguments: expression is a string expression to build the one-dimensional array; delimiter

(optional) is the string delimiter used to separate elements; count (optional) is the number of
elements to return; and compare (optional) is the value indicating the comparison constant, as
described in Table A.8.

41.1389-6 AppA 722 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Intrinsic Functions 723

The Sqr() Function

Description: Returns the square root of a number.

Syntax: Sqr(number) App

Arguments: number is any valid numeric expression greater than or equal to zero. A
The StrComp() Function
Description: Returns a value indicating the result of a string comparison.

Syntax: StrComp(string1, string2[, compare])

Arguments: string1 and string2 are any valid string expressions to be compared, and
compare (optional) is the value indicating the comparison constant, as described in Table A.8.

The StrReverse() Function

Description: Returns a string where the character order has been reversed.

Syntax: StrReverse(string)

Arguments: string is any valid string expression.

The String() Function

Description: Returns a repeating character string of the length specified.

Syntax: String(number, character)

Arguments: number is any valid numeric expression greater than zero, and character is the
ANSI character code used to build the string.

The Tan() Function

Description: Returns the tangent of an angle.

Syntax: Tan(number)

Arguments: number is any valid numeric expression that can express an angle in radians.

The Time() Function

Description: Returns the current system time.

Syntax: Time()

Arguments: None.

41.1389-6 AppA 723 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 724 Appendix A VBScript 2.0 Reference

The TimeSerial() Function

Description: Returns the time for a specific hour, minute, and second.

Syntax: TimeSerial(hour, minute, second)

Arguments: hour is any numeric expression between 0 and 23; minute is any numeric expres-
sion between 0 and 59; and second is any numeric expression between 0 and 59.

The TimeValue() Function

Description: Returns the current time.

Syntax: TimeValue(time)

Arguments: time is any valid time expression.

The Trim() Function

Description: Returns a string without leading and trailing spaces.

Syntax: Trim(string)

Arguments: string is any valid string expression.

The TypeName() Function

Description: Returns subtype information about a variable.

Syntax: TypeName(varName)

Arguments: varName is the required variable name.

The UBound() Function

Description: Returns the largest available subscript for a dimension of an array.

Syntax: UBound(arrayname[, dimension])

Arguments: arrayname is the name of a valid array, and dimension (optional) is a number
indicating the dimension to find the upper bound.

The UCase() Function

Description: Returns a string that has been converted to uppercase characters.

Syntax: UCase(string)

Arguments: string is any valid string expression.

41.1389-6 AppA 724 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Constants 725

The VarType() Function

Description: Returns the subtype of a variable.

Syntax: VarType(varName) App

Arguments: varName is the required variable name. A

The Weekday() Function
Description: Returns a whole number representing the day of the week.

Syntax: Weekday(date, [firstdayofweek])

Arguments: date is any valid date expression, and firstdayofweek is an optional date constant
to assign the first day of week, as seen in Table A.9.

The WeekdayName() Function

Description: Returns the specified day of the week.

Syntax: WeekdayName(weekday, [abbreviate, [firstdayofweek]])

Arguments: weekday is the numeric representation for the day of the week; abbreviate is an
optional Boolean value (if set to True, the weekday name will be abbreviated; if set to False, the
full weekday name is displayed); and firstdayofweek is an optional date constant to assign the
first day of week, as seen in Table A.9.

The Year() Function

Description: Returns the current year.

Syntax: Year(date)

Arguments: date is any valid date expression.

Constants
VBScript gives developers the ability to use constants to replace unmemorable or difficult-to-
read numeric or string values. These intrinsic constants can be referenced by either their
names or their values.

Color Constants
VBScript gives you the ability to use color constants within your scripts. These constants, as
shown in Table A.7, can be referenced directly in your scripts.

41.1389-6 AppA 725 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 726 Appendix A VBScript 2.0 Reference

Table A.7 Color Constants

Constant Value Description
vbBlack &h00 Black
vbRed &hFF Red
vbGreen &hFF00 Green
vbYellow &hFFFF Yellow
vbBlue &hFF0000 Blue
vbMagenta &hFF00FF Magenta
vbCyan &hFFFF00 Cyan
vbWhite &hFFFFFF White

Comparison Constants
Many intrinsic VBScript functions perform comparisons to complete a task. These give you the
ability to explicitly define the comparison method by using the comparison constants shown in
Table A.8.

Table A.8 Comparison Constants

Constant Value Description
VbBinaryCompare 0 Performs a binary comparison.
VbTextCompare 1 Performs a textual comparison.
VbDatabaseCompare 2 Performs a comparison based on information
extracted in the database where the compari-
son is to be performed.

The Date/Time Constants

VBScript 2.0 provides new date/time functions that help speed scripting development and
eliminate excessive coding that would otherwise have to be done by hand. Many of the new
functions utilize the date/time constants to help customize the functions (see Table A.9).

Table A.9 Date/Time Constants

Constant Value Description
vbSunday 1 Sunday.
vbMonday 2 Monday.
vbTuesday 3 Tuesday.

41.1389-6 AppA 726 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Constants 727

Constant Value Description

vbWednesday 4 Wednesday.
vbThursday 5 Thursday. App
vbFriday 6 Friday. A
vbSaturday 7 Saturday.
vbFirstJan1 1 Uses the week in which January 1 occurs
(default).
VbFirstFourDays 2 Uses the first week that has at least four days in
the new year.
VbFirstFullWeek 3 Uses the first full week of the year.
VbUseSystem 0 Uses the date format contained in the regional
settings on the host computer.
VbUseSystemDayOfWeek 0 Uses the day of the week specified in your
system settings for the first day of the week.

Date Format Constants

In addition to the new date/time functionality, VBScript 2.0 also introduces new date format
functions that can use the constants shown in Table A.10.

Table A.10 Date Format Constants

Constant Value Description
VbGeneralDate 0 Displays a date and/or time.
VbLongDate 1 Displays a date using the long date format.
VbShortDate 2 Displays a date using the short date format.
VbLongTime 3 Displays a time using the long time format.
VbShortTime 4 Displays a time using the short time format.

Notice that date and time settings are controlled within your regional settings found in the
Control Panel. Consider the following script:
<%
Response.Write FormatDateTime(now,VbGeneralDate) &”<BR>”
Response.Write FormatDateTime(now,VbLongDate) &”<BR>”
Response.Write FormatDateTime(now,VbShortDate) &”<BR>”
Response.Write FormatDateTime(now,VbLongTime) &”<BR>”
Response.Write FormatDateTime(now,VbShortTime) &”<BR>”
%>

41.1389-6 AppA 727 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 728 Appendix A VBScript 2.0 Reference

The previous script will produce the following results:

10/12/97 3:47:05 PM
Sunday, October 12, 1997
10/12/97
3:47:05 PM
15:47

File Input/Output Constants

The file input/output constants are used to tap into the functionality of the FileSystemObject.
The FileSystemObject provides access to files on the server. The FileSystemObject can open
a file with three different permission levels. The constants enabling read-only, write-new, and
file-append modes are shown in Table A.11.

Table A.11 File Input/Output Constants

Constant Value Description
ForReading 1 Opens a file for read-only access.
ForWriting 2 Opens a file for write access. If the file exists, the previous
content is overwritten.
ForAppending 8 Opens a file and appends the data stream to the end of the
file.

Miscellaneous Constant
VBScript uses the constant shown in Table A.12 to represent an error returned from an
ActiveX Automation object.

Table A.12 Miscellaneous Constant

Constant Value Description
VbObjectError &h80040000 Error message from an Automation Object

The MsgBox Constants

The message box uses two sets of constants. The display constants, as displayed in Table A.13,
are used to create the look and feel of the message box. The message box result constants,
shown in Table A.15, are used to determine user response to the displayed message box.

41.1389-6 AppA 728 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Constants 729

Table A.13 Message Box Display Constants

Constant Value Description
vbOKOnly 0 Displays OK button only. App

vbOKCancel 1 Displays OK and Cancel buttons. A

vbAbortRetryIgnore 2 Displays Abort, Retry, and Ignore buttons.
vbYesNoCancel 3 Displays Yes, No, and Cancel buttons.
vbYesNo 4 Displays Yes and No buttons.
vbRetryCancel 5 Displays Retry and Cancel buttons.
vbCritical 16 Displays Critical Message icon.
vbQuestion 32 Displays Warning Query icon.
vbExclamation 48 Displays Warning Message icon.
vbInformation 64 Displays Information Message icon.
vbDefaultButton1 0 First button is set as the default button.
vbDefaultButton2 256 Second button is set as the default button.
vbDefaultButton3 512 Third button is set as the default button.
vbDefaultButton4 768 Fourth button is set as the default button.
vbApplicationModal 0 The user must respond to the message box
before continuing work in the current
application.
VbSystemModal 4096 System modal. All applications are sus-
pended until the user responds to the
message box.

You can trap the user response by the return values of the message box. The constants shown
in Table A.14 are used to identify which button the user selected in the message box.

Table A.14 Message Box Results Constants

Constant Value Description
vbOK 1 OK button was clicked.
vbCancel 2 Cancel button was clicked.
vbAbort 3 Abort button was clicked.
vbRetry 4 Retry button was clicked.

continues

41.1389-6 AppA 729 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 730 Appendix A VBScript 2.0 Reference

Table A.14 Continued

Constant Value Description
vbIgnore 5 Ignore button was clicked.
vbYes 6 Yes button was clicked.
vbNo 7 No button was clicked.

String Constants
VBScript can use string constants to help control output to the current page or file, as shown in
Table A.15.

Table A.15 String Constants

Constant Value Description
vbCr Chr(13) Carriage return
vbCrLf Chr(13) & Chr(10) Carriage return–line feed combination
vbFormFeed Chr(12) Form feed; not useful in Microsoft
Windows
vbLf Chr(10) Line feed
vbNewLine Chr(13) & Chr(10) Platform-specific new line character or
Chr(10) whatever is appropriate for the platform
vbNullChar Chr(0) Character having the value 0
vbNullString String having value 0 Not the same as a zero-length string (“”);
used for calling external procedures
vbTab Chr(9) Horizontal tab
vbVerticalTab Chr(11) Vertical tab; not useful in Microsoft
Windows

Tristate Constants
VBScript combines Boolean functionality with the capability to use the default values of a func-
tion into one set of constants called tristate constants. The tristate constants are shown in Table
A.16. Typically, the default settings are extracted from the computer’s region settings as found
in the Control Panel.

41.1389-6 AppA 730 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 Constants 731

Table A.16 Tristate Constants

Constant Value Description
TristateTrue -1 True App

TristateFalse 0 False A
TristateUseDefault -2 Use default setting of calling function

VarType Constants
VBScript enables you to determine the Variant subtype a variable is stored as by using the
Vartype function. The Vartype function returns the subtype values shown in Table A.17.

Table A.17 VarType Constants

Constant Value Description
vbEmpty 0 Un-initialized (default)
vbNull 1 Contains no valid data
vbInteger 2 Integer subtype
vbLong 3 Long subtype
vbSingle 4 Single subtype
vbDouble 5 Double subtype
vbCurrency 6 Currency subtype

vbDate 7 Date subtype

vbString 8 String subtype
vbObject 9 Object
vbError 10 Error subtype

vbBoolean 11 Boolean subtype

vbVariant 12 Variant

vbDataObject 13 Data access object

vbDecimal 14 Decimal subtype
vbByte 17 Byte subtype
vbArray 8192 Array

41.1389-6 AppA 731 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 732 Appendix A VBScript 2.0 Reference

VBScript Objects
VBScript provides access to four built-in objects (see Table A.18). These objects are used to
encapsulate a specific area of code functionality. The objects use methods, shown in Table A.19,
to perform an action, and have properties, shown in Table A.20, to describe the object.

Table A.18 VBScript Objects

Name Description
Dictionary Object that stores data key and item pairs.
Err Contains runtime error information.
FileSystemObject Provides access to a computer’s file system.
TextStream Provides sequential file access.

VBScript Object Methods

VBScript uses methods to perform an action on or to control an object. Table A.19 provides a
quick guide to the VBScript 2.0 methods.

Table A.19 VBScript 2.0 Methods

Method Description
Add Adds a key and item pair to a Dictionary object.
Clear Clears all settings of the Err object.
Close Closes an open TextStream file.
CreateTextFile Creates a specified filename and returns a TextStream object.
Exists Returns whether a specified key exists in the Dictionary
object.
Items Returns an array containing all items in a Dictionary object.
Keys Returns an array containing all existing keys in a Dictionary
object.
OpenTextFile Opens a specified file and returns a TextStream object.
Raise Generates a runtime error.
Read Reads characters from a TextStream file.
ReadAll Reads an entire TextStream file.
ReadLine Reads an entire line from a TextStream file.

41.1389-6 AppA 732 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 VBScript Object Properties 733

Method Description
Remove Removes a specific key and item pair from a Dictionary
object.
App
RemoveAll Removes all key and item pairs from a Dictionary object.
A
Skip Skips a specified number of characters when reading a
TextStream file.
SkipLine Skips the next line when reading a TextStream file.
Write Writes a specified string to a TextStream file.
WriteBlankLines Writes a specified number of newline characters to a
TextStream file.
WriteLine Writes a specified string and newline character to a
TextStream file.

VBScript Object Properties

Properties of an object are used to describe specific information about an object. The proper-
ties of VBScript objects are shown in Table A.20.

Table A.20 Properties

Name Description
AtEndOfLine Returns whether the file pointer is positioned immediately on
the end-of-line marker in a TextStream file.
AtEndOfStream Returns whether the file pointer is positioned at the end of a
TextStream file.
Column Returns the column number of the character position in a
TextStream file.
CompareMode Sets or returns the comparison mode for comparing string
keys in a Dictionary object.
Count Returns the number of items in a Dictionary object.
Description Returns or sets a descriptive string associated with an error.
HelpContext Sets or returns a context ID for a topic in a help file.
HelpFile Sets or returns a path to a Help File.
Item Sets or returns an item for a specified key in a Dictionary
object.
Key Sets a key in a Dictionary object.

continues

41.1389-6 AppA 733 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 734 Appendix A VBScript 2.0 Reference

Table A.20 Continued

Name Description
Line Returns the current line number in a TextStream file.
Number Sets or returns a numeric value specifying an error.
Source Sets or returns the name of the object that generated the
error.

41.1389-6 AppA 734 10/26/98, 10:16 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App A lp#3
 B
735

APPENDIX

Active Server Object App

B
Reference

Active Server Pages derives its rich functionality from the

use of six inherent objects. These six objects do have to be
explicitly instantiated but are directly built into Active
Server Pages. These objects have their own predefined
roles and each has its own properties, methods, and
events. These objects form the foundation of ASP develop-
ment and consequently have their own chapters dedicated
to fully describe them. The following provides a brief
overview of each object’s functionality. ■

42.1389-6 AppB 735 10/26/98, 10:17 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 736 Appendix B Active Server Object Reference

Controlling the Application Object

The Application object is used to manage all information in the ASP application. The informa-
tion can be accessed and passed between different users in the application. The Application
object has two events, the Application_OnStart and Application_OnEnd events, as shown in
Table B.1. The Application object is created by the Active Server to represent the
instantiation of an Active Server application. The Application object is created when a user
requests any page in the ASP application’s virtual directory. Once the Application object is
started, all subsequent users can reference its objects and variables.

Table B.1 The Application Object’s Events

Event Description
Application_OnStart Initiated when the ASP application is first started.
Application_OnEnd Initiated when the ASP application is terminated.

These individual application events are found in the GLOBAL.ASA file and are executed in a
manner similar to any subroutine or function.
<SCRIPT LANGUAGE=”VBScript” RUNAT = “Server”>

Sub Application_OnStart ‘=== Defines Application Object’s OnStart Event

Application(“strMessage”)= “Welcome to Active Server Pages”
End Sub
Sub Application_OnEnd

End Sub

</SCRIPT>

However, if an event is called and no event procedure is defined, a runtime error will occur
because a reference is being made to an event that does not exist.

Declaring Application-Level Variables

Application-level variables are accessible by all users of the application. Application-level vari-
ables can be defined as needed in any .ASP file, not just in the Application_OnStart and
Application_OnEnd events of the GLOBAL.ASA file.

To declare an application variable, use the following syntax:

Application(varName)

Application_OnStart Event The Application_OnStart event requires the following syntax

within the GLOBAL.ASA file:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>
Sub Application_OnStart
. . .
End Sub </SCRIPT>

42.1389-6 AppB 736 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Controlling the Application Object 737

The Application_OnStart event is executed before any user or session event is triggered. Any
attempts to call a Session, Response, or Request object will generate a fatal runtime error
because these objects have not been created at this point. However, you are able to reference
the Server object.

Application_OnEnd Event The Application_OnEnd event requires the following syntax

within the GLOBAL.ASA file:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>
Sub Application_OnEnd
‘=== Your Script Here
End Sub App

</SCRIPT> B
The Application_OnEnd event occurs when the Web server is shut down. As a result, the
Application object is destroyed. The shutting down of the Web server can occur with the
termination of the HTTP service or when the GLOBAL.ASA file needs to be recompiled and
reloaded into memory.

Controlling Application-Level Variables with Methods

The Application object has two methods designed to control variable access: the Lock and
UnLock methods. These methods help to prevent multiple users from changing application-level
variables at the same time (see Table B.2).

Table B.2 The Application Object’s Methods

Name Description
Lock Prevents other clients from changing Application object properties.
UnLock Enables other clients to modify the Application object’s properties.

Lock Method The Lock method is to control the Application object’s properties by prevent-
ing concurrent users from modifying the same application-level variable at the same time. The
Lock method is implemented by using the following syntax:

Application.Lock

UnLock Method The Application.UnLock method releases control of the application vari-
ables. The UnLock method is implemented using the following syntax:
Application.UnLock

The Application.UnLock method is implicitly called when the processing of a page is complete
and the UnLock method has not been called or when the script processing times out. The de-
fault page-level script timeout value is 90 seconds.

42.1389-6 AppB 737 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 738 Appendix B Active Server Object Reference

Instantiating Application-Level Objects

You can create a reference to an application-level object’s automation object by embedding the
unique class identifier either by the ProgID or the ClassID within the object tag. The ProgID
represents the registered name of the object. The ClassID refers to the registered class num-
ber of the object. To create a server-side object, use the following syntax:
<OBJECT RUNAT=Server SCOPE = Application ID=Identifier
{PROGID=”progID”|CLASSID=”ClassID”}>
</OBJECT>

where Scope identifies the object’s lifetime and ID identifies the object’s instantiated name.

The <OBJECT> tag must be placed outside the <SCRIPT> and </SCRIPT> tags used to define the
GLOBAL.ASA file. If an <OBJECT> tag or any other tag is found between the <SCRIPT> tags, the
script interpreter will try to process the <OBJECT> tags and generate an error. This <SCRIPT>
and <OBJECT> tag layout is illustrated in the following code:
<SCRIPT LANGUAGE=”VBScript” RUNAT=”Server”>
Sub Application_OnStart
‘=== Application Level events and object placed here
End Sub

Application_OnStart
</SCRIPT>

<OBJECT RUNAT=Server SCOPE=Application ID=MyDataConn

PROGID=”RemoteConn.DataConn”>
</OBJECT>

The application-level object can now be accessed from any ASP file in the application by using
the ID name of the object.

Controlling the Server Object

The Server object is an ASP object used to control administrative features of the IIS Web
server and actions that deal with the HTTP service. The Server object represents a control
interface to the HTTP service. The Server object is referenced using the following syntax:
Server.Method/Property

where Server is the Active Server object.

The Server object has four methods that are used to control various aspects of the Web server,
as shown in Table B.3.

Table B.3 The Server Object’s Methods

Method Description
CreateObject Creates a Server instance of an object.
HTMLEncode Utilizes HTML encoding to deliver text to browser.

42.1389-6 AppB 738 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Controlling the Server Object 739

Method Description
MapPath Translates the virtual Web server path to the physical path on
the server.
URLEncode Utilizes URL encoding techniques.

The CreateObject Method

Perhaps one of the most exciting aspects of Active Server Pages is the capability to create an
instance of an ActiveX object. This capability to tap into ActiveX objects provides a virtually App
endless number of ways to deliver information to any Web client that were once only available B
to propriety client software or dedicated network connections. The Server object acts as a
translator between the in-house data and logic stores and virtually any HTML-compliant
browser. The CreateObject method uses the following syntax:
Server.CreateObject(progID)

where progID is the class or type of object to be instantiated. The progID also requires the
following special format:
appname.objecttype

where appname is the application name hosting the object and objecttype is the class or type of
the object to create. All COM-based objects are required to have one class type per application
name.

The CreateObject method only creates objects that have page-level scope. If you need to ac-
cess the same component multiple times, in most situations you would benefit from using an
application- or session-level object.

The MapPath Method

The Server object uses the MapPath method to track and manage path information on the
server. The MapPath method requires the following syntax:
Server.MapPath( path )

where Server is the Active Server object and path is a physical or virtual directory. There are
two basic rules to remember with the path variables.

■ Path arguments that start with a backslash (\) or a forwardslash (/) are used to repre-
sent virtual directories.
■ Path arguments that do not start with the backslash or forward slash represent relative
directories.

The ScriptTimeOut Property

The ScriptTimeOut property is a property of the Server object that prevents a process from
running endlessly. The ScriptTimeOut value requires the following syntax:
Server.ScriptTimeout = Seconds

42.1389-6 AppB 739 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 740 Appendix B Active Server Object Reference

where Seconds indicates the number of seconds allotted for page-level scripting to process.
After this amount of time has been reached, the Server object terminates the process and
writes an event to the event log.

Using Server Encoding Techniques

All Web technology is based on the transfer of simple text across the Internet or an intranet via
the TCP/IP protocol and the translation of that text into content within the Web browser. Be-
cause the browser only interprets text, embedded non-ASCII characters are misinterpreted as
text or simply are not available from the keyboard. The Server object uses the URLEncode and
HTMLEncode methods to ensure the proper character translation between the Web server and
browser.

The URLEncode Method

The URLEncode method of the Server object is used to deliver explicit information from the
Web server to the client via the URL. To ensure that all the characters entered by the user are
properly passed to the server, the characters must be scanned to ensure that only known char-
acters are being sent. To use the URL encoding methods, follow this syntax:
Server.URLEncode(string)

where Server is the Server object and string is the string to apply the URL encoding rules.

The URLEncode method performs the following processes to the data:

■ Spaces are transformed into plus (+) symbols.

■ Fields are left unencoded.
■ An unencoded equal sign assigns an unencoded field name with the data value.
■ Non-ASCII characters are transformed into escape codes.

Using the HTMLEncode Method

The HTMLEncode method is used by the server to explicitly define the characters to be dis-
played on a page. To use the HTMLEncode feature, use the following syntax:
Server.HTMLEncode(string)

where Server is the Server object and string is the string to encode. The HTMLEncode method
is important to ensure that the proper characters are displayed on the page and not processed
by the server.

42.1389-6 AppB 740 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Trapping the ObjectContext’s Events 741

The ObjectContext Object: Interacting with the

Transaction Server
With the Internet Information Server 4.0, Active Server Pages have a direct interface to con-
trolling components in the Microsoft Transaction Server. This interface is made possible
through the ObjectContext object. When the scripting engine processes this directive, the
Active Server Page is executed as a transaction on the Transaction Server. To initiate the trans-
action processing of the ObjectContext object, use the following directive syntax:
<%@ Transaction = Required %> App

ObjectContext uses two methods to control the objects managed by the Microsoft Transaction B
Server, as shown in Table B.4.

Table B.4 The ObjectContext Object’s Methods

Method Description
SetComplete Sets the work of an object as a success and permanently accepts
the changes to the resource.
SetAbort Sets the work of an object as a failure and returns the resource to
its original state, neglecting any changes made to the resource.

The SetComplete Method

The SetComplete method is used to commit an MTS object’s transaction as a success. The
changes made by the object method on a resource are permanently committed. To accept
changes made by object, use the following syntax:
ObjectContext.SetComplete

After the SetComplete method is executed, the OnTransactionCommit event is triggered.

The SetAbort Method

The SetAbort method is used to roll back changes made by an object managed by the MTS.
When the SetAbort method is called, the object has completed its work and is returned to its
original state. To reject changes made by the MTS object, use the following syntax:
ObjectContext.SetAbort

After the SetAbort method is executed, the OnTransactionAbort event is triggered.

Trapping the ObjectContext’s Events

ObjectContext has two events that are triggered after the ObjectContext’s methods are ex-
ecuted, as shown in Table B.5.

42.1389-6 AppB 741 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 742 Appendix B Active Server Object Reference

Table B.5 The ObjectContext Events

Event Description
OnTransactionCommit Event triggered after the ObjectContext.SetComplete
method is executed.
OnTransactionAbort Event triggered after the ObjectContext.SetAbort method is
executed.

The OnTransactionCommit Event

The OnTransactionCommit event is triggered by the ObjectContext object SetComplete
method. When the SetComplete method is processed, the script will process the
OnTransactionCommit subroutine.

Sub OnTransactionCommit()
Response.Redirect “CreateVirutalReceipt.ASP”
End Sub

The relationship between OnTransactionCommit event and ObjectContext.SetComplete

method is demonstrated in the following example:
<%@TRANSACTION = Required %>
<%
Sub OnTransactionCommit()
Dim strMessage
strMessage =”This was generated by executing the SetComplete method “
strMessage = strMessage & “to trigger the OnTransactionCommit event”
Response.Write strMessage
End Sub
%>
<HTML>
<HEAD>
<TITLE>Set Complete</TITLE>
</HEAD>
<BODY>

<%
ObjectContext.SetComplete
%>

</BODY>
</HTML>

The code example initiates transaction processing using the TRANSACTION = Required direc-
tive and executes the SetComplete method to trigger the OnTransactionCommit() event. In this
example, the OnTransactionCommit subroutine, writes the following text to the page:
This text message was generated by executing the SetComplete method to trigger
the OnTransactionCommit event

42.1389-6 AppB 742 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Communicating with Web Client: The Response Object 743

OnTransactionAbort Event
Similar to its counterpart, the OnTransactionAbort event is triggered when the ObjectContext
object SetAbort event is executed. When the SetAbort method is processed, the script will
process the OnTransactionCommit event.
Sub OnTransactionAbort()
Response.Redirect “InsufficientFunds.ASP”
End Sub

You can apply the same example shown in the OnTransactionCommit to OnTransactionAbort,
as shown in the following example: App
<%@TRANSACTION = Required %>
<%
B
Sub OnTransactionAbort()
Dim strMessage
strMessage =”This text message was generated by executing the SetAbort
➥method “
strMessage = strMessage & “to trigger the OnTransactionAbort event”
Response.Write strMessage
End Sub
%>
<HTML>
<HEAD>
<TITLE>Set Abort</TITLE>
</HEAD>
<BODY>

<%
ObjectContext.SetAbort
%>

</BODY>
</HTML>

The example initiates transaction processing using the TRANSACTION = Required directive and
executes the SetAbort method to trigger the OnTransactionAbort() event. In this example,
the OnTransactionAbort subroutine writes the following text to the page:
This text message was generated by executing the SetAbort method to trigger the
OnTransactionAbort event.

Communicating with Web Client: The Response

Object
The Active Server uses the Response object to control and manage the data sent to the
browser. The Response object is responsible for controlling the delivery of data, writing HTTP
header information, writing text, HTML, scripting variables, and non-textual information, and
controlling cookies on the client browser.

42.1389-6 AppB 743 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 744 Appendix B Active Server Object Reference

The Response Collection

The Response object uses the Cookie collection to manage and control both cookie files and the
data stored within the cookies. The Cookie collection enables single or multiple variables to be
stored and manipulated in temporary text files on the client’s browser. The Response collection
is shown in Table B.6.

Table B.6 The Response Collection

Collection Name Description
Cookies Sets cookies values.

The Cookies Collection

The Cookies collection enables single or multiple variables to be stored and manipulated in
temporary text files on the client’s browser. The Cookies collection requires the following
syntax:
Response.Cookies(Cookie)[(key)|.attribute] = value

where Response is the built-in Response object, Cookie is the name of the cookie file, key identi-
fies a dictionary element, attribute is a specific characteristic of the cookie, and value is the
value being assigned to the cookie.

Notice that you do not have to manually trap and replace non-ANSCI characters, that is, encode
the string data that will be transferred to the cookie. The encoding process automatically filters
out any non-ANSCI characters.

The Cookies collection also uses attributes to help manage cookies on the browser. Table B.7
displays the cookie attributes that can be set by the Response collection.

Table B.7 The Attributes of the Response Cookie Collection

Name Description
Expires Sets the date when the cookie will expire.
Domain Specifies cookie delivery to only members specified by this domain.
Path Determines the delivery path information.
Secure Specifies whether the cookie is secure.
HasKey Returns whether the cookie contains keys.

42.1389-6 AppB 744 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Communicating with Web Client: The Response Object 745

The Response Object’s Properties

The Response object has various properties that can be used to set characteristics of informa-
tion delivered by the server. Table B.8 illustrates the Response object’s properties.

Table B.8 The Response Properties

Property Name Description
Buffer Sets whether or not page output is buffered.
Charset Appends the name of the character set to the content-type App
header. B
Expires Sets the amount of time before the page cached on the browser
expires.
ExpiresAbsolute Sets the date and time a page cached on a browser expires.
IsClientConnected Determines whether the browser has disconnected from the
server.
Status Indicates the status line returned by the server.
PICS Adds the value of a PICS label to the PICS-label field of the
response header.

The Buffer Property The Buffer property is used to control and regulate when information is
sent to the requesting browser. The Web server can either stream information to the user as
the server is processing the script, or it can wait to release all the data after the entire script is
finished processing. The output buffering uses the following syntax:
Response.Buffer [= flag]

where Response is the Response object and flag can be set to either True or False. If True, the
server buffering is cached until the entire ASP page has completed processing or until the
Flush or End methods have been called. If the flag is set to False, server buffering is disabled.

The CacheControl Property The Response object uses the CacheControl property to enable a
proxy server to cache output from Active Server Pages. The CacheControl requires the follow-
ing syntax:
Response.CacheControl [= True/False ]

The CharSet Property The CharSet property of the Response object is used to set the charac-
ter set of the ContentType to be displayed on the requesting browser. The CharSet property
controls the ContentType by using the following syntax:
Response.Charset(CharSetName)

where CharSetName is the character set designated for a particular page.

42.1389-6 AppB 745 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 746 Appendix B Active Server Object Reference

The ContentType Property The ContentType Response object property enables you to specifi-
cally control the HTTP content-type that is sent to the browser. The browser uses the content-
type information to determine how it should interpret the information, such as treating the
information as text, HTML, or as an image file. The ContentType functionality is implemented
using the following syntax:
Response.ContentType [= ContentType ]

where ContentType is the browser specific MIME Content type.

The Expires Property The Expires property is an interesting feature that enables you to
specify the amount of time in minutes before the page is expired from the browser cache. The
Expires property uses the following syntax:

Response.Expires [= number]

where number is the number of minutes the page will remain active in the cache. Once this time
has expired, the browser will be forced to retrieve information from the hosting site.

The ExpiresAbsolute Property ExpiresAbsolute is an extension of the Expires property. The

ExpiresAbsolute property enables you to specify the exact time and date a page will expire in
the browser’s cache. The ExpiresAbsolute property requires the following syntax:
Response.ExpiresAbsolute [= [date] [time]]

where date is the date on which the page will expire and time indicates the time the page will
expire. However, note that the expiration date and time is converted to Greenwich Mean Time.

The PICS Property The PICS property of the Response object enables you to control rating
values of the PICS-label field in the response header. This functionality can be controlled by
implementing the following syntax:
Response.PICS(Picslabel)

where PicsLabel is the formatted PICS label.

The Status Property The Response object uses the Status property to control the status line
returned by the Web server. The status line is used to determine the results of the Web server
request and is specified by HTTP specifications. To implement this functionality, use the follow-
ing syntax:
Response.Status = StatusDescription

where StatusDescription is the status code and status code description. The following is a list
of common HTTP status lines that can be returned by the Web server:

400 Bad Request

410 Unauthorized - Login Failed
404 Not Found
406 Not Acceptable
412 Precondition Failed
414 Request-URL Too Long

42.1389-6 AppB 746 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Sending Output to the Browser: The Response Methods 747

500 Internal Server Error

501 Not Implemented
502 Bad Gateway

One of the most common uses of the Status property is to check or authenticate the rights of
the requesting user. For example, to force user validation for a page, use the following code to
prompt the user for a username and password dialog box:
<% Response.Status = “401 Unauthorized” %>

The IsClientConnected Property IsClientConnected tracks whether the requesting browser App
has disconnected from the server since the last time the server issued a Response.Write com-
mand. The IsClientConnected property requires the following syntax: B
Response.IsClientConnected()

The IsClientConnected property provides an extension of the Session.Timeout capability by

enabling the requesting browser to exceed the session timeout variable without losing server
connection and is often implemented before processing script. The following example tests the
Response object to see whether the connection is valid before calling a function to transfer data.

If Response.IsClientConnected Then
vTransfer = TranferInfo(“AcctRetail”, “AcctWholeSale”)
Else
VTransfer = ReConnect(“UserName”, “Department”)
End If

Sending Output to the Browser: The Response

Methods
The Response methods are used to explicitly control information flow from the server to the
browser. Table B.9 lists the available Response methods.

Table B.9 The Response Methods

Method Name Description
Write Writes a string to the current HTTP output.
BinaryWrite Writes information to the current HTTP output without any
character conversion.
Clear Erases any buffered HTML output.
End Stops the Web server from processing the script and returns
the current result.
Flush Bypasses buffering and sends output immediately to client.
Redirect Attempts to automatically route the browser to a URL.

continues

42.1389-6 AppB 747 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 748 Appendix B Active Server Object Reference

Table B.9 Continued

Method Name Description
AddHeader Writes a string to the HTML header.
AppendToLog Writes a string to the end of the Web server log entry.

The Write Method

The Write method of the Response object is used to send output to the browser. The Write
method requires the following syntax:
Response.Write variant

where variant is any Variant datatype supported by the Visual Basic scripting engine.

When using the Response.Write method, the Variant datatype to be delivered to the client
cannot contain the string %>. This string is interpreted by the scripting engine as an end script
delimiter. Instead, use the string %\> to write the closing script delimiter tag.

The Write method uses the Variant datatype to send information to the browser. The Variant
datatype, although the most versatile of datatypes, does have its limitations. The Variant
datatype itself can contain only 1,022 bytes of information. If you try to use a variable that is
larger than 1,022 bytes, a runtime error will occur.

The AddHeader Method

The AddHeader method enables you to add header information to the existing HTTP header. To
use the AddHeader method, use the following syntax:
Response.AddHeader name, value

where name is the name of the new header variable and value is the value assigned to the new
header variable.

The AddHeader method must be called before any content is sent to the browser. Failure to do
so will generate a runtime error because the HTTP protocol first sends header information to
the browser.

The AppendToLog Method

The AppendToLog method of the Response object enables you to append information directly to
the Web server log file. To use this functionality, use the following syntax:
Response.AppendToLog string

where Response is the built-in Response object and string is an 80-byte character string.

42.1389-6 AppB 748 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Sending Output to the Browser: The Response Methods 749

The BinaryWrite Method

The BinaryWrite method enables direct non-formatted output to be displayed to the request-
ing browser. This direct output is useful when displaying non-string information, such as vari-
ous image formats. To use the BinaryWrite method, use the following syntax:
Response.BinaryWrite data

where data is information that will be sent to the browser without any character conversion.

The Clear Method App

The Clear method is used to erase any HTML that has been buffered on the server. The Clear
property uses the following syntax: B
Response.Clear

where Response is the built-in Response object and Clear is the Clear method.

The End Method

The End method also is used to manage the buffered server output. The End method returns
the current buffered output up to the point where the End method is called. To use the End
method, use the following syntax:
Response.End

where Response is the built-in Response object and End activates the End method.

The Flush Method

The Flush method is used by the Response object to immediately send any buffered output to
the browser. The syntax for the Flush method is as follows:
Response.Flush

where Response is the built-in Response object and Flush calls the Flush method. The Flush
method can only be used if buffering has been activated.

The Redirect Method

The Redirect method is used to route the browser to another Web page. The Redirect
method is implemented by using the following syntax:
Response.Redirect URL

where Response is the built-in Response object and URL is the Uniform Resource Locator that
indicates where the browser is to be routed.

42.1389-6 AppB 749 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 750 Appendix B Active Server Object Reference

Retrieving Information Using the Request Object

The Request object is responsible for retrieving information from the Web browser. The
Request object uses collections, properties, and methods to retrieve information from the user.

Accepting User Information Using the Request Collections

The Request object uses separate objects that can be grouped together to interface with the
calling client (see Table B.10).

Table B.10 The Request Collection

Collection Name Description
ClientCertificate Retrieves the certification fields issued by the Web
browser.
Cookies Retrieves the values of the cookies sent in an HTTP
request.
Form Retrieves the values of form elements posted to the
HTTP request.
QueryString Retrieves the values of the variables in the HTTP query
string.
ServerVariables Retrieves the values of predetermined environment
variables.

The ClientCertificate Collection The Request object’s ClientCertificate collection is used

to provide proper security identification across unsecured environments. The
ClientCertificate keys are shown in Table B.11.

Table B.11 The ClientCertificate Certification Fields

Key Description
Subject Returns a list of subfield values that contain information
about the subject of the certificate.
Issuer Contains a list of subfield values containing information
about the issuer of the certificate.
ValidForm Returns when the certificate becomes valid.
ValidUntil Specifies when the certificate expires.
SerialNumber Returns the certification serial number as an ASCII
representation of hexadecimal bytes.
Certificate Returns the binary stream of the entire certificate
content in ASN.1 format.

42.1389-6 AppB 750 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Retrieving Information Using the Request Object 751

The use of subfields enables specific information to be retrieved from the Subject and Issuer
key fields mentioned above. Table B.12 presents the available subfields for the
ClientCertificate collection.

Table B.12 The ClientCertificate Collection Subfields

Value Description
C Specifies the name of the country of origin.
O Specifies the company or organization name. App
OU Specifies the name of the organizational unit. B
CN Specifies the common name of the user.
L Specifies a locality.
S Specifies a state or province.
T Specifies the title of the person or organization.
GN Specifies a given name.
I Specifies a set of initials.

If no certificate is sent, the ClientCertificate collection returns EMPTY.

The Cookies Collection The Cookies collection is used by the Request object to retrieve
values stored in text files on the client’s machine.

The Cookies collection requires the following syntax:

Request.Cookies(cookie)[(key)|.attribute]

Its properties are summarized in Table B.13.

Table B.13 Cookies Collection Properties

Property Description
HasKeys Returns if the cookie contains keys.

The Form Collection The Form collection helps to extract data from information submitted via
the HTTP Post method. The Forms collection requires the following syntax:
Request.Form(parameter)[(index)|.Count]

where parameter is the name of the Form collection, index is the specific form element, and
Count identifies the number of elements that exist on a form.

42.1389-6 AppB 751 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 752 Appendix B Active Server Object Reference

For example, the following creates an ASP variable name vUserName by extracting values en-
tered in HTML forms fields named txtUserFirstName and txtUserLastName:
<%
vUserName = Request.Form(“txtUserFirstName”) & “ “ &
Request.Form(“txtUserLastName”)
%>

The QueryString Collection The QueryString collection is used by the Request object to
extract variables from the HTTP query string. The query string is textual content that occurs
after the question mark character (?) in the URL string:
Request.QueryString(variable)[(index)|.Count]

where Request is the Request object, variable is the name of the variable name in the query
string, index is the element index, and Count specifies the number of variables in the query
string.

For example, to populate an ASP variable named myDBDrilldown from a QueryString variable
named myDatabaseStr, use the following syntax:
<% myDBDrillDown = Request.QueryString(“myDatabaseStr “) %>

If the URL http:/ASPapp/loadDB.asp?myDatabaseStr=AcctNumber was submitted to the

server, the text AcctNumber would be assigned to the ASP variable.

The ServerVariables Collection The ServerVariables collection is used to obtain server

environmental variables. To use these features, the ServerVariables collection requires the
following syntax:
Request.ServerVariables (ServerVariable)

where ServerVariable is the name of the server variable. Table B.14 displays a list of common
server variables.

Table B.14 The Server Variables

Variable Name Description
*ALL_HTTP Displays all HTTP headers sent by the client.
*ALL_RAW Retrieves all headers as they are sent by the client.
*APPL_MD_PATH Retrieves the metabase path for the ISAPI DLL.
*APPL_PHYSICAL_PATH Retrieves the physical path corresponding to the
Metabase.
AUTH_TYPE Displays the authentication method used by the server.
*AUTH_USER Retrieves the authenticated user name in Raw format.
AUTH_PASSWORD Displays the value entered in the client’s authentication
dialog using BASIC authentication security.
*CERT_COOKIE Returns the unique ID for client certificate.

42.1389-6 AppB 752 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Retrieving Information Using the Request Object 753

Variable Name Description

*CERT_FLAGS Displays whether the client certificate is valid.
*CERT_ISSUER Displays Issuer field of the client certificate.
CERT_KEYSIZE Returns the number of bits in Secure Sockets Layer
connection key size.
*CERT_SERIALNUMBER Displays the client certificate Serial number field.
*CERT_SERVER_ISSUER Displays the server certificate Issuer field.
App
*CERT_SERVER_SUBJECT Displays the server certificate Subject field.
*CERT_SUBJECT Displays the client certificate Subject field. B
CONTENT_LENGTH Returns the length of the content.
CONTENT_TYPE Returns the data type of the content.
GATEWAY_INTERFACE Returns the version of the CGI specifications used on
the server.
HTTP_<HeaderName> Returns the information in the HeaderName.
*HTTPS Returns whether or not the request came in through
secure channel (SSL).
*HTTPS_KEYSIZE Returns the number of bits in Secure Sockets Layer
connection key size.
*HTTPS_SECRETKEYSIZE Returns the number of bits in server certificate private
key.
*HTTPS_SERVER_ISSUER Returns the server certificate Issuer field.
*HTTPS_SERVER_SUBJECT Displays the server certificate Subject field.
*INSTANCE_ID Returns the ID for the IIS instance.
*INSTANCE_META_PATH Returns the metabase path for the instance of IIS that
responds to the request.
LOGON_USER Displays the NT login account the request is made from.
PATH_INFO Displays the server path information.
PATH_TRANSLATED Returns the translated version of the PATH_INFO.
QUERY_STRING Returns the query string in the URL.
REMOTE_ADDR Displays the IP address of the requesting machine.
REMOTE_HOST Displays the name of the requesting host.
REQUEST_METHOD Returns the method that initiated the request.
SCRIPT_NAME Displays the virtual path to the executing script.

continues

42.1389-6 AppB 753 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 754 Appendix B Active Server Object Reference

Table B.14 Continued

Variable Name Description
SERVER_NAME Returns the server’s host name, DNS alias, or IP
address.
SERVER_PORT Returns the server port number the request is made on.
SERVER_PORT_SECURE Returns a 1 if request is made on a secure port, 0 if
unsecured.
SERVER_PROTOCOL Returns the name and version of the requesting
protocol.
SERVER_SOFTWARE Returns the name and version of HTTP server.
URL Returns the base portion of the URL.

Where * indicates the new IIS 4.0 Server variables.

The Request Properties and Methods

The Request object has one property and one method. The TotalBytes property dictates the
total amount of bytes sent in the request. The BinaryRead method is used to accept and store
data sent from a browser.

The TotalBytes Property The TotalBytes property is used to return the number of bytes sent
by the client and requires the following syntax:
Request.TotalBytes

The BinaryRead Method The BinaryRead method is used to read binary information that is
sent to the server from a POST request. To use the BinaryRead method, use the following syn-
tax:
myBinArray = Request.BinaryRead(count)

where count is the number of bytes to place into an array named myBinArray.

Managing the Session Object

The Session object is responsible for managing information for a specific user session. These
variables are not accessible by other user sessions, but they can be passed from page to page
within the ASP application. The Session object is created when a new user enters an applica-
tion. The termination of the application or the application’s timeout threshold destroys the
Session object for the user.

42.1389-6 AppB 754 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Managing the Session Object 755

The Session Object: A Cookie-Dependent System

The Session object is based on using cookies to store and transfer a unique user ID between
the browser client and the Web server. This session ID is used to create and reference Server
objects specific to a particular user. If cookies are not permitted on the client browser (because
of firewall issues, browser incompatibility, or desktop/network security concerns), the Session
object is rendered useless. If cookies are not permitted, ASP applications can still be developed
and deployed. However, most of the state tracking will have to be done through more cumber-
some methods, such as using HTML hidden fields and passing information via the URL.

When a user enters the application, the server first checks to see whether the requesting user App
has a valid SessionID. If the SessionID is found in the HTTP header, the user is identified as an
active user and is able to continue in the application. If no SessionID is found in the header, the
B
server generates an identifier and sends it to the browser’s cookie. This identifier is needed to
create a unique Session object that represents the specific user session. The purpose of the
identifier is to generate an exclusive identification to tie the actions of the browser client to
corresponding objects on the Web server. Each time a browser request is made, this unique
token is passed to identify the appropriate request.

CAUTION
Session objects are only supported on Web browsers that support cookies.

In a cookie-compliant browser world, the Session object greatly helps manage the user in the
application. However, when dealing with a non-secure cookie-compliant browser, it is not pos-
sible to safely track user information via the Session object. Typically, this is of importance
where client impersonation might occur. The counterfeiting of a user can occur in two situa-
tions. The first is if the session ID is captured while the data is in transit between the client and
server. The second possible security violation occurs when the cookie file itself is copied from
the user machine to another machine. With two valid session cookies, the user is able to tem-
porary clone himself onto another machine.

Implementing Session Variables and Properties

The Session object is responsible for managing user information. The user can create and
manipulate the Session object’s methods and properties using the following syntax:
Session.property|method

where Session is the Session object. The Session object has four properties, as listed in
Table B.15.

42.1389-6 AppB 755 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 756 Appendix B Active Server Object Reference

Table B.15 The Session Object’s Properties

Property Description
SessionID Returns the unique user session identifier.
TimeOut Returns or sets the user timeout value in minutes.
CodePage Sets the language attribute for the deployed pages.
LCID Sets the local identifier used to set local date, currency, and
time formats.

Declaring Session Variables

To create or reference a session variable, use the Session object and the name of the variable
using the following syntax:
ObjectLevel(varName)

where ObjectLevel is the Session object and varName is the name of the variable.

Notice that you do not have to use the Dim, Redim, Public, Private, or Const declared state-
ments that are found in VBScript to create variables.

Because the session variables are only available to the specific user session, no Lock and
UnLock methods are needed to prevent simultaneous updates to the same variable. However,
great care should be taken to ensure the correct spelling of session- and application-level vari-
ables. If a session or application variable is misspelled, a new variable with that misspelling is
created.

The SessionID Property The Session object uses the SessionID to keep track of user infor-
mation from page to page within the ASP application. The server generates the SessionID
when a new session is started. The SessionID is available using the following syntax:
Session.SessionID

The TimeOut Property The TimeOut property sets or returns the amount of time, in minutes,
that the Session object can remain inactive before the user’s Session object is destroyed. To
set the TimeOut property, use the following syntax:
Session.Timeout [ = nMinutes]

where Session is the Session object and nMinutes is the timeout value specified in minutes.

The LCID Property The LCID property is used to set the local identifier properties for an ASP.
The local identifier is used to control the display formatting that is specific to a localized loca-
tion or region. The LCID property requires the following syntax:
Session.LCID(=LCIDcode)

where LCIDcode is the valid local identifier. The LCID is standard abbreviation that identifies
localized formatting issues such as time, date, and currency formats.

42.1389-6 AppB 756 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Managing the Session Object 757

The CodePage Property The CodePage property of the Session object is used to assign the
system code page for an ASP. To implement this feature, use the following syntax:
Session.CodePage = CodePage

where CodePage is valid code page for the scripting engine.

Trapping Session Events

The Session object uses two events, shown in Table B.16, to help monitor when a user enters
and exits the ASP application. A new Session object is created by the Active Server to repre-
sent when a new user has entered the ASP application. Once the Session object is started, App
information pertaining to that user’s session can be stored on the Web server. B
Table B.16 The Session Object’s Events
Event Description
Session_OnStart Triggered when a new user enters ASP application.
Session_OnEnd Initiated when the ASP Session Object is terminated.

The Session_OnStart Event To trap the start of the Session object, you must use the
GLOBAL.ASA file.
The Session_OnStart event requires a layout similar to the following
GLOBAL.ASA syntax:

<SCRIPT LANGUAGE = ScriptLanguage RUNAT=Server>

Sub Session_OnStart
‘=== The Code to be executed when a new user enters the
➥app
End Sub

Session_OnStart ‘=== Calls the procedure Session_OnStart

</SCRIPT>

Session events can reference any other ASP object in the OnStart and OnEnd events.

The Session_OnEnd Event The Session_OnEnd event, which is located in the optional
GLOBAL.ASA file, is triggered when the current Session objects are closed. The Session_OnEnd
event requires the following syntax:
<SCRIPT LANGUAGE=ScriptLanguage RUNAT=Server>

Sub Session_OnEnd
‘=== Closing code here
End Sub

</SCRIPT>

42.1389-6 AppB 757 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 758 Appendix B Active Server Object Reference

where ScriptLanguage is any script-compliant language. The Session_OnEnd event is triggered

when the Session object is abandoned or times out. During this pre-shutdown of the Session
object, all interaction with the client browser is prohibited while the existing requests in the
queue are processed.

Any ASP code requests to the Response or Request objects in the Session_OnEnd event will
generate an error. References to all objects internal to the Web server, such as the Server,
Application, or Session objects, are still valid. However, reference to the Server.MapPath
method will cause a type mismatch error to occur. Table B.17 illustrates the valid object calls
for the Session_OnEnd event.

Table B.17 Session_OnEnd Object Calls

Valid Objects Invalid Object Calls
Server Request

Session Response

Application

Controlling User Session Resources

The Abandon method is used to destroy and release the resources consumed by inactive
Session objects. The termination of the Session object is called using the following syntax:

Session.Abandon

where Session is the current Session object. In most situations, the Abandon method will be
executed implicitly when the Session.Timeout value has passed.

The Session.Abandon method has some special scoping characteristics that are important to
remember. First, when the Session.Abandon method is called, the Web server continues to
process the remaining ASP code on the page. This is unlike the Server.TimeOut event and the
Response.Redirect method, where code execution on that page is instantly terminated.

The second scoping issue of the Session object is that all object properties that were created
before the Session.Abandon method will be unavailable after the Abandon method is called.
The Abandon method destroys the properties of the Session object without destroying the
Session object itself.

Third, because the Session object is not yet destroyed, the Session object can be used to store
user properties. Only after the page has completely finished processing is the current Session
object placed in the queue for destruction.

42.1389-6 AppB 758 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 Managing the Session Object 759

Using Session-Level Objects

Session-level objects can be categorized into two classes based on how the item is declared, as
shown in Table B.18. Objects can be declared with or without the <OBJECT> tag. If the session-
level object is declared as an object using the <OBJECT> tag, the object is managed by the
StaticObject collection. If the object is not declared using the <OBJECT> tag, the session-level
item is managed by the Contents collection.

Table B.18 The Session Collection

App
Name Description
Contents Contains all session-level items that have not been created B
with the <OBJECT> tag.
StaticObject Contains all session-level objects declared with the <OBJECT>
tag.

The Contents Collection The Contents collection is used to contain all items that have been
created during a session that have not been declared with the <OBJECT> tag. The Contents
collection requires the following syntax:
Session.Contents(Key)

where Key is the name of the property to retrieve.

The StaticObject Collection The StaticObject collection is similar to the Contents collec-
tion, but StaticObject is used to manage all session-level objects that have been declared with
the <OBJECT> tag. The StaticObject collection uses the following syntax:
Session.StaticObjects(Key)

where Key is the property to retrieve. StaticCollection can also be used to reference a single
session variable or multiple session variables.

42.1389-6 AppB 759 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 760 Appendix B Active Server Object Reference

42.1389-6 AppB 760 10/26/98, 10:18 AM

Brands03/Stone3 SEU Active Server Pages #1389-6 Sring 11/13/97 AppB lp#3
 C
761

APPENDIX

ADO Reference

App
This reference summarizes the various objects and C
collections of ActiveX Data Objects. Along with these
collections and objects, this reference also lists the
various properties and methods of these objects and
collections. ■

43.1389-6 AppC 761 10/26/98, 10:18 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 762 Appendix C ADO Reference

Connection Object
The Connection object represents a connection to an OLE DB datasource.

Properties
Attributes

Used to set when and if to start a new transaction after a RollbackTrans or CommitTrans is
called. It can be set to either of the following XactAttributeEnum values, or the sum of both:

adXactCommitRetaining Starts a new transaction after a CommitTrans is called.

AdXactAbortRetaining Starts a new transaction after a RollbackTrans is called.

CommandTimeout

Specifies how long to wait in seconds for a command to execute before giving up and return-
ing an error. The default is 30. When set to 0, the wait is infinite.

ConnectionString

A string that contains the information necessary to connect to the datasource. This string is
made up of up to five optional name-value pairs, or a single datasource name (DSN). All other
name-value pairs are passed on directly to the provider.
Provider The name of the provider to use for the connection.
DSN The name of the ODBC datasource name to connect to.
User The username to connect with.
Password The password to log into the datasource with.
File The name of a file that contains connection-specific information.

ConnectionTimeout

Specifies how long to wait in seconds when trying to establish a connection before giving up
and returning an error. The default is 15. When set to 0, the wait is infinite.

DefaultDatabase

Specifies the default database for the connection. This is used when there are multiple data-
bases for a provider. When you set this value, SQL statements will not have to specify a database.

IsolationLevel

Specifies the level of isolation to apply to transactions in the connection. Can be set to any of
the following IsolationLevelEnum values:

adXactUnspecified This value is returned when ADO can’t determine the

IsolationLevel present in the provider.

adXactChaos Specifies that transactions can’t overwrite pending values

of more isolated transactions.

43.1389-6 AppC 762 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Connection Object 763

adXactBrowse Specifies that transactions can view uncommitted changes

in other transactions.
adXactReadUncommitted This value is the same as adXactBrowse.
adXactCursorStability The default value. Specifies that changes in other transac-
tions are readable only after they have been committed.
adXactReadCommitted This value is the same as adXactCursorStability.
adXactRepeatableRead Specifies that changes from other transactions are not
viewable until after commits. However, re-querying will
result in new recordsets.
adXactIsolated Indicates that all transactions are isolated from each
other.
adXactSerializable This value is the same as adXactIsolated.

Mode
App
This property sets the permissions for changing, deleting, and adding data through the con-
nection. It can be any of the following ConnectModeEnum values. C
adModeUnknown Indicates that the permissions cannot be determined.
adModeRead Indicates read-only permissions.
AdModeWrite Indicates write-only permissions.
AdModeReadWrite Indicates read and write permissions.
AdModeShareDenyRead Prevents others from opening the connection with read
permissions.
AdModeShareDenyWrite Prevents others from opening the connection with write
permissions.
AdModeShareExclusive Prevents others from opening the connection with read
and write permissions.

Provider

This property indicates the OLE DB provider to use for the connection. The default is
MSDASQL, which is the ODBC provider for OLE DB.

Version

This property returns the version of ADO in use.

Methods
BeginTrans

Syntax:
intTransactionLevel = dataConn.BeginTrans

43.1389-6 AppC 763 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 764 Appendix C ADO Reference

Used to start a new transaction. While the transaction is under way (until CommitTrans or
RollBackTrans is called), no changes can be written to the database. This method returns an
integer that indicates the transaction’s level, if the provider supports nested transactions.

Close

Syntax:
dataConn.Close

Closes the open connection, and releases all resources associated with the connection. This
method also closes any recordsets of the connection.

CommitTrans

Syntax:
dataConn.CommitTrans

Commits the transaction’s changes to the database.

Execute

Syntax:
Set rs = dataConn.Execute(CommandText, RecordsAffected, Options)

Executes a SQL statement, query, or stored procedure. This method returns a recordset. It can
be sent the following optional parameters:

CommandText A string that is the name of the query, stored procedure,

or SQL statement to execute.
RecordsAffected An integer that represents the number of records affected
by the operation.
Options Specifies what the CommandText parameter is. This value
is a CommandTypeEnum.

Open

Syntax:
dataConn.Open ConnectionString, UserID, Password

Opens the connection. Can be sent any of the following optional parameters:

ConnectionString A string containing connection information.

UserID The username to use when opening the connection.
Password The password to use when opening the connection.

RollbackTrans

Syntax:
dataConn.RollbackTrans

Reverses any changes made during the transaction.

43.1389-6 AppC 764 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Command Object 765

Errors Collection and Object

The Errors collection is part of the Connection object. The collection is a group of all the error
objects generated while connected to the datasource. An Error object represents an error.

Properties
Description

A short description of the error generated by either the provider or ADO.

HelpContext

Specifies the Context ID in a Microsoft help file that contains information on the error. Used in
connection with the HelpFile property.

HelpFile

Specifies the Microsoft help file that contains more information on the error. App
NativeError C
Returns a provider-specific error code.

Number

Returns a unique identification number for that error.

Source

Specifies the object or application that originally generated the error.

SQLState

Returns the ANSI SQL state error code for the error.

Command Object
This object is used to create and store a specific command to be executed on the datasource.

Properties
ActiveConnection

Specifies the connection object to use for the Command object. This also can be a valid connec-
tion string, in which case a connection object is created implicity.

CommandText

A string that is the command to be executed against the provider. This can be a SQL statement,
table name, or the name of a stored procedure.

43.1389-6 AppC 765 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 766 Appendix C ADO Reference

CommandTimeout

Specifies how long to wait in seconds for the command to execute before timing out and gener-
ating an error. The default is 30. If set to 0, the wait is infinite.

CommandType

Specifies the type of command that is contained in the CommandText property. Can be any of the
following CommandTypeEnum values:

adCmdText Indicates a SQL statement.

adCmdTable Indicates a table name.
adStoredProc Indicates a stored procedure.
adCmdUnknown Indicates that CommandText has an unknown type.

Prepared

Specifies whether ADO should create a compiled version of the CommandText before executing it.

Methods
CreateParameter

Syntax:
Set paramObj = dataCmd.CreateParameter(Name, [Type, Direction, Size, Value])

Used to create a new parameter object for the parameter collection. It can be sent five fields, all
of which are optional, except for name. The other values are

Type Indicates the datatype of the parameter.

Direction Specifies whether the parameter is an input, output, or
both.
Size Specifies the maximum size in bytes the parameter
can be.
Value The value of the parameter.
Execute

Syntax:
Set dataRS = dataCmd.Execute([RecordsAffected, Parameters, Options])

Similar to the connection object’s Execute method, this method executes the command found
in the CommandText property against the data provider. It can be sent the following optional
parameters:

RecordsAffected An integer that returns the number of records affected by

the command.
Parameters An array of parameters to be sent along with the com-
mand.

43.1389-6 AppC 766 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Parameters Collection and Objects 767

Options Specifies the type of command found in CommandText. See

the CommandType property for specifics of values.

Parameters Collection and Objects

The Parameters collection is a set of Parameter objects. Parameter objects represent param-
eters to be sent to a data provider for parameterized queries.

Properties
Attributes

Specifies what type of data the parameter accepts. Can be any combination of the following
ParameterAttributesEnums:

adParamSigned Specifies whether the parameter accepts signed values.

adParamNullable Specifies whether the parameter accepts null values. App
adParamLong Specifies whether the parameter accepts long values. C
Direction

Specifies if the parameter is an input, output, or both. Can be any of the following
ParameterDirectionEnum values:

adParamInput Input only.

adParamOutput Output only.
adParamInputOutput Both input and output.
adParamReturnValue Return value.

Name

Specifies the name of the parameter.

NumericScale

Indicates the number of decimal places the parameter can contain.

Precision

Specifies the maximum number of digits the parameter can contain.

Size

Specifies the maximum size in bytes the parameter may be.

Type

Specifies the datatype of the parameter.

Value

Specifies the value of the parameter.

43.1389-6 AppC 767 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 768 Appendix C ADO Reference

Methods
AppendChunk

Syntax:
paramObj.AppendChunk Data

This method is used to append a chunk of binary data to the parameter.

Append

Syntax:
dataCmd.Append ParamObj

This method is used to append a new Parameter object to the Parameters collection of the
Command object.

Delete

Syntax:
dataCmd.Delete Index

Used to delete a specific parameter from the Parameters collection. Index is the index of the
parameter object.

Recordset Object
The Recordset object is the set of all records returned from an executed command.

Properties
AbsolutePage

Specifies which page to move the current record pointer to.

AbsolutePosition

Specifies which record to move the current record pointer to.

ActiveConnection

Specifies which connection object the recordset is part of.

BOF

Specifies that the current record pointer is at the beginning of the recordset.

Bookmark

Specifies the bookmark for the record that has been bookmarked.

43.1389-6 AppC 768 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Recordset Object 769

CacheSize

Specifies the number of records from the recordset that are cached.

CursorType

Specifies the type of cursor to use in the recordset. Can be any of the following
CursorTypeEnum values:

adOpenForwardOnly Forward-only, read-only cursor.

adOpenKeyset Like the dynamic cursor, except that records added by
others are inaccessible.
adOpenDynamic A fully scrollable read-write cursor capable of seeing the
changes made by other users.
adOpenStatic This cursor retrieves a copy of the recordset. Changes
made by others are inaccessible.

EditMode App

Returns the editing status of the recordset. Can be any of the following editModeEnum values: C
adEditNone Indicates no editing is taking place.
adEditInProgress Indicates a change has been made in the current record,
but it hasn’t been saved to disk yet.
adEditAdd Indicates the AddNew method is currently taking place.

EOF

Specifies that the current record pointer is past the last record in the recordset.

Filter

Specifies a filter to apply to the recordset. Can be a string of AND clauses, an array of book-
marks, or any of the following FilterGroupEnum values:

adFilterNone Removes the current filter.

adFilterPendingRecords Shows all records that have changed but have not yet
been written to disk.
adFilterAffectedRecords Shows all records affected by the last Delete,
CancelBatch, UpdateBatch, or Resync.

adFilterFetchedRecords Shows the records currently in the cache.

LockType

Specifies the type of locking to apply to the recordset when it is open. Can be any of the follow-
ing LockTypeEnum values:

adLockReadOnly A read-only recordset

adLockPessimistic Each record is locked as it becomes the current record.

43.1389-6 AppC 769 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 770 Appendix C ADO Reference

adLockOptimistic Each record is locked only when the Update method is

called.
adLockBatchOptimistic A batch of records that are locked only when the
BatchUpdate method is called.

MaxRecords

Specifies the maximum number of records the recordset may contain.

PageCount

Returns the number of pages in the recordset.

PageSize

Specifies the number of records that a single page contains.

RecordCount

Returns the number of records in the recordset.

Source

Specifies the source (SQL statement, stored procedure, or table name) from which the records
came.

Status

Returns the status of the recordset.

Methods
AddNew

Syntax:
rs.AddNew Fields, Values

Adds a new record with the values specified.

CancelBatch

Syntax:
rs.CancelBatch recordsAffected

Cancels a batch update. Can return the number of recordsAffected by the cancel.

CancelUpdate

Syntax:
rs.CancelUpdate

Cancels the update.

43.1389-6 AppC 770 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Recordset Object 771

Clone

Syntax:
Set newRS = oldRS.Clone

Returns a copy of the original recordset.

Close

Syntax:
rs.Close

Closes the recordset.

Delete

Syntax:
rs.Delete recordsAffected
App
Deletes the current record.
C
GetRows

Syntax:
Set myArray = rs.GetRows(rows, start, fields)

Sets an array with a series of records from the recordset.

Move

Syntax:
rs.Move NumRecs, start

This method moves the current record pointer to the number of records specified.

MoveFirst

Syntax:
rs.MoveFirst

Moves the current record pointer to the first record within the recordset.

MoveLast

Syntax:
rs.MoveLast

Moves the current record pointer to the last record in the recordset.

MoveNext

Syntax:
rs.MoveNext

43.1389-6 AppC 771 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 772 Appendix C ADO Reference

Moves the current record pointer to the next record within the recordset.
NextRecordset

Syntax:
set newRs = oldrs.NextRecordset(recordsAffected)

Moves to the next recordset. This method can also return the recordsAffected by the move.

Open

Syntax:
rs.Open [Source, ActiveConnection, CursorType, LockType, Options]

Opens the recordset. This method can be passed the following optional parameters: Source,
ActiveConnection, CursorType, LockType, and Options.

Requery

Syntax:
rs.Requery

Requires the database and updates the recordset.

Resync

Syntax:
rs.Resync recordsAffected

Refreshes the recordset by re-syncing it with the database. This method can be passed a
recordsAffected parameter that will return the number of records affected by this method.

Supports

Syntax:
set bool = rs.Supports(cursorOption)

Used to determine what functionality the recordset supports. This method is passed a variable
cursorOption that is the option you wish to see if the provider supports it.

Update

Syntax:
rs.Update Fields, Values

Used to update the current record with the values specified for the fields specified.

UpdateBatch

Syntax:
rs.UpdateBatch recordsAffected

Used to write a batch update to disk. Can return the number of recordsAffected.

43.1389-6 AppC 772 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 Fields Collection and Objects 773

Fields Collection and Objects

A field is a specific column of data in a recordset. The Fields collection is the set of all field
objects in the recordset.

Properties
ActualSize

Returns the actual length of the field’s value.

Attributes

Specifies the attributes of the field.

DefinedSize

Returns the defined size for the field.

Name App

Returns the field’s name. C

NumericScale

Specifies the number of decimal points the field may contain.

OriginalValue

Returns the original value of the field’s value before it was updated.

Precision

Returns the number of digits the field’s value may contain.

Type

Returns the datatype of the field.

UnderlyingValue

Returns the value of the field in the database.

Value

Specifies the value of the field.

Methods
AppendChunk

Syntax:
field.AppendChunk data

Used to append a chunk of binary data to the field’s value.

43.1389-6 AppC 773 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 774 Appendix C ADO Reference

GetChunk

Syntax:
field.GetChunk(NumberOfBytes)

Used to retrieve a chunk of binary data from the field.

43.1389-6 AppC 774 10/26/98, 10:19 AM

Brands3/Stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 App C lp#3
 D
775

APPENDIX

Active Server
Component Reference

App

44.1389-6 AppD 775 10/26/98, 10:19 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 776 Appendix D Active Server Component Reference

Ad Rotator Component
Syntax:
Set ad = Server.CreateObject(“MSWC.AdRotator”)

The Ad Rotator component is used to rotate advertisements on a page.

Properties
Border

Syntax:
ad.Border(size)

Specifies the border size of the ad.

Clickable

Syntax:
ad.Clickable(bool)

Specifies whether the ad is clickable.

TargetFrame

Syntax:
ad.TargetFrame(frame)

Specifies the target frame for the ad link to be loaded in.

Methods
GetAdvertisement

Syntax:
ad.GetAdvertisement(pathAndFile)

This method retrieves the ad to be displayed from the list of ads in the file specified.

Files
Redirection File

This is a file the developer creates to capture which ads are being clicked, and then redirect to
the appropriate link.

Rotator Schedule File

Syntax:
Redirect URL
Width intWidth
Height intHeight

44.1389-6 AppD 776 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 Browser Capabilities Component 777

Border intBorder
*
adImageURL
adHomePageURL
Text
impressions

This file is used to set the properties of the various ads to be displayed. It has two sections:
The first section contains properties that pertain to all ads; the second section contains ad-
specific information.

Browser Capabilities Component

Syntax:
Set Browser = Server.CreateObject(“MSWC.BrowserType”)

This component is used to test whether the Web surfer’s browser supports certain functional-
ity. The following are some of the possible properties that can be tested for:

■ ActiveX controls
■ Beta
■ Browser
■ Cookies
App
■ Frames
■ JavaScript
D
■ Tables
■ VBScript
■ VBScript version

Each of these can become properties of this object. To see if the browser supports JavaScript,
you simply ask for that property’s value.

Files
BROWSCAP.INI

This file contains information on browsers. It is used to determine their capabilities.

Syntax:
; comments
HTTPUserAgentHeader
parent = browserDefinition
propertyX = valueX

Comments are any line that begins with a semi-colon (;). HTTPUserAgentHeader is the header for
any specific browser, and it is used to identify the browser. BrowserDefinition is used to
specify a parent browser that will inherit its functionality. For each property there is a corre-
sponding value (such as JavaScript = False.)

44.1389-6 AppD 777 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 778 Appendix D Active Server Component Reference

Content Linking Component

Syntax:
Set NextURL = Server.CreateObject(“MSWC.NextURL”)

This component is used to link URLs together in a like pages in a book or magazine.

Methods
GetListCount

Syntax:
NextURL.GetListCount(listFile)

Returns a count of all the URLs in the listFile.

GetNextURL

Syntax:
NextURL.GetNextURL(listFile)

Returns the next URL in the list in the listFile.

GetPreviousDescription

Syntax:
NextURL.GetPreviousDescription(listFile)

Returns a description of the previous URL in the listFile.

GetListIndex

Syntax:
NextURL.GetListIndex(listFile)

Returns the indexed position of the current page as found in the listFile.

GetNthDescription

Syntax:
NextURL.GetNthDescription(listFile, index)

Returns the description of the URL in the listFile at the position specified with the index
variable.

GetPreviousURL

Syntax:
NextURL.GerPreviousURL(listFile)

Returns the previous URL in the listFile.

44.1389-6 AppD 778 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 File Access Components 779

GetNextDescription

Syntax:
NextURL.GetNextDescription(listFile)

Returns the description of the next URL in the listFile.

GetNthURL

Syntax:
NextURL.GetNthURL(listFile, index)

Returns the URL in the index position of the listFile.

Files
Content Linking File

This file contains a list of URLs along with a description of each one.

Syntax:
URL Description

where URL is the URL of the next link (note that this URL cannot contain http:// or \\ ), and the
description is a string describing that URL.
App

D
File Access Components
File access with ASP is provided by two VBScript objects: FileSystem and TextStream.

FileSystemObject
Syntax:
Set filesys = CreateObject(“Scripting.FileSystemObject”)

This object is used to gain access to the file system. It can access existing files and create
new ones.

Properties
Drives

Lists the drives on the local system available to the object. You can iterate through this prop-
erty with the For...Each...Next construct.

Methods
CreateTextFile

Syntax:
filesys.CreateTextFile(filename, [overwrite, unicode])

44.1389-6 AppD 779 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 780 Appendix D Active Server Component Reference

This method is used to create a file. It is passed one required parameter, filename, and two
optional parameters: overwrite, which is a bool value on whether to overwrite an existing file if
it exists, and unicode, another bool, which, if True, writes the files as unicode.

OpenTextFile

Syntax:
filesys.OpenTextFile (filename, [iomode, create, format])

This method is used to open a file. It returns a TextStream object. It requires the filename
parameter, which indicates which file to open. Optional parameters are iomode, which specifies
whether the file is being opened ForReading or ForAppending; create, which is bool that indi-
cates whether or not to create the file if it doesn’t exist; and format, which indicates whether to
open the file as unicode, or ascii, or use the system default.

TextStream
Syntax:
texts = filesys.CreateTextFile(“c:\somefile.txt”, True)

TextStream is the object used to actually write and read from a file.

Properties
AtEndOfLine

This property returns a value of True if the file pointer is at the end of a line.

AtEndOfStream

This property returns True if the file pointer is at the end of the file.

Column

This property returns the column that the file pointer is currently on.

Line

This property returns the current line that the file pointer is on.

Methods
Close

Syntax:
Texts.Close

This method closes the file.

Read

Syntax:
Texts.Read(intNumChars)

44.1389-6 AppD 780 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 File Access Components 781

This method reads in the number of characters specified in intNumChars from the current
file pointer.

ReadAll

Syntax:
Texts.ReadAll

This method reads in the entire file and returns it as a string.

ReadLine

Syntax:
texts.RealLine

This method reads in the entire line and returns it as a string.

Skip

Syntax:
texts.Skip(intNumChars)

This method moves the current file pointer forward the number of characters specified in
the intNumChars variable.

SkipLine App
Syntax: D
texts.SkipLine

This method moves the current file pointer to the next line in the file.

Write

Syntax:
texts.Write(string)

This method writes out the string to the file at the position of the current file pointer.

WriteLine

Syntax:
texts.WriteLine(string)

This method writes out the string with a newline character added at the end. If string is
omitted, the newline character is written out alone.

WriteBlankLines

Syntax:
texts.WriteBlankLine(intNumLines)

This method writes out the number of blank lines specified in the intNumLines parameter.

44.1389-6 AppD 781 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 782 Appendix D Active Server Component Reference

44.1389-6 AppD 782 10/26/98, 10:20 AM

Brands03/stone3 SEU Active Server Pages #1389-6 SRing 11/13/97 AppD lp#3
 783

Index

Symbols
accessing objects
%> tag (HTML), 57
application-level arrays, Application, 192
<SCRIPT> tag, 26 321 ObjectContext, 192
@CodePage directive, databases, 518 Request, 193
206 accounts (Transaction Response, 193
@EnableSessionState Server components), Server, 192
directive, 206 605 Session, 193
calling, 606 planning, 30
@Language directive, 206
scripting, 24
@LCID directive, 207 ACTION method, 422
scripts, 24
@Transaction directive, activating GLOBAL.ASA, setting, 25
207 187 security, viewing code,
Active Data Object 208
(ADO), 347 versions, 208
A Active Messaging see also ASP
Component, 348 ActiveConnection
Abandon method
Active Server Pages property
Session object resources,
configuring, 199 Command object, 471,
259
Metabase, 199 765
Session_OnStart, 758
Registry, 201 Recordset object, 768
Abs( ) function
debugging, 193 ActiveX, 24
(VBScript), 710
Err object, 194 browser compatibility, 24
AbsolutePage property FileSystem object, 195 components, 383
Recordset object, 768 log files, 195 creating, 383-388
AbsolutePosition Registry settings, 195 datatype differences,
property defined, 9 389-392
Recordset object, 492, delimiters, 25 encapsulating
768 Internet Information calculations, 394-398
access Server, 30 Object Browser, 386
ASP applications, 197 issues, 208 SIMPLECLASS listing,
Database Object memory leaks, 208 385
Component, 675
security, 555

45.1389-6 index 783 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 784 ActiveX

TESTMULT.ASP AddProductInventory AdvancedDataSpace

listing, 387 method, 689 object, 526
TESTMULT.BAS ADO (Active Data aging (message queuing),
listing, 386 Object), 347, 430 617
controls access security, 555 ampersand operator (&),
VBScript, 52-54 Cold Fusion, 432 84
non-visual controls, 523 collections, 433
server-side controls Append method
Command object, 470
VBScript, 54-55 Parameters object, 768
Connection object, 452
wrapping functions, cursors, 487 AppendChunk method
392-394 dynamic, 489 Fields collection, 773
CONVERTITTEST1 forward-only, 488 Parameters object, 768
listing, 393 KeySet, 489 AppendToLog method
listing, 392, 393 static, 488 Response object, 289, 748
ActiveX Control Pad, 56 error handling, 463 application events,
ActualSize property inventory, 540 trapping, 212
Fields collection, 773 adding items, 542
deleting items, 540 application initialization
Ad Rotator Component, component (databases),
naming conventions, 444
336 674
object model, 432
banners, 414 Application object, 186,
objects, 433
files, 337 192, 212, 736
Command, 437
redirection, 339, 776 Application_OnEnd
Connection, 434
Schedule, 338, 777 event, 217
Field, 441
GetAdvertisement arrays, 222
Parameter, 438
method, 336, 776 collections, 226
Recordset, 440
properties, 340 Contents, 226
OLE DB, 430
Border, 776 StaticObject, 227
Recordset, 443
Clickable, 776 components, 220
recordset object, 486
TargetFrame, 776 declaring variables, 214
Remote Data Services
AddHeader method (RDS), 447 events
Response object, 289, 748 rollback processing, 446 Application_OnEnd,
adding inventory items search program, 448 737
(ADO), 542 transaction processing, Application_OnStart,
446 736
Additem.Asp (non-visual
Visual InterDev, 543 instantiating at the
Active Server Page),
Web applications, 431 application-level, 220
325
see also databases methods, 218, 737
addition operator (+), 70, CreateObject, 739
85 ADO object, 536
HTMLEncode, 740
Connection, 536
AddNew method Lock, 218, 737
LoadProductsTable, 536
Recordset object, 499, MapPath, 739
770 AdvancedDataControl ScriptTimeOut, 739
object, 525 UnLock, 219, 737
AddProductInventory
function, 542 AdvancedDataFactory URLEncode, 740
object, 526 OnStart event, 217

45.1389-6 index 784 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 BeginTrans command 785

Application_OnEnd event Array( ) function GLOBAL.ASA, 184

(Application object), (VBScript), 710 activating, 187
217, 737 arrays, 128 scoping, 190
Application_OnStart Application object, 222 HTML, 182
event (Application application-level, 318 interface utilities
object), 736 accessing, 321 (servers), 204
Session object, 262 procedure-level scoping,
application-level
session-level (online 191
arrays, 318
catalogs), 320 scripts, 182
accessing, 321
session-level scoping, 191
scoping, 190 Asc( ) function
text files, 180
variables (VBScript), 710
tracking session events,
declaring, 214 ASP 187
GLOBAL.ASA, 318 applications, accelerating, tracking user events, 186
applications 405-408
AtEndOfLine property
ASP accelerating, 405-408 listing, 405-408
File Access component,
Cold Fusion (ADO), 432 Database Object
780
intranets, 640 Component, 684
porting Visual Basic to intrinsic objects, 401-408 AtEndOfStream property
VBScript, 398-401 Message Queue Server, File Access component,
stateless, 431 629 780
Web-based applications, Microsoft Management Atn( ) function
12 Console (MMC), 670 (VBScript), 710
architectures, 43 parameter passing, 611 attributes
client/server, 44 Remote Data Services, Domain (cookies), 271
n-tier client/server 524 Expires (cookies), 270
model, 44 Transaction Server, 597 HasKeys (cookies), 272
n-tier Web access, 45 Transaction Server Path (cookies), 272
terminal access, 44 components, 610 RUNAT, 26
Web browsers, 45 transaction support, 240 Secure (cookies), 272
see also Active Server
argument lists (VBScript Attributes property
Pages
procedures), 115-117 Connection object, 462,
ASP applications 762
arithmetic operators, 70
access control, 197 Fields collection, 773
addition operator (+), 70
file level, 198 Parameter object, 478
concatenation operators,
packet filtering, 197 Parameters object, 767
84
application-level scoping,
division operator (/), 72
190
exponentiation operator
(^), 72
directives, 205 B
@CodePage, 206
integer division operator
@EnableSessionState,
(^), 73
206 banners, Ad Rotator
modulus operator
@Language, 206 Component, 336, 414
(MOD), 74
@LCID, 207
multiplication operator BeginTrans command,
@Transaction, 207
(*), 72 33
directories, 181
subtraction operator (-),
72

45.1389-6 index 785 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 786 BeginTrans method

BeginTrans method Buffer property CDbl function, 139

Connection object, 764 ContentType object, 280 VBScript, 711
BinaryRead method Expires object, 280 certificate servers, 17
Request object, 313, 754 Response object, 274, 745
CGI (Common Gateway
BinaryWrite method business objects, 523 Interface), 18
Response object, 290, 749 buttons (navigational), CharSet property
BOF property 423 Response object, 277, 745
Recordset object, 768 ChooseContent method
BOF/EOF property Content Rotator
Recordset object, 490
C Component, 373
Bookmark property Chr( ) function
Recordset object, 768 CacheControl property (VBScript), 711
Border property Response object, 276, 745 CInt function, 139
Ad Rotator Component, CacheSize property CInt( ) function
776 Recordset object, 769 (VBScript), 711
breakpoints (Script caching, VTM (Virtual Clear method
Debugger), 167 Table Manager), 524 Response object, 290, 749
BROWSCAP.INI calculations (server-side Clickable property
(Browser Capabilities scripts), 153 Ad Rotator Component,
Component), 342, 776
calculations,
417, 777
encapsulating (AX client-server architecture,
Browser Capabilities components), 394-398 44
Component listing, 395, 396, 397 client-side controls, see
(BROWSCAP.INI), 342,
Call statement (VBScript controls
417, 777
subroutines), 118-119 client-side scripts
properties, 345
calling accounts testing for browser
browsers, 13
(Transaction Server), compliance, 158
ActiveX compatibility, 24
606 validation subroutine,
architectures, 45
calling GLOBAL.ASA, 156-157
detecting, 417
187 VBScript, 155
IE security, 59
intranets, 641 CancelBatch method client-side VBScript
object models, 20 Recordset object, 770 versus server-side, 57
dynamic HTML object CancelUpdate method client/server
model, 22 Recordset object, 770 architecture, 13
Remote Data Services, databases, 509
CBool function, 137
527 ClientCertificate
VBScript, 710
routing, 26 collection
testing for scripting CByte function, 138
Request object, 300, 750
compliance, 158 VBScript, 711
CLng function, 139
user groups, 13 CCur function, 138
VBScript, 711
buffer (Clear method), VBScript, 711
290 Clone method
CDate function, 139
Recordset object, 771
VBScript, 711

45.1389-6 index 786 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 components 787

Close method Ad Rotator Component, CommandTimeout

Connection object, 764 336 property
File Access component, Browser Capabilities Command object, 474,
780 Component, 342 766
Recordset object, 771 Browser Capability Connection object, 762
Close property Component, 417 CommandType property
Connection object, 459 Content Linking Command object, 473,
Component, 360 766
closing queues (Message
Content Rotator
Queue Server), 628, CommitTrans method
Component, 371, 411
636 Connection object, 764
DCOM (Distributed
code delimiters, 25 COM), 23 Common Gateway
code libraries (non-visual Page Counter Interface (CGI), 18
Active Server Page), Component, 375 Common Object Model
325 Permission Checker (COM), 23
code stepping (Script Component, 357 comparison constants
Debugger), 167 Tools Component, 354 (VBScript), 726
CodePage property Command object, 470, comparison operators,
Session object, 256, 757 765 75
ADO, 437 equality operator, 77
coding
collections greater than operator (>),
ADO object model, 445
Parameters, 476 78
visual controls, 447
Connection object, 465 inequality operator (<>),
Cold Fusion (ADO), 432 creating, 470 78
collections methods less than operator (<), 78
ADO, 433 CreateParameter, 766 object comparison
Application object, 226 Execute, 475, 766 operator (IS), 79
Contents, 226 Parameters object, 476
component architecture,
StaticObject, 227 properties
ActiveConnection, 471, 30
Command object
765 scalability, 31
Parameters, 476
Connection object CommandText, 765 component threading
Error, 463 CommandTimeout, model
Errors, 765 474, 766 Session object, 262
Cookie, 268 CommandType, 473, components, 27
color constants 766 Application object, 220
(VBScript), 725 Prepared, 474, 766 Microsoft Management
recordsets, 475 Console (MMC), 670
Column property strings (CommandText), object-oriented
File Access component, 472 programming (OOP),
780
Command objects 30
COM (Common Object (properties), 471 properties, 239, 240
Model), 23 registering, 32
Active Messaging CommandText property
Transaction Server, 32,
Component, 348 Command object, 765
600, 603
ActiveX, 24 CommandText string
Command object, 472

45.1389-6 index 787 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 788 components

accounts, 605 collections constants

ASP, 610 Error, 463 VBScript, 725
calling accounts, 606 Errors, 765 color, 725
methods, 604 Command object, 465 comparison, 726
threading, 610 configuring, 456 date format, 727
transactions, 32 Error collection, 435 date/time, 726
see also component linking data, 454 file input/output, 728
architecture, 30 methods, 453 miscellaneous, 728
components (AX), 383 BeginTrans, 764 MsgBox, 728
listing, 402-408 Close, 764 string, 730
creating, 383-388 CommitTrans, 764 tristate, 730
Object Browser, 386 Execute, 460, 764 VarType, 731
datatype differences, Open, 764 VBScript message box
389-392 RollTransBack, 764 icons, 123
listing, 389-390 phone application, 465 VBScript message box
encapsulating properties, 452 results, 124
calculations, 394-398 Attributes, 462, 762 Content Linker
SIMPLECLASS listing, Close, 459 Component
385 CommandTimeout, (registration), 420
TESTMULT.ASP listing, 762 Content Linking
387 ConnectionString, 456, Component, 360, 778
TESTMULT.BAS listing, 762 Content Linking list, 361
386 ConnectionTimeout, methods
457, 762 GetNthURL, 779
concatentation operators,
DefaultDatabase, 458, GetListCount, 778
84
762 GetListIndex, 778
addition operator (+), 85
IsolationLevel, 458, GetPreviousURL, 778
ampersand operator (&),
486, 762 NextLink.DLL, 360
84
Mode, 456, 763
concurrency (databases), content linking file, 779
Open, 459
512 Provider, 763 Content Linking list
conditional statements Version, 763 (Content Linking
(VBScript), 704 transaction processing, Component), 361
configuring 461 Content Linking object
Active Server Pages, 199 connections (Transaction methods, 362
Metabase, 199 Server), 593 GetListCount, 363
Registry, 201 GetListIndex, 364
ConnectionString
Connection GetNextURL, 364
property
properties, 456 GetNthDescription, 368
Connection, 456
Session object, 261 GetNthURL, 365
Connection object, 762
GetPrevious, 368
conjunction operator ConnectionTimeout GetPreviousURL, 365
(AND), 81 property
Content Rotator
Connection object, 452, Connection object, 457,
Component, 371, 411
762 762
Content Schedule file,
ADO, 434
371
ADO object, 536

45.1389-6 index 788 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 databases 789

methods, 373 CreateObject method Database Designer

ChooseContent, 373 Application object, 739 (Visual InterDev), 572
GetAllContent, 374 Server object, 230 Database Object
Contents collection CreateObject( ) function Component, 675
Session object, 265 (VBScript), 712 ASP, 684
Session_OnStart, 759 CreateParameter method DCOMCFG, 682
Contents collections Command object, 766 error forms, 685
Application object, 226 methods
CreateParameter
DeleteProductInventory,
ContentType property property
690
Response object, 746 Parameter object, 476
AddProductInventory,
controls (VBScript), 52 CreateTextFile method 689
listing, 53-54 File Access component, GetProductInventory,
server-side, 54-55 780 688
conversion functions, CSng function, 139 online catalog, 693
136 VBScript, 712 registering, 678
CBool, 137 CStr function, 140 Regsrv32, 682
CByte, 138 VBScript, 712 testing components, 691
CCur, 138 Transaction Servers, 678
cursors
CDate, 139 validation servers, 686
ADO, 487
CDbl, 139 database servers, 15
dynamic, 489
CInt, 139 searches, 16
forward-only, 488
CLng, 139 Web servers, 16
KeySet, 489
CSng, 139
static, 488 databases, 506, 514
CStr, 140
CursorType property access
Cookie collection, 268 security, 555
Recordset object, 769
Response object, 744 accessing, 518
cookies ADO (Active Database
Domain attribute, 271 D Object), 347, 430
Expires attribute, 270 application initialization
HasKeys attribute, 272 Data Command (Visual component, 674
Path attribute, 272 InterDev), 574 client/server, 509
Secure attribute, 272 Data Form Wizard concurrency, 512
Session object, 249, 755 (Visual InterDev), 573 data modeling, 514
Cookies collection indexes, 517
data modeling
(Request object), 301, keys, 517
(databases), 514
751 locking, 487, 512
Data Range Footer object-oriented, 506
Cos( ) function
(Visual InterDev), 574 ODBC 3.0 (Open
(VBScript), 712
Data Range Header Database Connectivity
Counter Component, 378
(Visual InterDev), 574 3.0), 513
methods relational, 507
Get, 379 Data View (Visual
InterDev), 569 Remote Data Services,
Increment, 380
522
Remove, 381
security, 519
Set, 380

45.1389-6 index 789 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 790 databases

SQL, 507 dead-letter queues DeleteProductInventory

transaction processing, (message queuing), 618 function, 540
510 debugging, 193, 195 deleting inventory items
two-phase commit, 596 Err object, 194 (ADO), 540
Visual InterDev, 543-545, FileSystem object, 195 deleting queues
547, 549-553 Registry Settings, 195 (Message Queue
datatypes ScriptErrorsSentToBrowser Server), 629, 636
ActiveX components, object, 195
delimiters, 25
389-392 see also Script Debugger
listing, 389-391 delivering information,
decision structures
conversion functions, 136 server-side scripts, 152
(VBScript), 90
formatting functions, 133 If...Then, 90 Description property
Variant (VBScript), 60 If...Then...Else, 91 Errors collection, 765
subtypes, 61, 702 Select Case statement, 94 design-time controls
VarType function (Visual InterDev), 574
declarations, server-side
(VBScript), 62
scripts, 150 detecting browsers, 417
see also ADO
declaring development
date format constants
Application object environments
(VBScript), 727
variables, 214 Visual InterDev, 34
Date( ) function procedures, public/ Visual JavaScript, 38
(VBScript), 712 private (VBScript), dialog boxes (VBScript),
date/time constants 114-115 120
VBScript, 726 Session variables, 756 input boxes, 125-126
DateAdd( ) function variables message boxes, 120-124
(VBScript), 712 conventions, 66
Dim statement
Session object, 251
DateDiff( ) function (VBScript), 64
variables (VBScript), 64
(VBScript), 713 Direction property
Dim statement, 64
DatePart( ) function Option Explicit Parameter object, 478
(VBScript), 713 statement, 66 Parameters object, 767
DateSerial( ) function Private keyword, 65-66 directives
(VBScript), 713 Public keyword, 65 ASP applications, 205
DateValue( ) function ReDim function, 64-65 @CodePage, 206
(VBScript), 714 DefaultDatabase property @EnableSessionState,
Connection object, 458, 206
Day( ) function
762 @Language, 206
(VBScript), 714
DefinedSize property @LCID, 207
DCOM (Distributed @Transaction, 207
COM), 23 Fields collection, 773
Transaction, 240
DCOMCFG (Database Delete method
directories (ASP
Object Component), Parameters object, 768
applications), 181
682 Recordset object, 500,
771 disjunction operator
DCOMCNFG.EXE (OR), 82
(registration), 32 Delete property
Parameter object, 477 Distributed COM
(DCOM), 23

45.1389-6 index 790 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 events 791

distributed objects, 595 encryption errors

Distributed Transaction extranets, 650 ADO, 463
Coordinator (DTC), 512 Secure Socket Layers Script Debugger
(SSL), 17 logical, 174
distributed transaction
End method runtime, 173
processing, 587
Response object, 291, 749 syntax, 173
division operator (/), 72 trapping server-side
environment (Visual
Do...Loop loop scripts, 155
InterDev), 565
(VBScript), 99 Errors collection
EOF property
Domain attribute Connection object
Recordset object, 769
(cookies), 271 properties, 765
equality operator, 77 properties
Drive property
File Access component, equivalence operator Description, 765
779 (EQV), 83 HelpFile, 765
Err object, 108 HelpContext, 765
dynamic content, 10
debugging, 194 NativeError, 765
dynamic cursor, 489 Number, 765
properties, 107
dynamic HTML VBScript, 104 Source, 765
Internet Explorer, 524 SQLState, 765
Error (Connection
object model, 22 escape codes, 234
object), 435
stylesheets, 22
Fields, 773 event-driven
dynamic template, Parameters programming, 111-113
server-side scripts, 149 methods, 476 events, 113
properties, 476 Application object
Request object, 299, 750 Application_OnEnd,
E ClientCertificate, 300, 217, 737
750 Application_OnStart,
EditMode property
Cookies, 301, 751 736
Recordset object, 769
Form, 303, 751 ObjectContext object
Editor (Visual InterDev), QueryString, 305, 752 OnTransactionAbort,
577 ServerVariables, 309, 743
e-mail 752 OnTransactionCommit,
Active Messaging Response object, 268 742
Component, 348 Cookie, 744 OnStart (Application
see also message queuing Session object, 264 object), 217
encapsulating Contents, 265, 759 OnTransactionAbort,
calculations (AX StaticObject, 265, 759 242-244
components), 394-398 Error collection OnTransactionCommit,
listing, 395-397 Connection object, 435, 242-243
encoding methods 463 Session object, 256, 757
Server object, 234, 324 error forms (Database Session_OnEnd, 258,
HTMLEncode, 237 Object Component), 757
URLEncode, 235 685 Session_OnStart, 256,
757
Error object, 463

45.1389-6 index 791 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 792 exclusion operator

exclusion operator DefinedSize, 773 BROWSCAP.INI, 342

(XOR), 82 Name, 773 missing (Tools
exclusive locks, 512 NumericScale, 773 Component), 415
OriginalValue, 773 NextLink.DLL (Content
Execute method
Precision, 773 Linking Component),
Command object, 475,
Type, 773 360
766
UnderlyingValue, 773 FileSystem object
Connection object, 460,
Value, 773 debugging, 195
764
File Access component File Access components,
Exit statement
methods 779
VBScript, 102
Close, 780 Filter property
Exit statement (VBScript CreateTextFile, 780 Recordset object, 493,
procedures), 118 OpenTextFile, 780 769
Exp( ) function Read, 781
Filter( ) function
(VBScript), 714 ReadAll, 781
(VBScript), 714
Expires attribute ReadLine, 781
Skip, 781 filters (ISAPI), 20
(cookies), 270
SkipLine, 781 firewalls, 17
Expires property TextStream, 780
Response object, 746 Fix( ) function
Write, 781 (VBScript), 714
ExpiresAbsolute property WriteBlankLines, 781
Response object, 281, 746 Flush method
WriteLine, 781
Response object, 292, 749
Explain (VBScript dialog properties
box icon), 123 AtEndOfLine, 780 For Each...Next loop
AtEndOfStream, 780 (VBScript), 101
exponentiation operator
(^), 72 Column, 780 For...Next loop
Drives, 779 (VBScript), 96
external servers, 18 Line, 780 Form collection
extracting user File Access components, Request object, 303, 751
information, 329 779 FormatCurrency
extranets, 17, 649 objects function, 134
encryption, 650 FileSystemObject, 779
security, 649 FormatCurrency( )
file input/output function (VBScript),
constants 714
VBScript, 728
F file-level security, 198
FormatDateTime
function, 133
Field object (ADO), 441 file management (Visual FormatDateTime( )
Fields collection, 773 InterDev), 579 function (VBScript),
methods FileExists method 715
AppendChunk, 773 Tools Component, 354 FormatNumber function,
GetChunk, 774
files 136
properties
Ad Rotator Component, FormatNumber( )
ActualSize, 773
337 function (VBScript),
Attributes, 773
redirection, 339, 776 715
Schedule, 338, 777

45.1389-6 index 792 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 functions 793

FormatPercent function, CLng, 139 DateSerial( ), 713

135 CSng, 139 DateValue( ), 714
FormatPercent( ) CStr, 140 Day( ), 714
function (VBScript), DeleteProduceInventory, Exit statement, 118
715 540 Exp( ), 714
formatting, 133 Filter( ), 714
formatting functions, 133
FormatCurrency Fix( ), 714
FormatCurrency
function, 134 FormatCurrency( ),
function, 134
FormatDateTime 714
FormatDateTime
function, 133 FormatDateTime( ),
function, 133
FormatNumber 715
FormatNumber function,
function, 136 FormatNumber( ), 715
136
FormatPercent FormatPercent( ), 715
FormatPercent function,
function, 135 Hex( ), 715
135
identifier, 127 Hour( ), 716
forms IsArray, 128 InputBox(), 125-126,
error forms (Database IsDate, 128 716
Object Component), IsEmpty, 129 InStr( ), 716
685 IsNumeric, 131 InStrRev( ), 716
recordsets, 500 IsObject, 131 Int( ), 716
Request object, 328 VarType, 132 IsArray( ), 717
Forms Wizard (Visual IsNull, 129 IsDate( ), 717
InterDev), 551 ReDim (VBScript), 64-65 IsEmpty( ), 717
forward-only cursor, 488 VarType (VBScript), 62 IsNull( ), 717
VBScript, 114, 119, 705 IsNumeric( ), 717
forwarding (message
Abs( ), 710 IsObject( ), 717
queuing), 616
argument lists, 115-117 Join( ), 718
FrontPage, 35 Array( ), 710 LBound( ), 718
server-side scripts, 149 Asc( ), 710 LCase( ), 718
FTP (File Transfer Atn( ), 710 Left( ), 718
Protocol), 15 CBool( ), 710 Len( ), 718
sites (Microsoft CByte( ), 711 LoadPicture( ), 718
Management Console CCur( ), 711 Log( ), 719
(MMC)), 666 CDate( ), 711 LTrim( ), 719
Function statement CDbl( ), 711 Mid( ), 719
(VBScript), 119 Chr( ), 711 Minute( ), 719
CInt( ), 711 Month( ), 719
functions CLng( ), 711 MonthName( ), 719
AddProductInventory, Cos( ), 712 MsgBox( ), 120-124,
542 CreateObject( ), 712 720
conversion, 136 CSng( ), 712 nesting, 117
CBool, 137 CStr( ), 712 Now( ), 720
CByte, 138 Date( ), 712 Oct( ), 720
CCur, 138 DateAdd( ), 712 public/private, 114,
CDate, 139 DateDiff( ), 713 115
CDbl, 139 DatePart( ), 713 Replace( ), 720
CInt, 139

45.1389-6 index 793 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 794 functions

Right( ), 720 GetAdvertisement GetRows method

Rnd( ), 721 property Recordset object, 771
Round( ), 721 Ad Rotator Component, GLOBAL.ASA
RTrim( ), 721 776 activating, 187
ScriptEngine( ), 721 GetAllContent method application-level
ScriptEngineBuild Content Rotator variables, 318
Version( ), 721 Component, 374 ASP applications, 184
ScriptEngineMajor scoping, 190
GetChunk methods
Version( ), 721 session-level variables,
Fields collection, 774
ScriptEngineMinor 320
Version( ), 722 GetListCount method
Content Linking Gopher, 15
Second( ), 722
Sgn( ), 722 Component, 778 greater than operator (>),
Sin( ), 722 Content Linking object, 78
Space( ), 722 363 guaranteed message
Split( ), 722 GetListIndex method delivery (message
Sqr( ), 723 Content Linking queuing), 615
StrComp( ), 723 Component, 778
String( ), 723 Content Linking object,
StrReverse( ), 723 364 H
Tan( ), 723 GetNextDescription
Time( ), 723 HasAccess method
method, 779
TimeSerial( ), 724 Permission Checker
Content Linking object,
TimeValue( ), 724 Component, 358
367
Trim( ), 724 HasKeys attribute
GetNextURL method
TypeName( ), 724 (cookies), 272
Content Linking object,
UBound( ), 724 HelpContext property
364
UCase( ), 724 Errors collection, 765
VarType( ), 725 GetNthDescription
method HelpFile property
WeekDay( ), 725
Content Linking object, Errors collection, 765
WeekdayName( ), 725
Year( ), 725 368 Hex( ) function
wrapping, 392-394 GetNthURL method (VBScript), 715
CONVERTITTEST1 Content Linking history (VBScript), 51-52
listing, 393-394 Component, 779 hits, Page Counter
listing, 392-393 Content Linking object, Component, 419
365
Hits method
GetPreviousDescription Page Counter
G method, 368, 778 Component, 377
Get method GetProductInventory Home Directory
Counter Component, 379 method, 688 (Microsoft Management
GetAdvertisement GetPreviousURL method Console (MMC)), 658
method Content Linking Hour( ) function
Ad Rotator Component, Component, 778 (VBScript), 716
336 Content Linking object,
365

45.1389-6 index 794 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 Item property 795

HTML If...Then decision Internet Information

ASP applications, 182 structure (VBScript), 90 Server, 30, 39
dynamic If...Then…Else decision HTTP 1.1 specifications,
Internet Explorer, 524 structure (VBScript), 91 42
object model, 52 Microsoft Management
IIS, VBScript syntax, 58
tags Console (MMC), 39
<% %>, 57 Image Composer (Visual Secure Socket Layer, 41
<SCRIPT>, 57-58, 112 InterDev), 581 server-side Java, 43
HTML Layout Control Increment method Internet Service Manager
Editor (Visual Counter Component, 380 (interface utility), 204
InterDev), 578 indexes (databases), 517 intranets
HTML tools (FrontPage), inequality operator (<>), applications, 640
35 78 bandwidth, 640
HTMLEncode method input boxes (VBScript), browsers, 641
Application object, 740 125-126 project management, 643
Server object, 237 InputBox function inventory (ADO),
HTTP, 14 (VBScript), 125-126, 540-542
KeepAlive request, 15 716 ISAPI, 19
TCP/IP, 14 installing filters, 20, 661
HTTP 1.1 (Microsoft components (Transaction IsArray function, 128
Management Console Server), 600 IsArray( ) function
(MMC)), 670 registering, 600 (VBScript), 717
HTTP KeepAlive request, instantiating IsClientConnected
15 Application objects, 220 property
application-level object Response object, 284, 747
bandwidth
IsDate function, 128
I (intranets), 640
VBScript, 717
Session object, 261
icons IsEmpty function, 129
constants (VBScript InStr( ) function
VBScript, 717
message boxes), 123 (VBScript), 716
IsNull function, 129
VBScript dialog boxes, InStrRev( ) function
VBScript, 717
123 (VBScript), 716
IsNumeric function, 131
identifier functions, 127 Int( ) function
VBScript, 717
IsArray, 128 (VBScript), 716
IsDate, 128 IsObject function, 131
integer division operator
IsEmpty, 129 VBScript, 717
(^), 73
IsNull, 129 IsolationLevel property
InterDev (server-side
IsNumeric, 131 Connection object, 458,
scripts), 147
IsObject, 131 486, 762
VarType, 132 interface utilities
issues (Active Server
(servers), 204
IE (Internet Explorer), Pages), 208
59 Internet Explorer, 59
Item property
dynamic HTML, 524 dynamic HTML, 524
Parameter object, 476

45.1389-6 index 795 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 796 Java

J limitations wrapping functions,

Remote Data Services, CONVERTITTEST1,
Java, Internet 527 393-394
Information Server, 43 Line property LoadPicture( ) function
JavaScript, 36 File Access component, (VBScript), 718
Join( ) function 780 LoadProductsTable
(VBScript), 718 Link View tools (Visual ADO object, 536
journal queues (message InterDev), 580 locating queues (Message
queuing), 618 linking data (Connection Queue Server), 622
JScript, 36 object), 454 Lock method
Just in Time (JIT), 33 links (Content Linking Application object, 218,
Component), 360 737
listings locking
K ActiveX component databases, 512
datatype diffferences, exclusive, 512
KeepAlive request 389-390 recordsets, 487
(HTTP), 15 ActiveX components, shared, 512
keys (databases), 517 SIMPLECLASS, 385 LockType property
KeySet cursor, 489 ActiveX components, Recordset object, 769
TESTMULT.ASP, 387
keywords log files, debugging, 195
ActiveX components,
Private (VBScript), 65-66 TESTMULT.BAS, 386 Log( ) function
Public (VBScript), 65 ASP applications, (VBScript), 719
accelerating, 405-408 logical errors (Script
ASP intrinsic objects, Debugger), 174
L accessing via ActiveX logical operators, 79
components, 403 conjunction operator
encapsulating (AND), 81
LANGUAGE attribute, 25 calculations (AX disjunction operator
LANGUAGE directive, components), 395-397 (OR), 82
700 porting Visual Basic equivalence operator
LBound( ) function applications to (EQV), 83
(VBScript), 718 VBScript, 399-400 exclusion operator
LCase( ) function type casting, 390-391 (XOR), 82
(VBScript), 718 VBScript to manipulate negation operator (NOT),
server-side ActiveX 80
LCID property controls, 55
Session object, 254, 756 looping
VBScript to manipulate
Left( ) function VBScript, 96
HTML controls, 53-54
(VBScript), 718 Do...Loop, 99
VBScript variable scope,
For Each...Next, 101
Len( ) function 63
wrapping functions, For...Next, 96
(VBScript), 718 While...Wend, 100
392-393
less than operator (<), 78 LTrim( ) function
(VBScript), 719

45.1389-6 index 796 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 methods 797

M Message Queue Server, Content Linking object,

619 362
MapPath method prioritizing, 617 GetListCount, 363
Application object, 739 queue browsing, 617 GetListIndex, 364
Server object, 232 storing, 616 GetNextURL, 364
MaxRecords property transactional messages, GetNthDescription,
Recordset object, 770 618 368
messages GetNthURL, 365
Media Manager (Visual GetPreviousURL, 365
InterDev), 582 receiving (Message
Queue Server), 628 Content Rotator
memory leaks (Active sending (Message Queue Component, 373
Server Pages), 208 Server), 625 ChooseContent, 373
message boxes GetAllContent, 374
Metabase, configuring
(VBScript), 120-124 Counter Component
pages, 199
icon constants, 123 Get, 379
result constants, 124 methods Increment, 380
ACTION, 422 Remove, 381
Message Queue Explorer Ad Rotator Component Set, 380
(Message Queue GetAdvertisement, Err object, 108
Server), 629 336, 776 Fields collection
Message Queue Server, Application object, 218, AppendChunk, 773
619 737 GetChunk, 774
ActiveX components, 619 CreateObject, 739 File Access component
ASP, 629 HTMLEncode, 740 Close, 780
Message Queue Lock, 218, 737 CreateTextFile, 780
Explorer, 629 MapPath, 739 OpenTextFile, 780
MSMQQueueInfo ScriptTimeOut, 739 Read, 781
component, 620 UnLock, 219, 737 ReadAll, 781
queues URLEncode, 740 ReadLine, 781
closing, 628, 636 Command object Skip, 781
deleting, 629, 636 CreateParameter, 766 SkipLine, 781
locating, 622 Execute, 475, 766 TextStream, 780
opening, 624 Connection object, 453 Write, 781
receiving messages, BeginTrans, 764 WriteBlankLines, 781
628 Close, 764 WriteLine, 781
sending messages to, CommitTrans, 764 ObjectContext object
631 Execute, 460, 764 SetAbort, 741
sending messages, 625 Open, 764 SetComplete, 741
message queuing, RollTransBack, 764 Page Counter
613-614 Content Linking Component, 376
aging, 617 Component Hits, 377
dead-letter queues, 618 GetListCount, 778 Reset, 377
forwarding, 616 GetListIndex, 778 Parameter object, 477
guaranteed message GetNthURL, 779 Parameters collection,
delivery, 615 GetPreviousURL, 778 476
journal queues, 618

45.1389-6 index 797 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 798 methods

Parameters object SetComplete, 240-241 MonthName( ) function

Append, 768 Tools Component (VBScript), 719
AppendChunk, 768 FileExists, 354 Move method
Delete, 768 Owner, 356 Recordset object, 771
Permission Checker PluginExists, 356
MoveFirst method
Component Random, 357
Recordset object, 495,
HasAccess, 358 Transaction Server
771
POST, 422 components, 604
Recordset object returning data, 604 MoveLast method
AddNew, 499, 770 VBScript, 732 Recordset object, 495,
CancelBatch, 770 771
Microsoft Management
CancelUpdate, 770 Console (MMC), 39, MoveNext method
Clone, 771 654 Recordset object, 496,
Close, 771 application settings , 661 772
Delete, 500, 771 ASP components, 670 MovePrevious method
GetRows, 771 FTP sites, 666 Recordset object, 497
Move, 771 Home Directory, 658
MoveFirst, 495, 771 MsgBox constants
HTTP 1.1, 670 VBScript, 720, 728
MoveLast, 495, 771 ISAPI filters , 661
MoveNext, 496, 772 MsgBox( ) function
Script Debugger, 668
MovePrevious, 497 (VBScript), 120-124,
security, 664
NextRecordset, 772 720
transaction-based scripts,
Open, 772 668 MSMQQueueInfo
Requery, 772 virtual directories, 666 component, 620
Resync, 772 Web building sites, 656, MTS (Microsoft
Supports, 500, 772 666-667 Transaction Server)
Update, 498, 772 component properties,
Mid( ) function
UpdateBatch, 772 239-240
(VBScript), 719
Request object, 313
BinaryRead, 313, 754 middleware (n-tier), 586 multi-button forms
Response object, 284, 747 Minute( ) function Request object, 328
AddHeader, 289, 748 (VBScript), 719 multiplication operator
AppendToLog, 289, (*), 72
miscellaneous constants
748 (VBScript) 728 Music Producer (Visual
BinaryWrite, 290, 749 InterDev), 582
Clear, 290, 749 missing files (Tools
End, 291, 749 Component) 415
Flush, 292, 749 MMC (Microsoft N
Redirect, 293, 749 Management Console),
Write, 285, 748 39 n-tier
Server object, 230 Mode property client/server model, 44
CreateObject, 230 Connection object, 456, Web access, 45
encoding, 234 763 Name property
MapPath, 232 modulus operator Fields collection, 773
Session object (MOD), 74 Parameter object, 478
Abandon, 259, 758 Parameters object, 767
SetAbort, 241 Month( ) function
(VBScript), 719

45.1389-6 index 798 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 objects 799

naming conventions object Session, Connection, 452, 762

ADO, 444 component threading Command object, 465
VBScript, 66 model, 262 methods, 453
NativeError property object-oriented phone application, 465
Errors collection, 765 programming (OOP), properties, 452
30 transaction
navigating recordsets,
processing, 461
495 ObjectContext object,
Err, debugging, 194
navigation buttons, 423 192, 239, 741
Error properties, 463
events, 242-244
negation operator (NOT), File Access components
OnTransactionAbort,
80 FileSystemObject, 779
743
nesting procedures FileSystem debugging,
OnTransactionCommit,
(VBScript), 117 195
242-243, 742
ObjectContext, 192, 239,
networks methods, 240-241
741
extranets, 17, 649 SetAbort, 241, 741
events, 242-244
intranets, 649 SetComplete, 241, 741
methods, 240-241
Microsoft Management objects, 27 Parameter properties,
Console (MMC), 654 ADO, 432-433, 536 477
NextLink.DLL (Content Command, 437 Parameters, 476
Linking Component), Connection, 434, 536 properties, 113
360 Field, 441 Recordset, 768
NextRecordset method LoadProductsTable, properties, 489
Recordset object, 772 536 Remote Data Services
Parameter, 438 AdvancedDataControl,
non-visual ActiveX Recordset, 440
control, 523 525
Application, 186, 192, 212, AdvancedDataFactory,
Now( ) function 736 526
(VBScript), 720 Application_OnEnd AdvancedDataSpace,
Number property event, 217 526
Errors collection, 765 arrays, 222 Request, 193, 750
NumericScale property collections, 226 collections, 299, 750
Fields collection, 773 components, 220 extracting user
Parameter object, 478 declaring variables, information, 329
214 methods, 313
instantiating at the multi-button forms,
O application-level, 220 328
methods, 218 properties, 313
Object Browser, 386 Onstart event, 217 Response, 193
object comparison ASP intrinsic, 401-408 collections, 268
operator (IS), 79 business, 523 Domain attribute, 271
Command, 470, 765 Eesppn, 743
object models, 20, 52 Connection object, 465
dynamic HTML object Expire attribute, 270
creating, 470 HasKeys attribute, 272
model, 22 Parameters object, 476 methods, 284
object request brokers, properties, 471 Path attribute, 272
591 recordsets, 475

45.1389-6 index 799 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 800 objects

properties, 273 OLE DB (ADO), 430 exponentiation

Secure attribute, 272 On Error statement operator (^), 72
ScriptErrorMessage VBScript, 103 integer division
debugging, 195 operator (^), 73
online catalogs, 316
ScriptErrorsSentToBrowser modulus operator
application-level arrays,
debugging, 195 (MOD), 74
318
Server, 192, 229 multiplication operator
Database Object
encoding, 324 (-), 72
Component, 693
methods, 230 subtraction operator
searching “shopping
properties, 237 (-), 72
basket”, 326
Session, 193, 247, 755 comparison, 75
session-level
arrays, 262 equality operator, 77
arrays, 320
CodePage property, greater than operator
variables, 320
256 (>), 78
collections, 264 Onstart event inequality operator
configuring, 261 (Application object), (<>), 78
cookies, 249 217 less than operator (<),
events, 256 OnTransactionAbort 78
LCID property, 254 event, 242-244 object comparison
properties, 250, 755 ObjectContext object, 743 operator (IS), 79
resources, 259 OnTransactionCommit concatenation
session-level objects, event, 242-243 addition operator (+),
261 ObjectContext object, 742 85
SessionID property, ampersand operator
252 Open Database (&), 84
Connectivity 3.0 (ODBC
Timeout property, 252 logical, 79
3.0), 513
user information, 248 Conjunction operator
variables, 250, 755 Open method (AND), 81
Session object Connection object, 764 disjunction operator
session-level, 759 Recordset object, 772 (OR), 82
VBScript, 732 Open property equivalence operator
Oct( ) function Connection object, 459 (EQV), 83
(VBScript), 720 opening queues exclusion operator
(Message Queue (XOR), 82
OCX controls, 23
Server), 624 negation operator
ODBC 3.0 (Open (NOT), 80
Database Connectivity OpenTextFile method precedence, 85
3.0), 513 File Access component, VBScript, 703
780
ODBMS (object-oriented Option Explicit statement
databases), 506 operators (VBScript), 66
arithmetic, 70
OLE automation (Object ORBs (object request
addition operator (+),
Linking and brokers), 591
Embedding), 23 70
concatentation OriginalValue property
OLE Automation Servers, operators, 84 Fields collection, 773
384 division operator (/), Owner method
see also ActiveX, 72 Tools Component, 356
components

45.1389-6 index 800 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 properties 801

P Parameters object, 476 Private Communication

methods Technology (PCT), 17
packages (Transaction Append, 768 Private keyword
Server), 598 AppendChunk, 768 (VBScript), 65-66
packet filtering (security), Delete, 768
private procedures
197 properties
(VBScript), 114-115
Attributes, 767
Page Counter Direction, 767 procedure-level scope
Component, 375 Name, 767 (VBScript), 63
hits, 419 Precision, 767 procedure-level scoping,
methods, 376 Size, 767 191
Hits, 377 Type, 767
Reset, 377 procedures (VBScript)
Value, 767 argument lists, 115-117
PageCount property Path attribute (cookies), Exit statement, 118
Recordset object, 770 272 functions, 114, 119
pages (recordsets), 492 PERL CGI scripts, 18 nesting, 117
PageSize property public/private, 114-115
Permission Checker
Recordset object, 492, subroutines, 114, 118
Component, 357
770 calling, 118-119
HasAccess method, 358
Parameter object public/private, 114-115
phone application
ADO, 438 process directives,
Parameter object, 481
creating, 480 server-side scripts, 150
methods, 477 phone book application
project management
phone application, 481 Remote Data Services,
(intranets), 643
properties, 477 527
projects (Visual
Attributes, 478 PICS property
InterDev), 563
CreateParameter, 476 Response object, 281, 746
Delete, 477 properties, 113
PluginExists method
Direction, 478 Ad Rotator Component,
Tools Component, 356
Item, 476 340
porting applications
Name, 478 Border, 776
(Visual Basic to
Numeric Scale, 478 Clickable, 776
VBScript), 398-401
Precision, 479 TargetFrame, 776
POST method, 422 Browser Capabilities
Refresh, 477
Size, 479 precedence of operators, Component, 345
Type, 479 85 Command object
Precision property ActiveConnection, 471,
parameter passing (ASP), 765
611 Fields collection, 773
Parameter object, 479 CommandText, 765
Parameters collection, Parameters object, 767 CommandTimeout,
476 474, 766
Command object, 476 Prepared property
CommandType, 473,
methods, 476 Command object, 474,
766
properties, 476 766
Prepared, 474, 766
prioritizing (message Command objects, 471
queuing), 617

45.1389-6 index 801 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 802 properties

Connection object, 452 Parameter object, 477 CharSet, 277, 745

Attributes, 462, 762 Attributes, 478 ContentType, 280, 746
Close, 459 CreateParameter, 476 Expires, 280, 746
CommandTimeout, Delete, 477 ExpiresAbsolute, 281,
762 Direction, 478 746
ConnectionString, 456, Item, 476 IsClientConnected,
762 Name, 478 284, 747
ConnectionTimeout, Numeric Scale, 478 PICS, 281, 746
457, 762 Precision, 479 Status, 283, 746
DefaultDatabase, 458, Refresh, 477 Server object, 237
762 Size, 479 ScriptTimeout, 237
IsolationLevel, 458, Type, 479 Session object, 250, 755
762 Parameters collection, CodePage, 256, 757
Mode, 456, 763 476 LCID, 254, 756
Open, 459 Parameters object SessionID, 252, 756
Provider, 763 Attributes, 767 TimeOut, 756
Version, 763 Direction, 767 Timeout, 252
Err object, 107 Name, 767 VBScript, 733
Error object, 463 Precision, 767 protocols
Errors collection Size, 767 HTTP, 14
Description, 765 Type, 767 KeepAlive request, 15
HelpContext, 765 Recordset object, 489 Private Communication
HelpFile, 765 AbsolutePage, 768 Technology (PCT), 17
NativeError, 765 AbsolutePosition, 492, Secure Socket Layers
Number, 765 768 (SSL), 17
Source, 765 ActiveConnection, 768 TCP/IP, 14
SQLState, 765 BOF, 768
Provider property
Fields collection BOF/EOF, 490
Connection object, 763
ActualSize, 773 Bookmark, 768
Attributes, 773 CacheSize, 769 proxy servers, 16
DefinedSize, 773 CursorType, 769 Public keyword
Name, 773 EditMode, 769 (VBScript), 65
NumericScale, 773 EOF, 769 public procedures
OriginalValue, 773 Filter, 493, 769 (VBScript), 114-115
Precision, 773 LockType, 769
Type, 773 MaxRecords, 770
UnderlyingValue, 773 PageCount, 770 Q
Value, 773 PageSize, 492, 770
File Access component RecordCount, 491, 770 Query Designer (Visual
AtEndOfLine, 780 Source, 770 InterDev), 570
AtEndOfStream, 780 Status, 770 QueryString collection
Column, 780 Request object, 313 Request object, 305, 752
Drives, 779 TotalBytes, 313, 754
queues
Line, 780 Response object, 273, 745
browsing (message
MTS component Buffer, 274, 745
queuing), 617
properties, 239, 240 CacheControl, 276,
745

45.1389-6 index 802 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 Remove method 803

Message Queue Server, MoveLast, 495, 771 redirection file (Ad

620 MoveNext, 496, 772 Rotator Component),
closing, 628, 636 MovePrevious, 497 339
deleting, 629, 636 NextRecordset, 772 Refresh property
locating, 622 Open, 772 Parameter object, 477
opening, 624 Requery, 772
receiving messages, Resync, 772 registering
634 Supports, 500, 772 components, 32
sending messages, Update, 498, 772 Database Object
625, 631 UpdateBatch, 772 Component, 678
properties, 489 installing, 600
AbsolutePage, 768 Transaction Server, 597
R AbsolutePosition, 492, registration
768 Content Linker
Random method ActiveConnection, 768 Component, 420
Tools Component, 357 BOF, 768 DCOMCNFG.EXE, 32
Read method BOF/EOF, 490 REGSVR32.EXE, 32
File Access component, Bookmark, 768 Registry, configuring
781 CacheSize, 769 pages, 201
ReadAll method CursorType, 769 Registry Settings
File Access component, EditMode, 769 (debugging), 195
781 EOF, 769
Filter, 493, 769 Regsrv (Database Object
ReadLine method Component), 682
LockType, 769
File Access component, relational databases, 507
MaxRecords, 770
781
PageCount, 770 Remote Data Services,
receiving messages PageSize, 492, 770 447, 522
(Message Queue RecordCount, 491, 770 ASP, 524
Server), 628 Source, 770 browsers, 527
from queues (Message Status, 770 limitations, 527
Queue Server), 634 objects
recordsets
RecordCount property Command object, 475 AdvancedDataControl,
Recordset object, 491, forms, 500 525
770 locking, 487 AdvancedDataFactory,
recordset (ADO), 443 navigating, 495 526
Recordset object, 768 pages, 492 AdvancedDataSpace,
ADO, 440 Recreational Software 526
methods Advisory Council on the phone book application,
AddNew, 499, 770 Internet (RSACi), 281 527
CancelBatch, 770 servers (third party
ReDim function components), 527
CancelUpdate, 770 (VBScript), 64-65
Clone, 771 Virtual Table Manager
Redirect method (VTM), 524
Close, 771
Response object, 293, 749 Remove method
Delete, 500, 771
redirection file (Ad Counter Component, 381
GetRows, 771
Rotator Component),
Move, 771
776
MoveFirst, 495, 771

45.1389-6 index 803 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 804 Replace( ) function

Replace( ) function AppendToLog, 289, routing users, 26

(VBScript), 720 748 RSACi (Recreational
Requery method BinaryWrite, 290, 749 Software Advisory
Recordset object, 772 Clear, 290, 749 Council on the
End, 291, 749 Internet), 281
request brokers (n-tier),
Flush, 292, 749
586 RTrim( ) function
Redirect, 293, 749
Request object, 193, (VBScript), 721
Write, 285, 748
298, 750 Path attribute, 272 RUNAT attribute, 26
collections, 299, 750 properties, 273, 745 runtime errors (Script
ClientCertificate, 300, Buffer, 274, 745 Debugger), 173
750 CacheControl, 276,
Cookies, 301, 751 745
Form, 303, 751 CharSet, 277, 745 S
QueryString, 305, 752 ContentType, 280, 746
ServerVariables, 309, Expires, 280, 746
752 ExpiresAbsolute, 281, scalability (component
extracting user 746 architecture), 31
information, 329 IsClientConnected, Schedule file (Ad Rotator
methods, 313 284, 747 Component), 338, 777
BinaryRead, 313, 754 PICS, 281, 746
multi-button forms, 328 scope (VBScript
Status, 283, 746
properties, 313 variables), 62
Secure attribute, 272
TotalBytes, 313, 754 listing, 63-64
result constants
Requires a new scoping
(VBScript message
transaction property, application-level, 190
boxes), 124
239 GLOBAL.ASA, 190
Resync method procedure-level, 191
Requires a transaction Recordset object, 772 session-level, 191
property, 239
retrieving user Script Debugger, 37,
Reset method information 162
Page Counter server-side scripts, 152 breakpoints, 167
Component, 377
returning data code stepping, 167
resources (Transaction Server), logical errors, 174
Session object, 259 604 runtime errors, 173
Abandon method, 259 script flow, 167
Right( ) function
Response object, 193, (VBScript), 720 syntax errors, 173
268 windows, 165
Rnd( ) function
collections, 268 Script Debugger
(VBScript), 721
Cookie, 268, 744 (Microsoft), 56, 668
Domain attribute, 271 rollback processing
(ADO), 446 Script Wizard, 37, 56
Eesppn, 743
Visual InterDev, 576
Expire attribute, 270 RollTransBack method
HasKeys attribute, 272 Connection object, 764 script-level scope
methods, 284, 747 (VBScript), 63
Round( ) function
AddHeader, 289, 748 (VBScript), 721

45.1389-6 index 804 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 Server object 805

<SCRIPT> tag (HTML), variables, 60-66 Secure attribute

57-58, 112 VarType function, 62 (cookies), 272
ScriptEngine( ) function scripts Secure Socket Layers
(VBScript), 721 Active Server Pages, 24 (SSL), 17, 41, 60
ScriptEngineBuild ASP applications, 182 security
Version( ) function CGIs, 18 access, 555
(VBScript), 721 client-side, 155 Active Server Pages
testing for browser viewing code, 208
ScriptEngineMajor
compliance, 158 ASP applications
Version( ) function
validation subroutine, access control, 197
(VBScript), 721
156-157 file level, 198
ScriptEngineMinor flow (Script Debugger), packet filtering, 197
Version( ) function 167 databases, 519
(VBScript), 722 ISAPI, 19 access, 555
ScriptErrorMessage Microsoft Management extranets, 649
object, debugging, 195 Console (MMC), 668 encryption, 650
ScriptErrorsSentTo PERL, 18 firewalls, 17
Browser object, server-side Permission Checker
debugging, 195 calculations, 153 Component, 357
declarations, 150 Private Communication
scripting, 36, 700
delivering information, Technology (PCT), 17
Active Server Pages, 24
152 Secure Socket Layers
components, 27
dynamic template, 149 (SSL), 17
JavaScript, 36 error trapping, 155
JScript, 36 VBScript, 58
FrontPage, 149 IE, 59
LANGUAGE directive,
InterDev, 147 server-side, 59-60
700
process directives, 150
languages, 36 security (Microsoft
static pages, 147
objects, 27 Management Console),
user information, 152
Script Debugger, 37 664
setting, 25
Script Wizard, 37 VBScript (server-side), Select Case statement
tools, 37 (VBScript), 94
146
VBScript, 36, 703
viewing (security), 208 sending messages
Visual JavaScript, 38
ScriptTimeOut method (Message Queue
scripting languages Server), 625, 631
Application object, 739
VBScript, 49-51, 111-113
ScriptTimeout property Server object, 192, 229
ActiveX controls, 52-54
(Server object), 237 encoding, 324
events, 113
encoding methods, 234
functions, 114-119 search program (web
HTMLEncode, 237
history, 51-52 application), 448
URLEncode, 235
IIS syntax, 58 searching methods, 230
ReDim function, 64-65 database servers, 16 CreateObject, 230
security, 58-60 “shopping basket”, 326 MapPath, 232
subroutines, 114-119
Second( ) function properties, 237
user interaction,
(VBScript), 722 ScriptTimeout, 237
120-126

45.1389-6 index 805 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 806 Server Side Includes

Server Side Includes Message Queue Server, Session_OnStart, 256,

(SSI), 27 619 757
server variables, 309 ActiveX components, methods
619 Abandon, 758
server-side ActiveX
proxy, 16 properties, 250, 755
controls
Remote Data Services, CodePage, 256, 757
VBScript, 54
527 LCID, 254, 756
listing, 55
status line, 283 SessionID, 252, 756
server-side scripts Transaction Server, 32, TimeOut, 756
calculations, 153 510, 592 Timeout, 252
declarations, 150 ASP, 597 resources, 259
delivering information, BeginTrans command, Abandon method, 259
152 33 session-level objects, 261,
dynamic template, 149 components, 603 759
error trapping, 155 connections, 593 user information, 248
FrontPage, 149 Database Object variables, 250, 755
InterDev, 147 Component, 678 declaring, 251, 756
process directives, 150 distributed objects, Session_OnEnd event
static pages, 147 595 Session object, 258
user information, 152 installing components,
VBScript, 146 Session_OnStart event
600
declarations, 150 Session object, 256
Just in Time (JIT), 33
delivering information, packages, 598 session-level
152 registering objects, 261
dynamic template, 149 components, 597 Session_OnStart, 759
FrontPage, 149 transaction control, scoping, 191
InterDev, 147 596 variables, 320
process directives, 150 validation servers, 686 SessionID property
static pages, 147 Session object, 252, 756
ServerVariables
user information, 152
collection sessions, tracking events,
server-side security Request object, 309, 752 187
issues, 59-60
Session events Set method
server-side VBScript Session object, 757 Counter Component, 380
IIS syntax, 58
Session object, 193, SetAbort method, 241
versus client-side, 57
247, 755 ObjectContext object, 741
servers, 14 arrays, 262 SetComplete method,
certificate, 17 collections, 264 240-241
database, 15 Contents, 265, 759 ObjectContext object, 741
searches, 16 StaticObject, 265, 759
escape codes, 234 setting scripts, 25
component threading
external, 18 model, 262 Sgn( ) function
firewalls, 17 configuring, 261 (VBScript), 722
interface utilities, 204 cookies, 249 shared locks, 512
Internet Information events, 256, 757 shopping basket (online
Server, 30, 39 Session_OnEnd, 258, catalog), 326
757

45.1389-6 index 806 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 TimeOut property 807

Sin( ) function status line, 283 T

(VBScript), 722 Status property
Size property Recordset object, 770 tablegrams (Virtual Table
Parameter object, 479, Response object, 283, 746 Manager), 524
767 storing (message tags
Skip method queuing), 616 <SCRIPT> tag, 26
File Access component, StrComp( ) function delimiters, 25
781 (VBScript), 723 tags (HTML)
SkipLine method string constants <% %>, 57
File Access component, (VBScript), 730 <SCRIPT>, 57-58, 112
781 Tan( ) function
string subtypes,
SMPT (Active Messaging converting to, 140 (VBScript), 723
Component), 348 TargetFrame property
String( ) function
Source property (VBScript), 723 Ad Rotator Component,
Errors collection, 765 776
strings (CommandText),
Recordset object, 770 targetting user groups,
472
Space( ) function 13
StrReverse( ) function
(VBScript), 722 TCP/IP, 14
(VBScript), 723
Split( ) function HTTP, 14
stylesheets, 22
(VBScript), 722 terminal access
subroutines architecture, 44
SQL databases, 507
VBScript, 114, 118
SQLState property argument lists, 115-117 testing
Errors collection, 765 calling, 118-119 client-side scripts for
Sqr( ) function Exit statement, 118 browser compliance,
(VBScript), 723 nesting, 117 158
public/private, 114-115 components (Database
SSI (Server Side Object Component),
Includes), 27 subtraction operator (-), 691
SSL (Secure Sockets 72
text (CharSet property),
Layer), 60 subtypes 277
stateless (Web Variant datatype, 702
VBScript Variant text files (ASP
applications), 431 applications), 180
datatype, 60-62
statements (conditional), TextStream method
704 Support transactions
property, 240 File Access component,
static content, 10 780
supporting transactions
static cursor, 488 threading, Transaction
(ASP), 240
static pages, server-side Server components,
Supports method 610
scripts, 147
Recordset object, 500,
StaticObject collection 772 Time( ) function
Application object, 227 (VBScript), 723
syntax errors (Script
Session object, 265 TimeOut property
Debugger), 173
Session OnStart, 759 Session object, 756

45.1389-6 index 807 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 808 Timeout property

Timeout property transaction monitors, two-phase commit

Session object, 252 589 (databases), 596
TimeSerial( ) function n-tier, 586 type casting (AX
(VBScript), 724 transaction processing, components)
TimeValue( ) function 446 listing, 390-391
(VBScript), 724 Connection object, 461 Type property
databases, 510 Fields collection, 773
tools
distributed, 587 Parameter object, 479
FrontPage, 35
scripting, 37 Transaction Server, Parameters object, 767
Visual InterDev, 34, 569 510-511, 592 TypeName( ) function
Data Command, 574 ASP, 597 (VBScript), 724
Data Form Wizard, BeginTrans command, 33
573 components, 32, 603
Data Range Footer, accounts, 605 U
574 ASP, 610
Data Range Header, calling accounts, 606 UBound( ) function
574 methods, 604 (VBScript), 724
Data View, 569 returning data, 604 UCase( ) function
Database Designer, threading, 610 (VBScript), 724
572 connections, 593
UnderlyingValue property
design-time controls, Database Object
Fields collection, 773
574 Component, 678
distributed objects, 595 UnLock method
Image Composer, 581
Distributed Transaction Application object, 219,
Link View, 580
Coordinator (DTC), 512 737
Media Manager, 582
Music Producer, 582 installing components, Update method
Query Designer, 570 600 Recordset object, 498,
Just in Time (JIT), 33 772
Tools Component, 354
packages, 598 UpdateBatch method
methods
registering components, Recordset object, 772
FileExists, 354
597
Owner, 356 URLEncode method
transaction control, 596
PluginExists, 356 Application object, 740
Random, 357 transaction-based scripts Server object, 235
missing files, 415 (MMC), 668
users
TotalBytes property transactional messages extracting information,
Request object, 313, 754 (message queuing), 618 329
tracking transactions, 32 hits (Page Counter
session events, 187 supporting, 240 Component), 419
user events, 186 trapping application information, Session
events, 212 object, 248
transaction control
interaction (VBScript),
(Transaction Server), Trim( ) function
120
596 (VBScript), 724
icon constants, 123
Transaction directive, tristate constants input boxes, 125-126
240 (VBScript), 730 message boxes, 120-124

45.1389-6 index 808 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 VBScript 809

Request object, 298 VBA (Visual Basic for Asc( ), 710

retrieving information Applications), 55 Atn( ), 710
(server-side scripts), VBScript, compared, CBool( ), 710
152 55-56 CByte( ), 711
routing, 26 see also VBScript; Visual CCur( ), 711
searching (online Basic CDate( ), 711
catalog), 326 VBScript, 36, 49-51, CDble( ), 711
tracking events, 186 111-113 Chr( ), 711
ActiveX controls, 52 CInt( ), 711
listing, 53-54 CLng( ), 711
V client-side scripts, 155 Cos( ), 712
testing for browser CreateObject( ), 712
validation servers CSng( ), 712
(Database Object compliance, 158
validation subroutine, CStr( ), 712
Component), 686 Date( ), 712
156-157
validation subroutines client-side versus DateAdd( ), 712
(client-side scripts), server-side, 57 DateDiff( ), 713
156-157 conditional statements, DatePart( ), 713
Value property 704 DateSerial( ), 713
Fields collection, 773 constants, 725 DateValue( ), 714
Parameters object, 767 color, 725 Day( ), 714
comparison, 726 Exit statement, 118
variables
date format, 727 Exp( ), 714
Application object, 214
date/time, 726 Filter( ), 714
application-level
file input/output, 728 Fix( ), 714
(GLOBAL.ASA), 318
miscellaneous, 728 FormatCurrency( ),
Counter Component, 378
MsgBox, 728 714
server, 309
string, 730 FormatDateTime( ),
Session object, 250, 755
tristate, 730 715
declaring, 251, 756
VarType, 731 FormatNumber( ), 715
session-level
conversion functions, 136 FormatPercent( ), 715
(GLOBAL.ASA), 320
decision structures, 90 Hex( ), 715
VBScript, 60-62
If...Then, 90 Hour( ), 716
declaring, 64-66
If...Then...Else, 91 InputBox(), 125-126,
naming conventions,
Select Case statement, 716
66
94 InStr( ), 716
scope, 62-64
Err object, 104 InStrRev( ), 716
Variant datatype, 60-61
methods, 108 Int( ), 716
Variant datatype IsArray( ), 717
subtypes, 702 properties, 107
events, 113 IsDate( ), 717
VBScript, 60-61 IsEmpty( ), 717
Exit statement, 102
VarType constants formatting functions, 133 IsNull( ), 717
VBScript, 731 functions, 114, 119, 705 IsNumeric( ), 717
VarType function, 132 Abs( ), 710 IsObject( ), 717
VBScript, 62, 725 argument lists, 115-117 Join( ), 718
Array( ), 710 LBound( ), 718

45.1389-6 index 809 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 810 VBScript

LCase( ), 718 VarType( ), 62, 725 Vbscript operators, 69

Left( ), 718 Weekday( ), 725 VBScripts (server-side),
Len( ), 718 WeekdayName( ), 725 146
LoadPicture( ), 718 Year( ), 725
Version property
Log( ), 719 history, 51-52
Connection object, 763
LTrim( ), 719 IIS syntax, 58
Mid( ), 719 indentifier functions, 127 versions (Active Server
Minute( ), 719 looping, 96 Pages), 208
Month( ), 719 Do...Loop, 99 virtual directories
MonthName( ), 719 For Each...Next, 101 (Microsoft Management
MsgBox(), 120-124, For...Next, 96 Console), 666
720 While...Wend, 100 Virtual Table Manager
nesting, 117 methods, 732 (VTM), 524
Now( ), 720 objects, 732
Visual Basic
Oct( ), 720 On Error statement, 103
porting to VBScript,
public/private, 114-115 operators, 703
398-401
ReDim, 64-65 porting from Visual
Replace( ), 720 Basic, 398-401 listing, 399-400
Right( ), 720 listing, 399-400 VBScript, compared,
55-56
Rnd( ), 721 properties, 733
wrapping functions,
Round( ), 721 scripts (client-side), 155
392-394
RTrim( ), 721 security, 58
listing, 392-393
ScriptEngine( ), 721 IE, 59
ScriptEngineBuild server-side, 59-60 see also VBA; VBScript
Version(), 721 server-side scripting, 54 Visual Basic for
ScriptEngineMajor listing, 55 Applications, see VBA
Version( ), 721 statements, 704 visual controls, 447
ScriptEngineMinor subroutines, 114, 118 Visual InterDev, 34, 562
Version( ), 722 argument lists, 115-117 ADO, 543
Second( ), 722 calling, 118-119 databases, 543-553
Sgn( ), 722 Exit statement, 118 Editor, 577
Sin( ), 722 nesting, 117 environment, 565
Space( ), 722 public/private, 114-115 file management, 579
Split( ), 722 user interaction, 120 Forms Wizard, 551
Sqr( ), 723 input boxes, 125-126 HTML Layout Control
StrComp( ), 723 message boxes, Editor, 578
String( ), 723 120-124 projects, 563
StrReverse( ), 723 variables, 60-62 tools, 569
Tan( ), 723 declaring, 64-66 Data Command, 574
Time( ), 723 naming conventions, Data Form Wizard,
TimeSerial( ), 724 66 573
TimeValue( ), 724 scope, 62-64 Data Range Footer,
Trim( ), 724 Variant datatype, 60-61 574
TypeName( ), 724 Variant datatype, 702 Data Range Header,
UBound( ), 724 subtypes, 702 574
UCase( ), 724 Visual Basic, compared, Data View, 569
55-56

45.1389-6 index 810 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 Year( ) function 811

Database Designer, servers, 14-16 workspaces (Visual

572 certificate, 17 InterDev), 563
design-time controls, database, 15 wrappers, 392
574 external, 18 wrapping functions,
Image Composer, 581 firewalls, 17 392-394
Link View, 580 proxy, 16 CONVERTITTEST1
Media Manager, 582 sites (Script Debugger listing, 393-394
Music Producer, 582 download), 560 listing, 392-393
Query Designer, 570 static content, 10
wizards, 576 Write method
Web-based applications,
workspaces, 563 File Access component,
12
781
Visual JavaScript, 38 Weekday( ) function Response object, 285, 748
V TM (Virtual Table (VBScript), 725
WriteBlankLines method
Manager), 524 WeekdayName( ) File Access component,
function (VBScript), 781
725
W While...Wend loop
WriteLine method
File Access component,
Web (VBScript), 100 781
applications, 12 windows (Script
ADO, 431 Debugger), 165
client/server wizards X-Y-Z
architecture, 13, 17 Visual InterDev, 576
search program, 448 Year( ) function
Editor, 577
browsers, 13, 59 (VBScript), 725
HTML Layout Control
dynamic content, 10 Editor, 578
Microsoft Management Script, 576
Console, 656, 666-667 Script, 56

45.1389-6 index 811 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 812

45.1389-6 index 812 10/26/98, 10:20 AM

Brands3/nhq3 UsingActiveServer Pages polly 11.14.97 index LP#2
 Using
Special Edition
Contents i

Using
Active
Server Pages

00.1389-6 FM 1 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 ii Special Edition Using Active Server Pages

00.1389-6 FM 2 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Using
Special Edition
Contents iii

Using
Active
Server Pages

Scot Johnson, Keith Ballinger,

Davis Chapman, et al.

00.1389-6 FM 3 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 iv Special Edition Using Active Server Pages

Special Edition Using Active Server Pages

Copyright© 1997 by Que® Corporation.
All rights reserved. Printed in the United States of America. No part
of this book may be used or reproduced in any form or by any
means, or stored in a database or retrieval system, without prior
written permission of the publisher except in the case of brief quota-
tions embodied in critical articles and reviews. Making copies of
any part of this book for any purpose other than your own personal
use is a violation of United States copyright laws. For information,
address Que Corporation, 201 W. 103rd Street, Indianapolis, IN,
46290. You may reach Que’s direct sales line by calling 1-800-428-
5331 or by faxing 1-800-882-8583.
Library of Congress Catalog No.: 97-69808
ISBN: 0-7897-1389-6
This book is sold as is, without warranty of any kind, either express
or implied, respecting the contents of this book, including but not
limited to implied warranties for the book’s quality, performance,
merchantability, or fitness for any particular purpose. Neither Que
Corporation nor its dealers or distributors shall be liable to the
purchaser or any other person or entity with respect to any liability,
loss, or damage caused or alleged to have been caused directly or
indirectly by this book.
99 98 6 5 4 3 2
Interpretation of the printing code: the rightmost double-digit num-
ber is the year of the book’s printing; the rightmost single-digit
number, the number of the book’s printing. For example, a printing
code of 97-1 shows that the first printing of the book occurred in
1997.
All terms mentioned in this book that are known to be trademarks
or service marks have been appropriately capitalized. Que cannot
attest to the accuracy of this information. Use of a term in this book
should not be regarded as affecting the validity of any trademark or
service mark.
Screen reproductions in this book were created using Collage Plus
from Inner Media, Inc., Hollis, NH.

00.1389-6 FM 4 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents v

Contents at a Glance
I Introduction to Active Server Pages
1 Getting Started with Web Enabling: Creating Dynamic Applications
with Active Server Pages 09
2 Developing and Deploying Active Server Page Applications 29

II VBScript: The Foundation of Active Server Pages

3 VBScript Programming Basics: Variables and Scope 49
4 Working with Operators 69
5 Controlling Program Flow with VBScript 89
6 Working with Functions, Subroutines, and Dialog Boxes 111
7 Datatype Conversion Features 127
8 Putting It All Together with VBScript 143
9 Using the Script Debugger 161

III Active Server Objects: The Essential Building Blocks

10 The Composition of an Active Server Application 179
11 Controlling the Application: The Application Object 211
12 Controlling the Server with the Server Object and Using Transaction
Processing with the ObjectContext Object 229
13 Managing the User Session: The Session Object 247
14 Speaking to the Web Client: The Response Collection 267
15 Retrieving User Information: The Request Object 297
16 Putting It All Together: Building an Online Catalog 315

IV Active Server Components

17 Implementing Existing Server-Side Components 335
18 Creating Your Own Active Server Components 383
19 Putting It All Together: Creating an Online Registration System 409

V Database Management with Active Server Pages

20 Database Access with ADO 429
21 Working with ADO’s Connection Object 451
22 Using Command Objects 469

00.1389-6 FM 5 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 vi Special Edition Using Active Server Pages

23 Working with ADO’s Recordset Object 485

24 Managing Your Database 505
25 Accessing Databases with Remote Data Services 521
26 Putting It All Together: Creating a Dynamic Database Application 535

VI Active Server Pages and the Enterprise Solution

27 Visual InterDev: The Active Server Pages Integrated Development
Environment 561
28 Working with the Microsoft Transaction Server 585
29 Working with the Message Queue Server 613
30 Application Development for the Corporate Intranet/Extranet 639
31 Using Internet Information Server 4.0 to Manage Your Web Applications 653
32 Putting It All Together: Creating an N-Tier Online Catalog Application 673

Appendixes
A VBScript 2.0 Reference 699
B Active Server Object Reference 735
C ADO Reference 761
D Active Server Component Reference 775

Index 783

00.1389-6 FM 6 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents vii

Table of Contents

Introduction 1 The Evolution of Active Server Pages 18

Common Gateway Interface 18
The Purpose of This Book 2 Internet Server Application
Who Should Use This Book 2 Programming Interface 19
The HTML Object Model 20
How the Book Is Organized 3 Dynamic HTML Object Model 22
Part I Introduction to Active Active Server Pages and The
Server Pages 3 Active Server Model:
Part II VBScript: The Foundation A COM-Based Evolution 23
of Active Server Pages 3 The Composition of Active
Part III Active Server Objects: Server Pages 24
The Essential Building Blocks 3
Part IV Active Server From Here… 27
Components 4
Part V Database Management 2 Developing and Deploying Active
with Active Server Pages 4 Server Page Applications 29
Part VI Active Server Pages and the
Enterprise Solution 4 Planning Active Server Web Sites and
Applications 30
Conventions Used in This Book 4 Why Use Components? 30
Why Use Transactions? 32
Creating Active Server Pages 34
I Introduction to Active Server Visual InterDev 34
Pages FrontPage 35
ActiveX Scripting Languages: Linking
1 Getting Started with Web Enabling: It All Together 36
Creating Dynamic Applications with VBScript 36
Active Server Pages 9 JavaScript 36
JScript 36
Static versus Dynamic Content 10
Static Content 10 Scripting Tools 37
Dynamic HTML Content 10 Microsoft Script Wizard and the
Script Debugger 37
Designing and Deploying Web-Enabled Visual JavaScript 38
Applications: A Changing Architecture
for Changing Needs 12 The Internet Information Server 39
The Web Browser 13 The Microsoft Management
The Web Server 14 Console 39
The HTTP-TCP/IP Connection 14 Secure Socket Layer 3.0 and
The Database Server 15 Client Authentication 41
Proxy Servers 16 HTTP 1.1 Support 42
Firewalls 17 Server-Side Java 43
Certificate Servers 17
External Servers 18

00.1389-6 FM 7 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 viii Special Edition Using Active Server Pages

Extending Active Server Pages into Division 72

Your Existing Network 43 Exponentiation 72
Terminal Access 44 Integer Division Operator 73
Client/Server 44 Modulus Operator 74
N-Tier Client/Server 44
Comparison Operators 75
Web Access 45
Equality Comparison 77
N-Tier Web Access 45
Inequality Comparison 78
From Here… 45 Greater Than Comparison 78
Less Than Comparison 78
Object Comparison 79
II VBScript: The Foundation of Logical Operators 79
Active Server Pages Negation (NOT ) 80
Conjunction (AND) 81
Disjunction (OR ) 82
3 VBScript Programming Basics: Exclusion (XOR ) 82
Variables and Scope 49 Equivalence (EQV ) 83
What Is VBScript? 50 Concatenation Operators 84
VBScript’s History 51 & Operator 84
+ Operator 85
Extending Functionality on the
Web Client and Web Server 52 Operator Precedence 85
Communication Between Objects From Here… 87
and Events 52
Differences Between VBA, VB,
5 Controlling Program Flow with
and VBScript 55
Server- and Client-Side Syntax 57 VBScript 89
Security 58 Using Decision Structures 90
VBScript Variable Datatypes 60 The If...Then Condition 90
Variant 60 The If...Then...Else Statement 91
Subtypes 60 The Select Case Statement 94
Identifying the Datatype 62 Implementing Looping Logic 96
Using Variables 62 The For...Next Statement 96
Variable Scope 62 The Do...Loop Statement 99
Declaring Variables 64 The While...Wend Statement 100
Declaration Limitations 66 The For Each...Next Statement 101

From Here… 66 Interrupting the Program Flow 101

The Exit Statement 102
4 Working with Operators 69 The On Error Statement 103
Arithmetic Operators 70 Error Trapping with the Err Object 104
Addition 70 The Err Object’s Properties 107
Subtraction 72 Using the Err Object Methods 108
Multiplication 72
From Here… 109

00.1389-6 FM 8 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents ix

6 Working with Functions, 8 Putting It All Together with

Subroutines, and Dialog VBScript 143
Boxes 111 Application Overview 144
Introducing Event-Driven Creating Server-Side Active Server
Programming 112 Scripts with VBScript 146
Using Procedures: Subroutines Creating the Visual InterDev
and Functions 114 Project 147
General Procedure Rules 114 Building Data Retrieval and
Subroutines 118 Display 150
Functions 119 Processing the Logic 153
Minor Error Trapping 155
Using Dialog Boxes to Interact
with Users 120 Creating Client-Side VBScript 155
Message Box 120 Restructuring the HTML 156
Input Box 125 Building the Validation Subroutine 157

From Here… 126 Testing for a VBScript-Compliant

Browser 158
7 Datatype Conversion From Here… 159
Features 127
9 Using the Script Debugger 161
Identifying the Datatype 127
The IsArray Function 128 Using the Script Debugger 162
The IsDate Function 128 Managing the Script Components 165
The IsEmpty Function 129 Controlling Script Flow 167
The IsNull Function 129
Types of Errors 172
The IsNumeric Function 131
The IsObject Function 131 Runtime Errors 173
The VarType Function 132 Syntax Errors 173
Logical Errors 174
The Formatting Functions 133
The FormatDateTime Function 133 From Here… 175
The FormatCurrency Function 134
The FormatPercent Function 135
The FormatNumber Function 136 III Active Server Objects: The
Datatype Conversions 136 Essential Building Blocks
Cbool Function 137
CByte Function 138 10 The Composition of an Active Server
CCur Function 138 Application 179
CDate Function 139
CDbl, CInt, CLng, CSng 139 The Components of an Active Server
CStr Function 140 Page Application 180
Text Files 180
From Here… 141 Physical, Root, and Virtual
Directories 181
ASP Scripting Languages and
HTML 182

00.1389-6 FM 9 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 x Special Edition Using Active Server Pages

The Active Server Page Project File: Controlling Application-Level Variables

GLOBAL.ASA 184 with Methods 218
Tracking Application Events 186 Lock Method 218
Tracking Session Events 187 UnLock Method 219
Activating the GLOBAL.ASA File 187
Using Components at the
Understanding Scoping Issues in Application Level 220
Active Server Applications 189 Instantiating Application-Level
GLOBAL.ASA Scoping Level 190 Objects 220
Application-Level Scope 190 Using Arrays at the Application
Session-Level Scope 191 Level 222
Procedure-Level Scope 191
The Application Collection 226
The Fundamental Active Server The Contents Collection 226
Page Objects 192 The StaticObject Collection 227
The Application Object 192
From Here… 228
The Server Object 192
The ObjectContext Object 192
The Session Object 193 12 Controlling the Server with the
The Response Object 193 Server Object and Using Transaction
The Request Object 193 Processing with the ObjectContext
Debugging Active Server Pages 193 Object 229
Trapping Errors with the
Controlling the Server Object
Err Object 194
Using Methods 230
Trapping Errors with the Web
Expanding the Web Server with
Server Log File 195
the CreateObject Method 230
Trapping Errors by Writing
Managing Server Information
Text Files 195
with MapPath 232
Controlling Access to the Site 197
Using the Server Encoding Methods 234
Web Server Packet Filtering 197
Using the URLEncode Method 235
File Level Access Control 198
Using the HTMLEncode Method 237
Configuring Active Server Pages 199
The Server Object Properties 237
The Metabase 199
The Registry Settings 201 Using the ObjectContext Object for ASP
Web Server Interface Utilities 204 Transaction Processing 239
Using ASP Directives 205 Configuring the MTS components 239
Handling Known Active Server Setting Up ASP to Support
Page Issues 208 Transactions 240
From Here… 209 Controlling Success and Failure
with Methods 240
11 Controlling the Application: The The SetComplete Method 241
The SetAbort Method 241
Application Object 211
Trapping the ObjectContext’s Events 242
Trapping Application Events 212
OnTransactionCommit Event 242
Declaring Application-Level
OnTransactionAbort Event 243
Variables 214
Application_OnStart Event 217 From Here… 244
Application_OnEnd Event 217

00.1389-6 FM 10 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xi

13 Managing the User Session: Using the Response Object’s

Properties 273
The Session Object 247 The Buffer Property 274
Using Session Variables to Track The CacheControl Property 276
User Workflow 248 The CharSet Property 277
Identifying the User 248 The ContentType Property 280
The Problem with Cookies 249 The Expires Property 280
The ExpiresAbsolute Property 281
Implementing Session Variables The PICS Property 281
and Properties 250 The Status Property 283
Declaring Session Variables 251 The IsClientConnected Property 284
The SessionID Property 252
The Timeout Property 252 Sending Output to the Browser:
The LCID Property 254 The Response Methods 284
The CodePage Property 256 The Write Method 285
The AddHeader Method 289
Trapping Session Events 256 The AppendToLog Method 289
Session_OnStart 256 The BinaryWrite Method 290
Session_OnEnd 258 The Clear Method 290
Controlling User Session Resources 259 The End Method 291
The Abandon Method 259 The Flush Method 292
Configuring the Session Object 261 The Redirect Method 293

Using Session-Level Objects 261 From Here… 295

Instantiating Session-Level
Objects 261 15 Retrieving User Information:
Instantiating Session-Level The Request Object 297
Objects 262
Storing Arrays in the Session Understanding the Request Object 298
Object 262
Accepting User Information Using
Managing Session Variables Using the Request Collections 299
Collections 264 The ClientCertificate Collection 300
The Contents Collection 265 Retrieving Cookie Information
The StaticObject Collection 265 Using the Cookies Collection 301
The Form Collection 303
From Here… 266 The QueryString Collection 305
The ServerVariables Collection 309
14 Speaking to the Web Client:
The Request Properties and Methods 313
The Response Collection 267 The TotalBytes Property 313
The Response Collection 268 The BinaryRead Method 313
The Expires Attribute 270 From Here… 314
The Domain Attribute 271
The Path Attribute 272
The Secure Attribute 272 16 Putting It All Together: Building an
The HasKeys Attribute 272 Online Catalog 315
The ASP Online Catalog Application:
An Overview 316

00.1389-6 FM 11 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xii Special Edition Using Active Server Pages

Pre-Loading Application The Tools Component 354

Variables Using the GLOBAL.ASA 318 The FileExists Method 354
Creating Application-Level The Owner Method 356
Variables and Arrays 318 The PluginExists Method 356
Creating the User’s Shopping Basket The Random Method 357
Using Session-Level Arrays and
The Permission Checker Component 357
Variables 320
The HasAccess Method 358
Creating and Managing Active Server
The Content Linking Component 360
Pages and Workflow 321
The NextLink Files 360
Accessing Application-Level
The Content Linking List 361
Arrays 321
The Content Linking Methods 362
Using the Server Object’s Encoding
Techniques 324 The Content Rotator Component 371
Using ASPs as Non-Visual Code The Content Schedule File 371
Libraries 325 The Content Rotator Methods 373
Examining the Shopping Basket 326 The ChooseContent Method 373
Using the Request Object in a The GetAllContent Method 374
Multi-Button Form 328
Extracting Information Using the The Page Counter Component 375
Request Object 329 The Page Counter Component
Methods 376
From Here… 330 The Hits Method 377
The Reset Method 377
The Counter Component 378
IV Active Server Components The Get Method 379
The Increment Method 380
17 Implementing Existing Server-Side The Set Method 380
Components 335 The Remove Method 381

The Ad Rotator Component 336 From Here… 381

The GetAdvertisement Method 336
The Ad Rotator Component 18 Creating Your Own Active Server
Required Files 337 Components 383
The Rotator Schedule File 338
The Redirection File 339 Creating ActiveX Components 384
The Ad Rotator Properties 340 Dealing with Datatype Differences 389
The Browser Capabilities Component 342 Wrapping Built-In Visual Basic
The BROWSCAP.INI File 342 Functions 392
The Browser Capabilities
Component Properties 345 Encapsulating Complex Calculations 394

The Database Access Component 347 Porting Existing Applications 398

The Active Messaging/SMTP Accessing the ASP Intrinsic Objects from

Component 348 Within an ActiveX Component 401
Speeding Up Your ASP Program 405
From Here… 408

00.1389-6 FM 12 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xiii

19 Putting It All Together: Creating an 21 Working with ADO’s Connection

Online Registration System 409 Object 451
Rotating the Content with the Content Properties and Methods of the
Rotator Component 411 Connection Object 452
Connection Object Properties 452
Rotating the Banner with the Ad
Connection Object Methods 453
Rotator Component 414
Linking to Your Data with the
Verifying that a File Exist with the
Connection Object 454
Tools Component 415
Configuring the Connection
Detecting the Browser with the Browser Object 456
Capabilities Component 417 Controlling the Connection 459
Determining Page Hits with the Page Error Handling with ADO 463
Counter Component 419 The Properties and Methods of
the Error Collection 463
Build a Web Registration with the The Properties of the Error Object 463
Content Linker 420
The Command Object and the
From Here… 425 Connection Object 465
Revisiting the Employee Phone
V Database Management with Application 465

Active Server Pages From Here… 467

20 Database Access with ADO 429 22 Using Command Objects 469

Understanding ADO 430 The Command Object 470
OLE DB 430 Creating a Command Object 470
Web Application Development 431 Properties and Methods of the
Products for Web Development 432 Command Object 471
Using the Command Object to
The ADO Object Model 432 Access Data 475
Properties and Methods of ADO
Objects 433 The Parameters Collection and the
ADO in Action 443 Parameter Object 476
Benefits of Coding with the ADO Methods and Properties of the
Object Model 445 Parameters Collection 476
Methods and Properties of the
Transaction and Rollback Processing 446 Parameter Object 477
Parameters in Action 480
Remote Data Services 447
Revisiting the Phone Application 481
Creating a Simple Data-Driven ASP
Program 448 From Here… 482
From Here… 449

00.1389-6 FM 13 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xiv Special Edition Using Active Server Pages

23 Working with ADO’s Recordset Data Access 518

Database Security 519
Object 485
From Here… 519
Retrieving Data into a Recordset 486
Isolation 486
Locking 487 25 Accessing Databases with Remote
Identifying Cursor Types 487 Data Services 521
Recordset Properties 489 Understanding Remote Data Services 522
BOF/EOF 490 Creating Business Objects 523
RecordCount 491 Understanding IE 4 and Dynamic
AbsolutePosition 492 HTML 524
AbsolutePage 492 Active Server Pages and RDS 524
Filter 493
Using the RDS Components 524
Navigating the Recordset 495 Advanced Data Virtual Table
MoveFirst 495 Manager 524
MoveLast 495 AdvancedDataControl Object 525
MoveNext 496 AdvancedDataSpace Object 526
MovePrevious 497 AdvancedDataFactory Object 526
Manipulating the Recordset 498 Understanding the Limitations
Update 498 of RDS 527
AddNew 499
Extending the Phone Book
Delete 500
Application 527
Supports 500
From Here… 534
The Recordset in Action 500
From Here… 504 26 Putting It All Together: Creating a
Dynamic Database
24 Managing Your Database 505 Application 535
Understanding Databases 506
Loading the Products Table 536
Types of Databases 506
Create the ActiveX Data Object
SQL 507
Connection 536
Client/Server Development 509
Modifying the Inventory Database 540
Transaction Processing with
Deleting Items from the Inventory 540
Transaction Server 510
Adding Records to the Database 542
Transactions 511
Inside Transaction Server 511 Using Visual InterDev to Access
the Data 543
Concurrency Issues 512
Using Visual InterDev’s Design Data
ODBC 3.0 513 Control to Retrieve Records 543
Using the Visual InterDev Forms
Designing and Configuring Wizard to Create the Product
Databases 514 Administration Interface 551
Data Modeling 514
Keys and Indexes 517 Controlling Access to the Site 555
From Here… 557

00.1389-6 FM 14 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xv

VI Active Server Pages and the Creating Transaction Server

Components 603
Enterprise Solution Calling Component Methods 604
Returning Data from Components 604
27 Visual InterDev: The Active Server The Account Components 605
Pages Integrated Development Calling the Account Components
from ASP 606
Environment 561
Building Transaction Server Components
A Development Environment for for Use with Active Server Pages 610
Active Server Pages 562 Threading and Transaction Server
An Introduction to Visual InterDev 562 Components 610
Creating Projects and Workspaces 563 Parameter Passing with Active
The Visual InterDev Environment 565 Server Pages 611
Visual Database Tools 569 From Here… 612
The Integrated Data View 569
The Query Designer 570
The Database Designer 572
29 Working with the Message Queue
Data Form Wizards 573 Server 613
Database Design-Time Controls 574 What Is Message Queuing? 614
Productivity Features 575 Guaranteed Message Delivery 615
The Script Wizard 576 Store and Forward 616
The Visual InterDev Editor 577 Queue Browsing 617
HTML Layout Control Editor 578 Message Prioritizing and Aging 617
Microsoft Image Composer 581 Transactional Messages 618
Microsoft Music Producer 582 Journal and Dead-Letter Queues 618
Microsoft Media Manager 582
Microsoft’s Message Queue Server 619
From Here… 583 ActiveX Components 619
Creating Queues 620
28 Working with the Microsoft Locating Queues 622
Opening Queues 624
Transaction Server 585 Sending Messages 625
Understanding Distributed Transaction Receiving Messages 628
Processing 587 Closing Queues 628
Transaction Monitors 589 Deleting Queues 628
Object Request Brokers 591 Message Queue Server and Active
Introducing Microsoft Transaction Server Pages 629
Server 592 Creating a Message Queue 629
Managing Database Connections 593 Sending a Message to the Queue 631
Managing Distributed Objects 595 Receiving the Message from the
Transaction Coordination 596 Queue 634
Closing and Deleting the Queue 636
Integrating Active Server Pages with
Transaction Server 597 From Here… 637
Registering Components with
Transaction Server 597

00.1389-6 FM 15 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xvi Special Edition Using Active Server Pages

30 Application Development for the Creating the Customized Database Access

Component 675
Corporate Intranet/Extranet 639
Registering the Component 678
Developing Intranet Applications 640
Installing Components on the Microsoft
Leveraging Bandwidth 640
Transaction Server 678
When You Know What Browser
Using the Regsrv32 and DCOMCFG
They Have... 641
Utilities for Component
Sample Applications for the
Configuration 682
Intranet 643
Project Management Tool 643 Integrating the New ActiveX Components
into the ASPs 684
Building an Extranet 649
Calling the Component from the Active
Securing Your Extranet from
Server Page 684
Intruders 649
Creating the Generic Error Form 685
Confidential Information 650
Creating the Validation Server 686
From Here… 651
Creating the Order Processing and
31 Using Internet Information Inventory Control Database
Component 688
Server 4.0 to Manage Your Web Creating the GetProductInventory
Applications 653 Method 688
The Microsoft Management Console 654 Creating the AddProductInventory
Network Service Management 654 Method 689
Web Site Management 656 Creating the DeleteProductInventory
Configuring Active Server Application Method 690
Settings 661 Building and Testing Components 691
Setting Security Accounts 664 Putting It All Together: Filling the
Creating New Sites and Virtual Order 693
Directories 666
Changing the Default Site Values 667 From Here… 695

New Active Server Pages

Functionality 668
The Script Debugger 668
Appendixes
Transactional Scripts, Microsoft
Transaction Server, and Isolated ASP A VBScript 2.0 Reference 699
Applications 668
Scripting Active Server Pages 700
New ASP Components 670
Operators 703
Utilizing New HTTP 1.1 Support 670
Statements and Conditional
From Here… 671
Statements 704

32 Putting It All Together: Creating an Intrinsic Functions 705

The Abs() Function 710
N-Tier Online Catalog The Array() Function 710
Application 673 The Asc() Function 710
Create the Application Initialization The Atn() Function 710
Component 674 The CBool() Function 710

00.1389-6 FM 16 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xvii

The CByte() Function 711 The MsgBox() Function 720

The CCur() Function 711 The Now() Function 720
The CDate() Function 711 The Oct() Function 720
The CDbl() Function 711 The Replace() Function 720
The Chr() Function 711 The Right() Function 720
The CInt() Function 711 The Rnd() Function 721
The CLng() Function 711 The Round() Function 721
The Cos() Function 712 The RTrim() Function 721
The CreateObject() Function 712 The ScriptEngine() Function 721
The CSng() Function 712 The ScriptEngineBuildVersion()
The CStr() Function 712 Function 721
The Date() Function 712 The ScriptEngineMajorVersion()
The DateAdd() Function 712 Function 721
The DateDiff() Function 713 The ScriptEngineMinorVersion()
The DatePart() Function 713 Function 722
The DateSerial() Function 713 The Second() Function 722
The DateValue() Function 714 The Sgn() Function 722
The Day() Function 714 The Sin() Function 722
The Exp() Function 714 The Space() Function 722
The Filter() Function 714 The Split() Function 722
The Fix() Function 714 The Sqr() Function 723
The FormatCurrency() Function 714 The StrComp() Function 723
The FormatDateTime() Function 715 The StrReverse() Function 723
The FormatNumber() Function 715 The String() Function 723
The FormatPercent() Function 715 The Tan() Function 723
The Hex() Function 715 The Time() Function 723
The Hour() Function 716 The TimeSerial() Function 724
The InputBox() Function 716 The TimeValue() Function 724
The InStr() Function 716 The Trim() Function 724
The InStrRev() Function 716 The TypeName() Function 724
The Int() Function 716 The UBound() Function 724
The IsArray() Function 717 The UCase() Function 724
The IsDate() Function 717 The VarType() Function 725
The IsEmpty() Function 717 The Weekday() Function 725
The IsNull() Function 717 The WeekdayName() Function 725
The IsNumeric() Function 717 The Year() Function 725
The IsObject() Function 717
Constants 725
The Join() Function 718
Color Constants 725
The LBound() Function 718
Comparison Constants 726
The LCase() Function 718
The Date/Time Constants 726
The Left() Function 718
Date Format Constants 727
The Len() Function 718
File Input/Output Constants 728
The LoadPicture() Function 718
Miscellaneous Constant 728
The Log() Function 719
The MsgBox Constants 728
The LTrim() Function 719
String Constants 730
The Mid() Function 719
Tristate Constants 730
The Minute() Function 719
VarType Constants 731
The Month() Function 719
The MonthName() Function 719 VBScript Objects 732

00.1389-6 FM 17 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xviii Special Edition Using Active Server Pages

VBScript Object Methods 732 Retrieving Information Using the

Request Object 750
VBScript Object Properties 733 Accepting User Information
Using the Request Collections 750
B Active Server Object The Request Properties and
Reference 735 Methods 754
Controlling the Application Object 736 Managing the Session Object 754
Declaring Application-Level The Session Object:
Variables 736 A Cookie-Dependent System 755
Controlling Application-Level Variables Implementing Session Variables
with Methods 737 and Properties 755
Instantiating Application-Level Declaring Session Variables 756
Objects 738 Trapping Session Events 757
Controlling the Server Object 738 Controlling User Session
The CreateObject Method 739 Resources 758
The MapPath Method 739 Using Session-Level Objects 759
The ScriptTimeOut Property 739
Using Server Encoding Techniques 740
C ADO Reference 761
The URLEncode Method 740 Connection Object 762
Using the HTMLEncode Method 740 Properties 762
Methods 763
The ObjectContext Object: Interacting
with the Transaction Server 741 Errors Collection and Object 765
The SetComplete Method 741 Properties 765
The SetAbort Method 741
Command Object 765
Trapping the ObjectContext’s Events 741 Properties 765
The OnTransactionCommit Event 742 Methods 766
OnTransactionAbort Event 743
Parameters Collection and Objects 767
Communicating with Web Client: Properties 767
The Response Object 743 Methods 768
The Response Collection 744
The Cookies Collection 744 Recordset Object 768
The Response Object’s Properties 745 Properties 768
Methods 770
Sending Output to the Browser:
The Response Methods 747 Fields Collection and Objects 773
The Write Method 748 Properties 773
The AddHeader Method 748 Methods 773
The AppendToLog Method 748
The BinaryWrite Method 749 D Active Server Component
The Clear Method 749 Reference 775
The End Method 749
The Flush Method 749 Ad Rotator Component 776
The Redirect Method 749 Properties 776
Methods 776
Files 776

00.1389-6 FM 18 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xix

Browser Capabilities Component 777

Files 777
Content Linking Component 778
Methods 778
Files 779
File Access Components 779
FileSystemObject 779
Properties 779
TextStream 780

Index 783

00.1389-6 FM 19 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xx Special Edition Using Active Server Pages

Credits
PRESIDENT PRODUCT MARKETING MANAGER
Richard Swadley Kourtnaye Sturgeon

SENIOR VICE PRESIDENT/PUBLISHING TECHNICAL EDITOR

Don Fowley Ramesh Chandak

GENERAL MANAGER SOFTWARE SPECIALIST

Joe Muldoon Brandon K. Penticuff

MANAGER OF PUBLISHING OPERATIONS ACQUISITIONS COORDINATOR

Linda H. Buehler Carmen Krikorian

EXECUTIVE EDITOR SOFTWARE RELATIONS MANAGER

Brad Jones Susan D. Gallagher

EDITORIAL SERVICES DIRECTOR SOFTWARE COORDINATOR

Carla Hall Andrea Duvall

MANAGING EDITOR EDITORIAL ASSISTANT

Jodi Jensen Jennifer L. Chisholm

ACQUISITIONS MANAGER BOOK DESIGNER

Cheryl D. Willoughby Ruth Harvey

ACQUISITIONS EDITOR COVER DESIGNER

Kelly Marshall Sandra Schroeder

PRODUCT DIRECTOR PRODUCTION TEAM

Al Valvano Elizabeth Deeter
Shawn Ring
PRODUCTION EDITOR
Heather Kaufman Urschel INDEXER
Christine Nelson
COORDINATOR OF EDITORIAL SERVICES
Charlotte Clapp

WEBMASTER
Thomas H. Bennett

Composed in Century Old Style and ITC Franklin Gothic by Que Corporation.

00.1389-6 FM 20 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xxi

To my family, for remembering to slide food under the door, and to

Isaiah, the best 13-year-old Internet developer I know!

To my friends, who refused to believe my existence was more than bits

and bytes.

And to all the other stones that have been overturned in my life to
make me the person I am today. Scot Johnson

To my lovely wife, Lara, for believing in me. Keith Ballinger

To my son Jacob, who makes it all worth it. Steven Potts

00.1389-6 FM 21 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xxii Special Edition Using Active Server Pages

About the Authors

Scot Johnson (yes, only one “t”) ([email protected]) can often be found standing
on a soap box preaching how Internet technologies can play a critical role in current business
and communication environments. In his experience, Scot architects, develops, and manages
the transition of traditional desktop and mainframe applications to the Web. Scot has written
articles for various trade magazines and has been published in Visual InterDev Unleashed. In
his free time, Scot enjoys the sweet sounds of a Fender Stratocastor, dry martinis, fine cigars,
and, of course, a medium-rare steak.

Keith Ballinger is a partner with Vision Information Services, a company focused on Web
development and creating tools for Web developers. Keith currently spends much of his time at
client sites helping them to create the new generation of Web-based applications. When he’s
not working, Keith enjoys spending time with his wife and friends. He also sings and plays
guitar in the Portland, Oregon-based band, Conduit.

Davis Chapman first began programming computers while working on his Master’s Degree
in Music Composition. Writing applications for computer music, he discovered that he enjoyed
designing and developing computer software. It wasn’t long before he came to the realization
that he stood a much better chance of eating if he stuck with his newfound skill and demoted
his hard-earned status as a “starving artist” to a part-time hobby. Since that time, Davis has
focused on the art of software design and development, with a strong emphasis on the practical
application of client/server technology. Davis was the lead author for Que’s Building Internet
Applications with Delphi 2 and Web Development with Visual Basic 5, as well as a contributing
author on Que’s Running a Perfect Web Site, Second Edition, and Platinum Edition Using Visual
Basic 5. He has been with the Dallas, Texas-based consulting firm B.R. Blackmarr & Associ-
ates for the past seven years, and can be reached at [email protected].
Steven Potts is an Atlanta-based consultant specializing in Windows programming and
Internet software development. He has coauthored over a dozen books on a variety of comput-
ing topics, including Visual C++, Visual Basic, and Active Server Pages.

00.1389-6 FM 22 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 Contents xxiii

Acknowledgments
I would like to thank Corey Burnett for ensuring that I wasn’t writing technical jargon only to
myself. I would also like to thank the incredible staffs at Automated Concepts, Inc., Financial
Dynamics, Inc., and InterSoft Inc., for unsurpassed technical expertise, support, and inspira-
tion. Scot Johnson

Thanks to everyone at Que, especially Kelly and Heather: You both have been great! I also
couldn’t have done this without all the great support my family, friends, and business partners
have given me. Also, a special thanks to John Richardson for showing me how a real engineer
works and so much else. Keith Ballinger

00.1389-6 FM 23 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1

 xxiv Special Edition Using Active Server Pages

Tell Us What You Think!

As the reader of this book, you are our most important critic and commentator. We value your
opinion and want to know what we're doing right, what we could do better, what areas you'd
like to see us publish in, and any other words of wisdom you're willing to pass our way.

As the Executive Editor for the Programming team at Macmillan Computer Publishing, I wel-
come your comments. You can fax, email, or write me directly to let me know what you did or
didn't like about this book—as well as what we can do to make our books stronger.

Please note that I cannot help you with technical problems related to the topic of this book, and
that, due to the high volume of mail I receive, I might not be able to reply to every message.

When you write, please be sure to include this book's title and authors, as well as your name
and phone or fax number. I will carefully review your comments and share them with the
authors and editors who worked on the book.

Fax: 317-817-7070
Email: [email protected]

Mail: Brad Jones

Executive Editor
Macmillan Computer Publishing
201 W.103rd Street
Indianapolis, IN 46290 USA

CLICK HERE TO VIEW PAGE ONE

00.1389-6 FM 24 10/26/98, 8:43 AM

B6 SEU active Server Pages #1389-6 mary ellen 9/8/98 FM lp#1