0% found this document useful (0 votes)

96 views

SAS - Macro Language 1

Uploaded by

Cristiano

Available Formats

Download as PDF, TXT or read online on Scribd

0% found this document useful (0 votes)

96 views

SAS - Macro Language 1

Uploaded by

Cristiano

Available Formats

Download as PDF, TXT or read online on Scribd

You are on page 1/ 314

SAS® Macro Language 1:

Essentials

Course Notes
SAS® Macro Language 1: Essentials Course Notes was developed by Jim Simon. Additional contributions
were made by Linda Mitterling and Theresa Stemler. Editing and production support was provided by the
Curriculum Development and Support Department.
SAS and all other SAS Institute Inc. product or service names are registered trademarks or trademarks of
SAS Institute Inc. in the USA and other countries. ® indicates USA registration. Other brand and product
names are trademarks of their respective companies.
SAS® Macro Language 1: Essentials Course Notes
Copyright © 2016 SAS Institute Inc. Cary, NC, USA. All rights reserved. Printed in the United States of
America. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in
any form or by any means, electronic, mechanical, photocopying, or otherwise, without the prior written
permission of the publisher, SAS Institute Inc.
Book code E70750, course code LWMC1/MC1, prepared date 03Jun2016. LWMC1_003

ISBN 978-1-62960-434-3
For Your Infor mation iii

Table of Contents
Course Description ......................................................................................................vi

Prerequisites .............................................................................................................. vii

Chapter 1 Introduction....................................................................................... 1-1

1.1 Course Logistics............................................................................................... 1-3

Demonstration: Creating Course Data Files .................................................... 1-9
Exercises ..................................................................................................1-10

1.2 Purpose of the Macro Facility ............................................................................1-12

1.3 Program Flow .................................................................................................1-17

Exercises ..................................................................................................1-28

1.4 Solutions ........................................................................................................1-30

Solutions to Exercises.................................................................................1-30

Chapter 2 Macro Variables ................................................................................ 2-1

2.1 Introduction to Macro Variables .......................................................................... 2-3

2.2 Automatic Macro Variables ................................................................................ 2-7

2.3 Macro Variable References................................................................................2-10

Exercises ..................................................................................................2-29

2.4 User-Defined Macro Variables ...........................................................................2-31

Exercises ..................................................................................................2-45

2.5 Delimiting Macro Variable References ................................................................2-47

Exercises ..................................................................................................2-54

2.6 Macro Functions .............................................................................................2-56

Exercises ..................................................................................................2-69

2.7 Solutions ........................................................................................................2-70

Solutions to Exercises.................................................................................2-70

Chapter 3 Macro Definitions.............................................................................. 3-1

3.1 Defining and Calling a Macro............................................................................. 3-3

Exercises ..................................................................................................3-28
iv For Your Information

3.2 Macro Parameters ............................................................................................3-30

Demonstration: Macros with Positional Parameters.........................................3-35
Demonstration: Macros with Keyword Parameters..........................................3-38
Demonstration: Macros with Mixed Parameter Lists .......................................3-42
Exercises ..................................................................................................3-44

3.3 Solutions ........................................................................................................3-46

Solutions to Exercises.................................................................................3-46

Chapter 4 DATA Step and SQL Interfaces ......................................................... 4-1

4.1 Creating Macro Variables in the DATA Step ......................................................... 4-3

Demonstration: SYMPUTX Routine ............................................................4-10
Demonstration: SYMPUTX Routine ............................................................4-13
Demonstration: SYMPUTX Routine ............................................................4-15
Demonstration: Passing Values between Steps................................................4-19
Exercises ..................................................................................................4-20

4.2 Indirect References to Macro Variables ...............................................................4-23

Demonstration: Indirect References to Macro Variables ...................................4-34
Exercises ..................................................................................................4-36

4.3 Creating Macro Variables in SQL.......................................................................4-39

Exercises ..................................................................................................4-46

4.4 Solutions ........................................................................................................4-49

Solutions to Exercises.................................................................................4-49

Chapter 5 Macro Programs ............................................................................... 5-1

5.1 Conditional Processing ...................................................................................... 5-3

Exercises ..................................................................................................5-15

5.2 Parameter Validation ........................................................................................5-17

Exercises ..................................................................................................5-22

5.3 Iterative Processing..........................................................................................5-25

Exercises ..................................................................................................5-36

5.4 Global and Local Symbol Tables........................................................................5-39

Exercises ..................................................................................................5-53
For Your Infor mation v

5.5 Solutions ........................................................................................................5-56

Solutions to Exercises.................................................................................5-56

Chapter 6 Learning More ................................................................................... 6-1

6.1 SAS Resources................................................................................................. 6-3

6.2 Beyond This Course.......................................................................................... 6-6

Chapter 7 Supplemental Materials .................................................................... 7-1

7.1 Program Flow .................................................................................................. 7-3

vi For Your Information

Course Description
This course focuses on the components of the SAS macro facility and how to design, write, and debug
macro systems. Emphasis is placed on understanding how programs with macro code are processed.

To learn more…

For information about other courses in the curriculum, contact the SAS
Education Division at 1-800-333-7660, or send e-mail to [email protected].
You can also find this information on the web at http://support.sas.com/training/
as well as in the Training Course Catalog.

For a list of other SAS books that relate to the topics covered in this
course notes, USA customers can contact the SAS Publishing Department
at 1-800-727-3228 or send e-mail to [email protected]. Customers outside
the USA, please contact your local SAS office.
Also, see the SAS Bookstore on the web at http://support.sas.com/publishing/
for a complete list of books and a convenient order form.
For Your Infor mation vii

Prerequisites
Before attending this course, you should have completed the SAS ® Programming 2: Data Manipulation
Techniques course or have equivalent knowledge. Specifically, you should be able to
 use a DATA step to read from or write to a SAS data set or external file
 use DATA step programming statements such as IF-THEN/ELSE, DO WHILE, DO UNTIL, and
iterative DO
 use SAS data set options such as DROP=, KEEP=, and OBS=
 use character functions such as SUBSTR, SCAN, INDEX, and UPCASE
 form subsets of data using the WHERE clause
 create and use SAS date values and constants
 use SAS procedures such as SORT, PRINT, CONTENTS, MEANS, FREQ, TABULATE, and CHART.
viii For Your Information
Chapter 1 Introduction

1.1 Course Logistics ....................................................................................................... 1-3

Demonstration: Creating Course Data Files ................................................................ 1-9
Exercises............................................................................................................. 1-10

1.2 Purpose of the Macro Facility ................................................................................. 1-12

1.3 Program Flow .......................................................................................................... 1-17

Exercises............................................................................................................. 1-28

1.4 Solutions ................................................................................................................. 1-30

Solutions to Exercises ........................................................................................... 1-30
Solutions to Student Activities (Polls/Quizzes) ........................................................... 1-32
1-2 Chapter 1 Introduction

1.1 Course Logistics

Objectives
 Describe the data that is used in the course.
 Identify the interfaces that are available for workshops.
 Specify the naming convention that is used for
course files.
 Define the three levels of exercises.
 Explain the extended learning for this course.
 Execute a SAS program to create course data files.
 Execute a SAS program to define the data location.

3
3

Orion Star Sports & Outdoors

This course focuses on a fictitious global sports
and outdoors retailer that has traditional stores,
an online store, and a catalog business.

Orion Star Data

Large amounts of data are stored in various formats.

1,000
employees

150,000
orders 90,000
customers

64 suppliers

5
5

SAS Programming Interfaces

In this course, three interfaces are available for you
to use during workshops.

SAS SAS
SAS
Windowing Enterprise
Studio
Environment Guide

 To enable you to learn more about each of these

interfaces, self-directed exercises are available
in the course data location.
6

SAS Windowing Environment

The SAS windowing environment is an application that
is accessed from different operating environments.

The SAS windowing environment

provides a full programming
interface that can be used to write,
edit, and submit SAS code.

SAS Enterprise Guide

SAS Enterprise Guide is a client application that is
accessed from the Windows operating environment.

SAS Enterprise Guide provides the

following:
 a point-and-click interface with
menus and task wizards to
generate SAS code
 a full programming interface that
can be used to write, edit, and
submit SAS code

SAS Studio
SAS Studio is a web client that is accessed through
an HTML5-compliant web browser.

SAS Studio provides the following:

 a point-and-click interface with
menus and task wizards to
generate SAS code
 a full programming interface that
can be used to write, edit, and
submit SAS code

1.01 Multiple Choice Poll

Which interface will you use during this course?
a. SAS Studio
b. SAS Enterprise Guide
c. SAS windowing environment

Three Levels of Exercises

The course is designed to have you complete only one
set of exercises. Select the level that is most appropriate
for your skill set.
Level 1 Provides step-by-step instructions.

Level 2 Provides less information and guidance.

Challenge Provides minimal information and guidance.

You might need to use SAS Help and
documentation.

12
12

Extending Your Learning

After class, you will have access to an extended learning
page that was created for this course. The page includes
 course data and program files
 a PDF file of the course notes
 other course-specific resources.

Course Data Creation

You must execute programs to create the files and define
the location of the course data.

Data
location?

cre8data.sas
libname.sas

14
14

Creating Course Data Files

This demonstration illustrates how to create the course data files and define the data location.
1. In your SAS session, open the cre8data program.
2. Find the following %LET statement:
%let path=s:/workshop;
3. If your data files are to be created at a location other than s:\workshop, change the value that
is assigned to the path micro variable to reflect the location. If your data files are created in
s:\workshop, then no change is needed.

 The cre8data program uses forward slashes for portability across operating environments.
UNIX and Linux require forward slashes. Windows accepts forward slashes and might convert
them to back slashes.

4. Submit the program.

5. View the results and verify that the output contains a list of data files.
6. Open the libname program, which was created by the cre8data program.
7. Submit the program.
8. View the log and verify that there are no errors or warnings.

Exercises

Required Exercise

You must complete the exercise to create the course data files. If you do not create the data files,
all programs in this course will fail.

1. Creating Course Data Files

a. In your SAS session, open the cre8data program.
b. Find the following %LET statement:
%let path=s:/workshop;
c. If your data files are to be created at a location other than s:\workshop, change the value that
is assigned to the path micro variable to reflect the location. If your data files are created in
s:\workshop, then no change is needed.

d. Submit the program.

e. View the results and verify that the output contains a list of data files.

f. Open the libname program, which was created by the cre8data program.
%let path=s:/workshop;
libname orion "s:/workshop";
g. Submit the program.
h. View the log and verify that there are no errors or warnings.

1.2 Purpose of the Macro Facility

Objectives
 State the purpose of the macro facility.
 View examples of macro applications.

Purpose of the Macro Facility

The macro facility is a text processing facility for
automating and customizing the generation of SAS
code, minimizing the amount of code that you must
enter.

The macro facility supports the following:

 symbolic substitution within SAS code
 automated production of SAS code
 conditional construction of SAS code
 dynamic generation of SAS code
19

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
1.2 Purpose of the Macro Facility 1-13

Substituting System Values

Automatic macro variables minimize entering
of system values.
proc print data=orion.customer;
title "Customer List";
footnote1 "Created 10:24 Monday, 19MAY2014";
footnote2 "on the WIN System Using SAS 9.3";
run;

Substituting User-Defined Values

User-defined macro variables minimize entering
of repetitive values.

proc freq data=orion.order_fact;

where year(order_date)=2011;
table order_type;
title "Order Types for 2011";
run;
proc means data=orion.order_fact;
where year(order_date)=2011;
class order_type;
var Total_Retail_Price;
title "Price Statistics for 2011";
run;

Conditional Processing
Macro programs support conditional construction
of SAS code.

Example: Generate a detailed report every day

and a summary report on Friday.

Daily Report proc print data=orion.orders;

run;

Is it proc means data=orion.orders;

Yes run;
Friday?
22

Repetitive Processing
Macro programs support repetitive generation
of SAS code.

proc print data=orion.year2012;

run;

proc print data=orion.year2013;

run;

proc print data=orion.year2014;

run;

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
1.2 Purpose of the Macro Facility 1-15

Data-Driven Applications
Macro programs support dynamic data-driven generation
of SAS code.
Example: Create separate subsets of a selected data set
for each unique value of a selected variable.
data AU CA DE IL TR US ZA;
set orion.customer;
select(country);
when("AU") output AU;
when("CA") output CA;
when("DE") output DE;
when("IL") output IL;
when("TR") output TR;
when("US") output US;
when("ZA") output ZA;
otherwise;
end;
run;
24

Efficiency of Macro-Based Applications

Macro techniques can reduce program development
and maintenance time.

 The efficiency of a SAS program depends

on the efficiency of the underlying SAS code,
regardless of whether the SAS code was entered
manually or generated by macro techniques.
25

Developing Macro Applications

If a macro application generates SAS code, use
a five-step approach to rapid development and debugging.
Step 1 Write and debug a SAS program without macro coding.
Generalize the program by replacing hardcoded values
Step 2
with macro variable references. (Chapter 2)
Create a macro definition with macro parameters.
Step 3
(Chapter 3)
Add macro-level programming for conditional and iterative
Step 4
processing. (Chapter 5)
Step 5 Add data-driven customization. (Chapter 5)

1.02 Short Answer Poll

The macro facility is a ____ processing facility for
automating and customizing SAS code.

1.3 Program Flow

Objectives
 Identify the tokens in a SAS program.
 Describe how a SAS program is tokenized, compiled,
and executed.

Program Flow
When you submit a SAS program, it is copied
to a memory location called the input stack.

Input Stack

 A SAS program can be any combination

of the following:
 DATA steps and PROC steps
 global statements
 Structured Query Language (SQL)
 SAS Component Language (SCL)

38
 SAS macro language

Program Flow

Input Stack

Batch or
SUBMIT Stored Noninteractive
Com m and Process Subm ission
data bonus; %STPBEGIN; //SYSIN DD *
set orion.staff; title 'New Data'; options nodate;
bonus=salary*.1; proc print data=new; title 'Staff';
run; run; proc sql;
proc print; proc means data=new; select *
title 'Bonuses'; run; from orion.staff;
run; %STPEND; quit;

Program Flow
When SAS code is in the input stack, a component
of SAS called the word scanner does the following:
 reads the text in the input stack, character
by character, left to right, top to bottom
 breaks the text into fundamental units called tokens

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;

Program Flow
The word scanner recognizes four classes of tokens:
 name tokens

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;
42 ...

Name tokens contain one or more characters beginning with a letter or underscore and continuing with
letters, underscores, or numerals.

 Format and informat names (COMMA11.2, $5.) contain a period.

Program Flow
The word scanner recognizes four classes of tokens:
 name tokens
 special tokens

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;
43 ...

Special tokens can be any character, or combination of characters, other than a letter, underscore,
or numeral.

Program Flow
The word scanner recognizes four classes of tokens:
 name tokens
 special tokens
 literal tokens

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;
44 ...

A literal token is a string of characters enclosed in single or double quotation marks.

 The string is treated as a unit by the compiler.

Program Flow
The word scanner recognizes four classes of tokens:
 name tokens
 special tokens
 literal tokens
 number tokens

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;
45

Number tokens can be integer numbers or floating-point numbers containing a decimal point,
an exponent, or both.

 A SAS date constant ('01jan2012'd) is a number token.

1.03 Poll
Blanks are tokens.
 True
 False

Tokenization
A token ends when the word scanner detects
 a blank
 the beginning of another token.

data
Word bonus
;
Scanner

set orion.staff;
bonus=salary*.1;
run;
Input proc print;
Stack title 'Bonuses';
run;

 The maximum length of a token is 32,767 characters.

Program Flow
The word scanner passes tokens, one at a time,
to the appropriate compiler, as the compiler demands.
data bonus;
Compiler

set
orion
Word .
Scanner staff
;

bonus=salary*.1;
Input Stack run;
proc print;
title 'Bonuses';
run;

Program Flow
Compiler
Request tokens until
semicolon received

Repeat for each Syntax

statement check

SAS
Step boundary
encountered;
Suspend compilation

Repeat for If no compilation

each step errors, execute
compiled code
53

Program Flow
How does the macro language fit in?

data bonus;
Compiler

set
orion
Word .
Scanner staff
;

bonus=salary*.1;
Input Stack run;
proc print;
title 'Bonuses';
run;

Macro Triggers
During word scanning, two token sequences are
recognized as macro triggers:

&name-tok en a macro variable reference

%name-tok en a macro statement, function, or call

 The word scanner passes macro triggers

to the macro processor.

Program Flow
Where does the macro processor fit in?

Compiler

Word
Scanner

Input
Stack

Macro Processor
The macro processor executes macro triggers, including
macro variable resolution, macro language statements,
macro functions, and macro calls, requesting tokens
as necessary.

Compiler

Macro Processor
Word
Scanner

Input
Stack
57

Macro Statements
The following are characteristics of macro statements:
 begin with a percent sign (%) followed by a name token
 end with a semicolon
 represent macro triggers
 are executed by the macro processor

%PUT Statement
Example: Write text to the SAS log.
SAS Log
12 %put Invalid code!;
Invalid code!
%PUT text;

The %PUT statement does the following:

 writes text to the SAS log
 writes to column one of the next line
 writes a blank line if no text is specified

 Quotation marks are not required around text.

 The %PUT statement is valid in open code
(anywhere in a SAS program).

Program Flow
The %PUT statement is submitted.

Compiler
Macro Processor

Word
Scanner

Input %put Invalid code!;

Stack
60 ...

Program Flow
The statement is tokenized.

Compiler
Macro Processor

Word %
Scanner put
Invalid
code
!
;
Input
Stack
61 ...

Program Flow
The macro trigger is passed to the macro processor.

Compiler
Macro Processor

Word %put
Scanner
Invalid
code
!
;

Input
Stack
62 ...

Program Flow
The macro processor does the following:
 requests tokens until a semicolon is encountered
 executes the macro language statement

Compiler
Macro Processor

Word %put Invalid code!;

Scanner

Input
Stack
63

Exercises

Level 1

2. Writing Text to the SAS Log with the %PUT Statement

Submit a %PUT statement that writes your name to the SAS log.
3. Writing NOTE, WARNING, and ERROR Messages to the SAS Log with the %PUT Statement

a. Open the program m101e02 shown below into the Editor window.
%put NOTE: Is this a SAS note?;
%put WARNING: Is this a SAS warning?;
%put ERROR: Is this a SAS error?;
b. Submit the program and review the results in the SAS log. What is unusual about the results?
________________________________________________________________________
________________________________________________________________________
________________________________________________________________________
c. Replace the colon so that a hyphen follows the keywords NOTE, WARNING, and ERROR.
Submit the program and review the results in the SAS log. How do the results change?
______________________________________________________________________________
______________________________________________________________________________
______________________________________________________________________________

d. Modify the program so that the keywords NOTE, WARNING, and ERROR are in lowercase.
Submit the program and review the results in the SAS log. Did the change affect the results?
_________________________________________________________ ________________
______________________________________________________________________________
______________________________________________________________________________

Level 2

4. Writing Special Characters to the SAS Log with the %PUT Statement
a. Submit the following %PUT statement:
%put Can you display a semicolon ; in your %PUT statement?;
b. Does the %PUT statement generate any text? If so, what is the text displayed in the SAS log?
______________________________________________________________________________

______________________________________________________________________________
c. Does the %PUT statement generate any error messages? If so, what is the cause of the error?
______________________________________________________________________________
______________________________________________________________________________

d. Is the second %PUT interpreted as text or a macro keyword?

______________________________________________________________________________
______________________________________________________________________________

1.4 Solutions
Solutions to Exercises
1. Creating Course Data Files
This is exercise is self-guided, so no solution is needed.
2. Writing Text to the SAS Log with the %PUT Statement
%put Jane Doe;
3. Writing NOTE, WARNING, and ERROR Messages to the SAS Log with the %PUT Statement
a. Open the program into the Editor window.
%put NOTE: Is this a SAS note?;
%put WARNING: Is this a SAS warning?;
%put ERROR: Is this a SAS error?;
b. The keywords NOTE, WARNING, and ERROR make the results of the %PUT statements
resemble standard SAS NOTE, WARNING, and ERROR messages. Depending on the operating
environment, the messages might be color-coded or just displayed in bold.

c. The hyphen forces the keywords NOTE, WARNING, and ERROR to be suppressed.
d. The keywords NOTE, WARNING, and ERROR are case-sensitive and do not have the desired
effect when entered in lowercase.

4. Writing Special Characters to the SAS Log with the %PUT Statement
a. Submit the following %PUT statement:
%put Can you display a semicolon ; in your %PUT statement?;
b. The first semicolon in the %PUT statement is treated as a special token, not as plain text,
and ends the statement. The %PUT statement generates the following text:
Partial SAS Log
1 %put Can you display a semicolon ; in your %PUT statement?;
Can you display a semicolon

c. An error message is generated after the first semicolon because the word IN is interpreted
as an invalid keyword.

Partial SAS Log

1 %put Can you display a semicolon ; in your %PUT statement?;
--
180
ERROR 180-322: Statement is not valid or it is used out of proper order.

d. Because of the first semicolon, the second %PUT is interpreted as a macro keyword that
generates the following text:
Partial SAS Log
statement?

 The method for interpreting special tokens as plain text is addressed later in the course.

Solutions to Student Activities (Polls/Quizzes)

1.02 Short Answer Poll – Correct Answer

The macro facility is a text processing facility for
automating and customizing SAS code.

1.03 Poll – Correct Answer

Blanks are tokens.
 True
 False

Blanks are not tokens. Blanks delimit tokens.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
Chapter 2 Macro Variables
2.1 Introduction to Macro Variables................................................................................ 2-3

2.2 Automatic Macro Variables ....................................................................................... 2-7

2.3 Macro Variable References ..................................................................................... 2-10

Exercises............................................................................................................. 2-29

2.4 User-Defined Macro Variables................................................................................. 2-31

Exercises............................................................................................................. 2-45

2.5 Delimiting Macro Variable References .................................................................... 2-47

Exercises............................................................................................................. 2-54

2.6 Macro Functions...................................................................................................... 2-56

Exercises............................................................................................................. 2-69

2.7 Solutions ................................................................................................................. 2-70

Solutions to Exercises ........................................................................................... 2-70
Solutions to Student Activities (Polls/Quizzes) ........................................................... 2-77
2-2 Chapter 2 Macro Variables

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.1 Introduction to Macro Variables 2-3

2.1 Introduction to Macro Variables

Objectives
 Describe the purpose of macro variables.
 Describe where macro variables are stored.
 Identify the two types of macro variables.

Business Scenario
Production programs might need to be updated with
information, such as the current date, each time they
are run. Use macro variables to minimize typing
and make the code more flexible and reusable.

Macro Variable

4
4

What Are Macro Variables?

Macro variables store text, including the following:
 complete or partial SAS statements
 complete or partial SAS steps

How Are Macro Variables Stored?

Macro variables are stored in a memory area called
the global symbol table.

Global Symbol Table

Macro
Variables

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.1 Introduction to Macro Variables 2-5

Automatic Macro Variables

When a SAS job or session begins, the global symbol
table is created and initialized with automatic macro
variables.
Global Symbol Table
SYSDATE 03FEB15
Automatic SYSDATE9 03FEB2015
Variables SYSDAY Tuesday
SYSTIME 10:47
SYSUSERID joeuser
. .
. .

User-Defined Macro Variables

User-defined macro variables can be added to the global
symbol table.

Global Symbol Table

. .
. .
Automatic SYSTIME 10:47
Variables SYSUSERID joeuser
. .
. .
User-Defined OFFICE Sydney
Variables DATE1 19may2014
UNITS 4

The global symbol table is deleted at the end of your

SAS job or session.
8

Global Macro Variables

Macro variables in the global symbol table
 are global in scope (always available)
 have a minimum length of 0 characters (null value)
 have a maximum length of 65,534 (64K) characters
 store numeric tokens as text.

Global Symbol Table

. .
. .
Automatic SYSTIME 10:47
Variables SYSUSERID joeuser
. .
. .
User-Defined OFFICE Sydney
Variables DATE1 19may2014
9
UNITS 4

2.01 Short Answer Poll

What are the two types of macro variables?

2.2 Automatic Macro Variables

Objectives
 Identify selected automatic macro variables.
 Display automatic macro variables in the SAS log.

Business Scenario
Use automatic macro variables to minimize typing.

SYSDATE
SYSDATE9
SYSDAY
???

15
15

Automatic Macro Variables

The following are true for automatic macro variables:
 system-defined
 created at SAS invocation
 global in scope (always available)
 assigned values by SAS
 can be assigned values by the user in some cases

Automatic Macro Variables

Some automatic macro variables have fixed values that
are set at SAS invocation.
Name Description
SYSDATE Date of SAS invocation (03FEB15)
SYSDATE9 Date of SAS invocation (03FEB2015)
SYSDAY Day of the week of SAS invocation (Tuesday)
SYSTIME Time of SAS invocation (10:47).
SYSSCP Operating system abbreviation (WIN, OS, HP 64)
SYSVER Release of SAS software (9.4)
SYSUSERID Login or user ID of current SAS process

 The macro variables SYSDATE, SYSDATE9, and

17
SYSTIME store text, not SAS date or time values.

Automatic Macro Variables

Some automatic macro variables have values that change
automatically based on activity during your SAS session.
Name Description
SYSLAST Name of the most recently created SAS data set
in the form libref.name. If no data set has been
created, the value is _NULL_.
SYSPARM Value specified at SAS invocation.
SYSERR SAS DATA or PROC step return code (0=success).
SYSLIBRC LIBNAME statement return code (0=success).

Automatic Macro Variables

The _AUTOMATIC_ argument in the %PUT statement
writes the names and values of all automatic macro
variables to the SAS log.

Partial SAS Log

639582 %put _automatic_; %PUT _AUTOMATIC_;
AUTOMATIC AFDSID 0
AUTOMATIC AFDSNAME
AUTOMATIC AFLIB
AUTOMATIC AFSTR1
AUTOMATIC AFSTR2
AUTOMATIC FSPBDV
AUTOMATIC SYSADDRBITS 64
AUTOMATIC SYSBUFFR
AUTOMATIC SYSCC 3000
AUTOMATIC SYSCHARWIDTH 1
AUTOMATIC SYSCMD
AUTOMATIC SYSDATASTEPPHASE
AUTOMATIC SYSDATE 03FEB15
AUTOMATIC SYSDATE9 03FEB2015
AUTOMATIC SYSDAY Tuesday
19

2.02 Poll
Submit the statement below.
%put _automatic_;

Does SYSTIME match the current time?

 Yes
 No

2.3 Macro Variable References

Objectives
 Explain how macro variable references are handled
by the word scanner and macro processor.

Business Scenario
Use automatic macro variables to generate footnotes with
the following information:
 report time
Customer Country
 report day
Country Frequency Percent
 report date ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
AU 8 10.39
CA 15 19.48
DE 10 12.99
IL 5 6.49
TR 7 9.09
US 28 36.36
ZA 4 5.19

Created 10:47 Tuesday, 03FEB2015

25
25

Macro Variable References

Macro variable references begin with an ampersand (&)
followed by a macro variable name.

&myvar macro variable reference

Macro variable references

 are also called symbolic references
 can appear anywhere in a program
 are not case sensitive
 represent macro triggers
 are passed to the macro processor.

Macro Variable References

When the macro processor receives a macro variable
reference, it does the following:
 searches the symbol table for the macro variable

Symbol Table
&sysday
sysday Tuesday

Macro Variable References

When the macro processor receives a macro variable
reference, it does the following:
 searches the symbol table for the macro variable

Symbol Table
&sysday
sysday Tuesday

 resolves the macro variable by substituting its value

Symbol Table
Tuesday
sysday Tuesday

Macro Variable References

When the macro processor receives a macro variable
reference, it does the following:
 searches the symbol table for the macro variable

Symbol Table
&sysday
sysday Tuesday

 resolves the macro variable by substituting its value

Symbol Table
Tuesday
sysday Tuesday

 issues a warning to the SAS log if the macro variable

is not found in the symbol table
WARNING: Apparent symbolic reference SYDAY not resolved.

Macro Variable References

Example: Write the day of the week to the SAS log.
SAS Log
14 %put Today is &sysday;
Today is Tuesday
15 %put &=sysday;
SYSDAY=Tuesday

 The form of the %PUT statement that is shown

on line 15 was introduced in SAS 9.3.

An equal sign between the ampersand and the macro variable name displays the macro variable’s name
followed by the macro variable’s value.

Substitution within a Macro Statement

Compiler
Macro Processor
Word
Scanner

Symbol Table
%put Today is &sysday;
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

31 ...

Substitution within a Macro Statement

When a macro trigger is encountered, it is passed
to the macro processor for evaluation.

Compiler
Macro Processor
Word %put
Scanner

Symbol Table
Today is &sysday;
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

32 ...

Substitution within a Macro Statement

The macro processor requests additional tokens.

Compiler
Macro Processor
Word ; %put Today is &sysday
Scanner

Symbol Table
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

33 ...

Substitution within a Macro Statement

The macro variable reference triggers the macro
processor to search the symbol table for the reference.

Compiler
Macro Processor
Word ; %put Today is &sysday
Scanner

Symbol Table
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

34 ...

Substitution within a Macro Statement

The macro processor resolves the macro variable
reference, substituting its value.

Compiler
Macro Processor
Word ; %put Today is Tuesday
Scanner

Symbol Table
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

35 ...

Substitution within a Macro Statement

The macro processor requests additional tokens until
a semicolon is encountered. Then it executes the
%PUT statement, writing the resolved text to the SAS log.

Compiler
Macro Processor
Word %put Today is Tuesday;
Scanner

Symbol Table
Input
SYSDAY Tuesday
Stack
SYSLAST _NULL_

2.03 Short Answer Poll

Open and submit m102d01. What are the footnotes
in the PROC FREQ output?

proc freq data=orion.Customer;

table Country / nocum;
footnote 'Created &systime &sysday, &sysdate9';
run;

m102d01
37

Substitution within a SAS Literal

To reference macro variables within a literal, enclose
the literal in double quotation marks.
proc freq data=orion.Customer;
table Country / nocum;
footnote "Created &systime &sysday, &sysdate9";
run;

m102d01
39

Substitution within a SAS Literal

PROC FREQ Output
Customer Country

Country Frequency Percent

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
AU 8 10.39
CA 15 19.48
DE 10 12.99
IL 5 6.49
TR 7 9.09
US 28 36.36
ZA 4 5.19

Created 10:47 Friday, 06JAN2014

m102d01
40

Substitution within a SAS Literal

Compiler
Macro Processor
Word
Scanner

Symbol Table
proc print;
Input title "Today is &sysday"; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

41 ...

Substitution within a SAS Literal

SAS statements are passed to the compiler.

proc print;
Compiler title

Macro Processor
Word "
Today
Scanner is

Symbol Table
Input &sysday"; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

42 ...

Substitution within a SAS Literal

The macro trigger is passed to the macro processor.

proc print;
Compiler title

Macro Processor
Word " &sysday
Today
Scanner is

Symbol Table
Input "; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

43 ...

Substitution within a SAS Literal

The macro processor searches the symbol table.

proc print;
Compiler title

Macro Processor
Word " &sysday
Today
Scanner is

Symbol Table
Input "; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

44 ...

Substitution within a SAS Literal

The resolved reference is passed back to the input stack.

proc print;
Compiler title

Macro Processor
Word "
Today
Scanner is

Symbol Table
Input Tuesday"; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

45 ...

Substitution within a SAS Literal

Word scanning continues.

proc print;
Compiler title

Macro Processor
Word "
Today
Scanner is
Tuesday
"

Symbol Table
Input ; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

46 ...

Substitution within a SAS Literal

The literal is passed to the compiler as a unit.

proc print;
Compiler title "Today is Tuesday"

Macro Processor
Word
Scanner

Symbol Table
Input ; SYSDAY Tuesday
Stack run;
SYSLAST WORK.ALL

47 ...

Substitution within a SAS Literal

When a step boundary is encountered, compilation ends
and execution begins.

proc print;
Compiler title "Today is Tuesday";

Macro Processor
Word
Scanner
run;

Symbol Table
Input
SYSDAY Tuesday
Stack
SYSLAST WORK.ALL

2.04 Multiple Choice Poll

Macro variable references are resolved by which
of the following?
a. SAS compiler
b. macro processor
c. word scanner

Substitution within SAS Code

Example: Generalize PROC PRINT to print the last created
data set, using the automatic macro variable
SYSLAST.

Compiler proc print data=ORION.ALL;

title "Listing of ORION.ALL";

Macro Processor
Word
Scanner run;

Symbol Table
proc print data=&syslast;
Input title "Listing of &syslast";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

51 ...

Substitution within SAS Code

SAS statements are passed to the compiler. When
a macro trigger is encountered, it is passed to the macro
processor for evaluation.

Compiler proc print data=ORION.ALL;

title "Listing of ORION.ALL";

Macro Processor
Word &syslast
Scanner run;

Symbol Table
;
Input title "Listing of &syslast";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

52 ...

Substitution within SAS Code

The macro variable reference triggers the macro processor
to search the symbol table for the reference.

Compiler proc print data=ORION.ALL;

title "Listing of ORION.ALL";

Macro Processor
Word &syslast
Scanner run;

Symbol Table
;
Input title "Listing of &syslast";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

53 ...

Substitution within SAS Code

The macro processor resolves the macro variable
reference, passing its resolved value back to the input
stack.

Compiler proc print data=ORION.ALL;

title "Listing of ORION.ALL";

Macro Processor
Word
Scanner run;

Symbol Table
ORION.ALL;
Input title "Listing of &syslast";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

54 ...

Substitution within SAS Code

Word scanning continues.

Compiler proc print data=ORION.ALL;

Macro Processor
Word
Scanner run;

Symbol Table
Input title "Listing of &syslast";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

55 ...

Substitution within SAS Code

A step boundary is encountered. Compilation ends.
Execution begins.

proc print data=ORION.ALL;

Compiler title "Listing of ORION.ALL";

Macro Processor
Word
Scanner run;

Symbol Table
Input SYSDAY Tuesday
Stack SYSLAST ORION.ALL

Unresolved Reference
Example: Reference a nonexistent macro variable within
a SAS literal.

proc print data=ORION.ALL;

Compiler title "Listing of ORION.ALL";

Macro Processor
Word "
Expenses
Scanner for
R

Symbol Table
proc print data=orion.exp;
Input title "Expenses for R&D";
SYSDAY Tuesday
Stack run; SYSLAST ORION.ALL

57 ...

Unresolved Reference
The macro trigger is passed to the macro processor for
evaluation.

proc print data=orion.exp;

Compiler title

Macro Processor
Word " &D
Expenses
Scanner for
R

Symbol Table
";
Input run;
SYSDAY Tuesday
Stack SYSLAST _NULL_

58 ...

Unresolved Reference
The macro processor issues a warning to the SAS log
when it cannot resolve a reference.

WARNING: Apparent symbolic

proc print data=orion.exp; reference D not resolved.
Compiler title

Macro Processor
Word " &D
&D
Expenses
Scanner for
R

Symbol Table
";
Input run;
SYSDAY Tuesday
Stack SYSLAST _NULL_

59 ...

Unresolved Reference
If the macro processor cannot resolve a reference,
the tokens are passed back to the word scanner.
The word scanner passes them to the compiler.
WARNING: Apparent symbolic
proc print data=orion.exp; reference D not resolved.
Compiler title "Expenses for R&D"

Macro Processor
Word ;
run
Scanner ;

Symbol Table
Input SYSDAY Tuesday
Stack SYSLAST _NULL_

Unresolved Reference
Example: Reference a nonexistent macro variable within
SAS code.
proc print data=&sydlast;
title "Listing of &syslast";
run;

Unresolved Reference
SAS Log
1 proc print data=&sydlast;
-
22
200
WARNING: Apparent symbolic reference SYDLAST not resolved.
ERROR: File WORK.SYDLAST.DATA does not exist.
ERROR 22-322: Expecting a name.
ERROR 200-322: The symbol is not recognized and will be ignored.
2 title "Listing of &syslast";
3 run;

NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.65 seconds
cpu time 0.09 seconds

Exercises

Submit a LIBNAME statement to assign the orion libref to the course SAS library according
to instructions provided by the instructor.
libname orion '________________________________';

Level 1
1. Displaying Automatic Macro Variables
a. Use the %PUT statement to list all automatic macro variables in the SAS log.
b. What are the values of the following automatic macro variables?
 SYSLAST _____________________________________________________________
 SYSUSERID _____________________________________________________________
 SYSTIME _____________________________________________________________
 SYSDATE9 _____________________________________________________________
c. Are the values for the automatic macro variables SYSTIME and SYSDATE9 accurate?
_____________________________________________________________________________
______________________________________________________________________________

Level 2
2. Using Automatic Macro Variables
a. Using the SORT procedure, sort the data set orion.continent by Continent_Name.
 Use the OUT= option in the PROC SORT statement so that you do not overwrite
the original data set.
b. Using the PRINT procedure and an automatic macro variable, print the most recently created data
set and display the data set name in the title.
c. Submit the program and examine the results.
3. Using Automatic Macro Variables
a. What is the value of the automatic macro variable SYSLAST after the following DATA step
is submitted?
_____________________________________________________________________
data new;
set orion.continent;
run;
b. What is the value of the automatic macro variable SYSLAST after the following PROC PRINT
step is submitted?
_____________________________________________________________________
proc print data=orion.continent;
run;

Challenge
4. Using SAS Date Constants
a. Open the m102e04 program shown below into the Editor window.
proc print data=orion.employee_payroll;
format Birth_Date Employee_Hire_Date date9.;
run;
b. Modify the program so that it subsets the data to return only the employees hired between
January 1, 2007, and today. Use the automatic macro variable SYSDATE9 to return today’s date.
PROC PRINT Output
Employee_ Birth_ Employee_ Employee_ Marital_
Obs Employee_ID Gender Salary Date Hire_Date Term_Date Status Dependents

310 121034 M 27110 23AUG1988 01JAN2007 . S 0

361 121085 M 32235 12NOV1986 01JAN2007 . S 0
364 121088 M 27240 10JUN1988 01JAN2007 . S 0

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-31

2.4 User-Defined Macro Variables

Objectives
 Create user-defined macro variables.
 Display values of user-defined macro variables
in the SAS log.

Business Scenario
Further automate your programs with user-defined macro
variables.

&myvar

67
67

%LET Statement
The %LET statement creates a macro variable
and assigns it a value.

%let name=Ed Norton;

%LET variable=value;

Symbol Table
name Ed Norton

 %LET statements are valid in open code (anywhere

in a SAS program).

User-Defined Macro Variables

User-defined macro variables have the following
characteristics:
 Macro variable names follow SAS naming conventions.
 If the macro variable already exists, its value
is overwritten.
 If the variable or value contain macro triggers,
the triggers are evaluated before the %LET statement
executes.

%let name=Ed Norton;

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-33

User-Defined Macro Variable Values

Macro variable values have the following characteristics:
 Minimum length is 0 characters (null value).
 Maximum length is 65,534 characters (64K).
 Number tokens are stored as text strings.
 Mathematical expressions are not evaluated.
 Case is preserved.
 Leading and trailing blanks are removed.
 Quotation marks, if any, are stored as part
of the value.
%let name=Ed Norton;

%LET Statement Examples

Determine the name and value of each macro variable
created by these %LET statements.
Global Symbol Table
%let name= Ed Norton ;
%let name2=' Ed Norton ';
%let title="Joan's Report";
%let start=;
%let sum=3+4;
%let total=0;
%let total=&total+∑
%let x=varlist;
%let &x=name age height;

71 ...

%LET Statement Examples

Leading and trailing blanks are removed.

Global Symbol Table

nam e Ed Norton
%let name= Ed Norton ;
%let name2=' Ed Norton ';
%let title="Joan's Report";
%let start=;
%let sum=3+4;
%let total=0;
%let total=&total+∑
%let x=varlist;
%let &x=name age height;

72 ...

%LET Statement Examples

Quotation marks are stored as part of the value.

Global Symbol Table

nam e Ed Norton
%let name= Ed Norton ;
%let name2=' Ed Norton '; nam e2 ' Ed Norton '
%let title="Joan's Report";
%let start=;
%let sum=3+4;
%let total=0;
%let total=&total+∑
%let x=varlist;
%let &x=name age height;

73 ...

Storing quotation marks as part of a macro variable’s value is not recommended.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-35

%LET Statement Examples

Quotation marks are stored as part of the value.

Global Symbol Table

nam e Ed Norton
%let name= Ed Norton ;
%let name2=' Ed Norton '; nam e2 ' Ed Norton '
%let title="Joan's Report"; title "Joan's Report"
%let start=; start
%let sum=3+4;
%let total=0;
%let total=&total+∑
%let x=varlist;
%let &x=name age height;

74 ...

Storing quotation marks as part of a macro variable’s value is not recommended.

%LET Statement Examples

A null value is stored.

Global Symbol Table

75 ...

%LET Statement Examples

Mathematical expressions are not evaluated.

Global Symbol Table

nam e Ed Norton
%let name= Ed Norton ;
%let name2=' Ed Norton '; nam e2 ' Ed Norton '
%let title="Joan's Report"; title "Joan's Report"
%let start=; start
%let sum=3+4; sum 3+4
%let total=0;
%let total=&total+∑
%let x=varlist;
%let &x=name age height;

76 ...

%LET Statement Examples

Numeric tokens are stored as character strings.

Global Symbol Table

77 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-37

%LET Statement Examples

The macro trigger is evaluated before the %LET statement
executes. The previous value of total is replaced.
Global Symbol Table
nam e Ed Norton
%let name= Ed Norton ;
%let name2=' Ed Norton '; nam e2 ' Ed Norton '
%let title="Joan's Report"; title "Joan's Report"
%let start=; start
%let sum=3+4; sum 3+4
%let total=0;
%let total=&total+∑ total 0+3+4
%let x=varlist;
%let &x=name age height;

78 ...

%LET Statement Examples

The text value is stored.

Global Symbol Table

79 ...

%LET Statement Examples

The macro variable’s name resolves to varlist.

Global Symbol Table

2.05 Short Answer Poll

What is the value of the macro variable z?

%let x=15;
%let y=10;
%let z=&x-&y;

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-39

%LET Statement Examples

Use a macro variable to select a numeric value.

Step 1 Hardcode the numeric value.

proc print data=orion.Order_Fact;

where Quantity > 4;
var Order_Date Product_ID Quantity;
title "Orders exceeding 4 units";
run;

m102d02
83

%LET Statement Examples

Use a macro variable to select a numeric value.

Step 2 Create a macro variable.

%let units=4;

proc print data=orion.Order_Fact;

where Quantity > 4;
var Order_Date Product_ID Quantity;
title "Orders exceeding 4 units";
run;

m102d02
84

 Macro variables store numbers as text, not as numeric values.

%LET Statement Examples

Use a macro variable to select a numeric value.

Step 3 Reference the macro variable.

%let units=4;

proc print data=orion.Order_Fact;

where Quantity > &units;
var Order_Date Product_ID Quantity;
title "Orders exceeding &units units";
run;

Partial SAS Log

NOTE: There were 6 observations read from the data set ORION.ORDER_FACT.
WHERE Quantity>4;

m102d02
85

 Macro variables store numbers as text, not as numeric values.

%LET Statement Examples

Use a macro variable to select a character value.

Step 1 Hardcode the character value.

proc print data=orion.Employee_Addresses;

where City="Sydney";
var Employee_Name;
title "Sydney Employees";
run;

m102d03
86

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-41

%LET Statement Examples

Use a macro variable to select a character value.

Step 2 Create a macro variable.

%let office=Sydney;

proc print data=orion.Employee_Addresses;

where City="Sydney";
var Employee_Name;
title "Sydney Employees";
run;

m102d03
87

%LET Statement Examples

Use a macro variable to select a character value.

Step 3 Reference the macro variable.

%let office=Sydney;

proc print data=orion.Employee_Addresses;

where City="&office";
var Employee_Name;
title "&office Employees";
run;

Partial SAS Log

NOTE: There were 67 observations read from the data set
ORION.EMPLOYEE_ADDRESSES.
WHERE City='Sydney';

m102d03
88

%LET Statement Examples

Use macro variables to select date values.
%let date1=25may2011;
%let date2=15jun2011;

proc print data=orion.Order_Fact;

where Order_Date between "&date1"d and "&date2"d;
var Order_Date Product_ID Quantity;
title "Orders between &date1 and &date2";
run;
Partial SAS Log
NOTE: There were 11 observations read from the data set
ORION.ORDER_FACT.
WHERE (Order_Date>='25MAY2011'D and Order_Date<='15JUN2011'D);

Orders between 25may2011 and 15jun2011

Order_
Obs Date Product_ID Quantity

526 05JUN2011 240100100679 2

527 06JUN2011 230100400007 1 m102d04
89

2.06 Multiple Choice Poll

Which WHERE statement is correct?
a. where City=&office;
b. where City='&office';
c. where City="&office";

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-43

Displaying Macro Variables

The _USER_ argument in the %PUT statement writes
the names and values of all user-defined macro
variables to the SAS log.

SAS Log
%PUT _USER_;
175 %put _user_;
GLOBAL DATE1 25may2011
GLOBAL DATE2 15jun2011
GLOBAL OFFICE Sydney
GLOBAL UNITS 4

Use _ALL_ to display all user-defined and automatic macro variables in the SAS log, as follows:
%put _all_;

Displaying Macro Variables

The SYMBOLGEN system option writes macro variable
values to the SAS log as they are resolved.

SAS Log
176 options symbolgen;
OPTIONS SYMBOLGEN;
177 %let office=Sydney;
178 proc print data=orion.employee_addresses;
179 where city="&office";
SYMBOLGEN: Macro variable OFFICE resolves to Sydney
180 var employee_name;
SYMBOLGEN: Macro variable OFFICE resolves to Sydney
181 title "&office Employees";
182 run;

 The default value is NOSYMBOLGEN.

After debugging, issue the following statement to turn off the SYMBOLGEN option:
options nosymbolgen;

Deleting User-Defined Macro Variables

The %SYMDEL statement deletes one or more
user-defined macro variables from the global symbol table.
%SYMDEL macro-variables;

%symdel office units;

Global Symbol Table

DATE1 25may2011
DATE2 15jun2011
OFFICE Sydney
UNITS 4

 Delete unneeded macro variables from the global

symbol table to release memory.
94

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.4 User- Defined Macro Variables 2-45

Exercises
Level 1
5. Defining and Using Macro Variables for Character Substitution
a. Open the m102e05 program shown below into the Editor window. Submit the program
and examine the output that it creates.
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains 'Gold';
title 'Gold Customers';
run;
b. Precede the program with a %LET statement to assign the value Gold to type. Modify the program
so that the two occurrences of Gold are replaced by references to the macro variable type. Submit
the program. It produces the same output as before.
c. Include the appropriate system option to display resolved values of macro variables in the SAS log.
Resubmit the program and examine the log.
d. Modify the value of type to Internet. Resubmit the program and examine the log.
e. Turn off the system option from part c above.

Level 2
6. Defining and Using Macro Variables for Numeric Substitution
a. Open the m102e06 program shown below into the Editor window. Edit the program to display only
the Gold-level customers between the ages of 30 to 45.
 Customer ages range from 19 to 73.
%let type=Gold;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
SAS Output
Gold Customers between 30 and 45

Customer_ Customer_
Obs Customer_Name Gender Age

3 Cornelia Krahl F 33
11 Oliver S. Füßling M 43
32 James Klisurich M 38
35 Viola Folsom F 38
40 Kyndal Hooks F 43
57 Rita Lotz F 43
75 Angel Borwick F 38

b. Modify the program so that the values 30 and 45 are replaced by references to the macro variables
age1 and age2, respectively.
c. Include the appropriate system option to display resolved values of macro variables in the SAS log.
Resubmit the program and examine the log.
d. Modify the values of age1 and age2. Resubmit the program and examine the log.
e. Turn off the system option from part c above.
7. Deleting Macro Variables
a. Open the program m102e07 program shown below into the Editor window. Submit the program
to create the macro variables.
%let pet1=Paisley;
%let pet2=Sitka;
b. Delete all user-defined macro variables.
c. Use the %PUT statement to verify that the macro variable deletion is successful.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.5 Delimiting Macro Variable References 2-47

2.5 Delimiting Macro Variable References

Objectives
 Code macro variable references with leading
or trailing text.
 Code adjacent macro variable references.

Business Scenario
The production program below requires the user
to repetitively enter the desired year, month,
and analysis variable. Automate the program with
user-defined macro variables. year

proc chart data=orion.y2010jan;

hbar week / sumvar=sale;
run; month
proc plot data=orion.y2010jan;
plot sale*day;
run;

analysis variable

What is different about this scenario?

Combining Macro Variables with Text

A macro variable reference can appear anywhere,
but additional syntax is sometimes required.

proc chart data=orion.y&year&month;

hbar week / sumvar=&var;
run;
proc plot data=orion.y&year&month;
plot &var*day;
run;

Leading text
Adjacent macro variable references
Trailing text

Which of the above situations might pose a problem?

100

Leading Text
Leading text is never a problem.
%let month=jan;
proc chart data=orion.y2010&month;
hbar week / sumvar=sale;
run;
proc plot data=orion.y2010&month;
plot sale*day;
run;
proc chart data=orion.y2010jan;
hbar week / sumvar=sale;
run;
proc plot data=orion.y2010jan;
plot sale*day;
run;

101

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.5 Delimiting Macro Variable References 2-49

Adjacent Macro Variable References

Adjacent macro variable references are never a problem.
%let year=2010;
%let month=jan;
proc chart data=orion.y&year&month;
hbar week / sumvar=sale;
run;
proc plot data=orion.y&year&month;
plot sale*day;
run;
proc chart data=orion.y2010jan;
hbar week / sumvar=sale;
run;
proc plot data=orion.y2010jan;
plot sale*day;
run;

102

Trailing Text
Trailing text can be a problem. Why?
Why is trailing text not a problem here?
%let year=2010;
%let month=jan;
%let var=sale;
proc chart data=orion.y&year&month;
hbar week / sumvar=&var;
run;
proc plot data=orion.y&year&month;
plot &var*day;
run;
proc chart data=orion.y2010jan;
hbar week / sumvar=sale;
run;
proc plot data=orion.y2010jan;
plot sale*day;
run;
103

Trailing Text
Trailing text is a problem if it changes the reference.
It is not a problem here because of the special character.
%let year=2010;
%let month=jan;
%let var=sale;
proc chart data=orion.y&year&month;
hbar week / sumvar=&var;
run;
proc plot data=orion.y&year&month;
plot &var*day;
run;
proc chart data=orion.y2010jan;
hbar week / sumvar=sale;
run;
proc plot data=orion.y2010jan;
plot sale*day;
run; special character
104

Setup for the Poll

SAS software includes the following procedures:

Create a macro variable that toggles between

a Base SAS or SAS/GRAPH procedure.

105

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.5 Delimiting Macro Variable References 2-51

2.07 Short Answer Poll

Create a macro variable that toggles between
a Base SAS or SAS/GRAPH procedure.
/* GRAPHICS should be null or G */
%let graphics=g;
proc &graphicschart data=orion.y&year&month;
hbar week / sumvar=&var;
run;
proc &graphicsplot data=orion.y&year&month;
plot &var*day;
run;

What is the problem with this reference?

106

Trailing Text Solution

A period (.) is a special delimiter that ends a macro variable
reference.

&myvar. special delimiter

The period does not appear as text when the macro

variable is resolved.

 The word scanner recognizes the end of a macro

variable reference when it encounters a character
that cannot be part of the reference.

108

Trailing Text Solution

Use a period to delimit the end of the macro variable
reference.
/* GRAPHICS should be null or G */
%let graphics=g;
proc &graphics.chart data=orion.y&year&month;
hbar week / sumvar=&var;
run;
proc &graphics.plot data=orion.y&year&month;
plot &var*day;
run; proc gchart data=orion.y2010jan;
hbar week / sumvar=sale;
run;
proc gplot data=orion.y2010jan;
plot sale*day;
run;

109  The generated code does not include the period.

2.08 Short Answer Poll

Modify the previous example to include a macro variable
that stores a libref.
%let lib=orion;
%let graphics=g;
libname &lib "&path";
proc &graphics.chart data=&lib.y&year&month;
hbar week / sumvar=&var;
run;
proc &graphics.plot data=&lib.y&year&month;
plot &var*day;
run;

What is the problem this time?

110

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.5 Delimiting Macro Variable References 2-53

Corrected Program
Use another period.

delimiter text

proc &graphics.chart data=&lib..y&year&month;

proc gchart data=orion.y2010jan;

112

Exercises

Level 1
8. Consecutive Macro Variable References
a. Open the m102e08 program shown below into the Editor window. Submit the program
and examine the output that it creates.
proc print data=orion.organization_dim;
where Employee_Hire_Date='01AUG2006'd;
id Employee_ID;
var Employee_Name Employee_Country Employee_Hire_Date;
title 'Personal Information for Employees Hired in AUG
2006';
run;
b. Precede the program with %LET statements to assign the value AUG to month and the value 2006
to year. Modify the program so that the two occurrences of AUG and 2006 are replaced by
references to the macro variables month and year, respectively. Submit the program. It produces
the same output as before.
c. Modify the value of month to JUL and year to 2003. Resubmit the program.

Level 2
9. Macro Variable References with Delimiters
a. Open the m102e09 program shown below into the Editor window. Submit the program
and examine the output that it creates.
proc print data=orion.organization_dim;
id Employee_ID;
var Employee_Name Employee_Country Employee_Gender;
title 'Listing of All Employees From
Orion.Organization_Dim';
run;
b. Modify the program so that all occurrences of Organization and Employee are replaced with macro
variable references called dsn and var, respectively. Submit the program and verify the output.
 When substituting for the hardcoded Employees in the TITLE statement, be sure to keep
the ending s as part of the title text.
c. Modify the value of dsn to Customer and var to Customer. Resubmit the program.

Challenge
10. Macro Variable References with Multiple Delimiters
a. Open the m102e10 program shown below into the Editor window. This program analyzes
the orion.staff data to find the employee with the most seniority within a job title. Submit
the program and examine the output that it creates.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
2.5 Delimiting Macro Variable References 2-55

proc sort data=orion.staff out=staffhires;

by Job_Title Emp_Hire_Date;
run;
data FirstHired;
set staffhires;
by Job_Title;
if First.Job_Title;
run;
proc print data=FirstHired;
id Job_Title;
var Employee_ID Emp_Hire_Date;
title "First Employee Hired within Each Job Title";
run;
b. Create a macro variable that acts as a toggle such that the program selects employees with either
the most seniority or the least seniority within their job title. Make sure that the title reflects this
difference.
Partial PROC PRINT Output
First Employee Hired within Each Job Title

Emp_Hire_
Job_Title Employee_ID Date

Account Manager 120746 01APR2002

Accountant I 120752 01AUG1975
Accountant II 120771 01DEC1976
Accountant III 120757 01JAN1974
Administration Manager 120104 01JAN1981
Applications Developer I 120797 01DEC1977
Applications Developer II 120796 01MAR1983
Applications Developer IV 120802 01JAN1978

Partial PROC PRINT Output

Last Employee Hired within Each Job Title

Emp_Hire_
Job_Title Employee_ID Date

Account Manager 120746 01APR2002

Accountant I 120761 01JUL2006
Accountant II 120754 01MAY2006
Accountant III 120755 01AUG1983
Administration Manager 121000 01DEC1993
Applications Developer I 120801 01JUL1999
Applications Developer II 120812 01AUG2001
Applications Developer IV 120794 01JUL2003

2.6 Macro Functions

Objectives
 Use macro language functions to manipulate text.
 Use macro language functions to perform arithmetic
operations.
 Use macro language functions to execute SAS
functions.

116

Business Scenario
The production program below requires manual entry
of the year and data set name. How do you eliminate
manual data entry?

hardcoded
data orders;
set orion.Orders;
where year(Order_Date)=2011;
Lag=Delivery_Date-Order_Date;
run;

proc means data=orders maxdec=2 min max mean;

class Order_Type;
var Lag;
title "Report based on ORDERS data";
run;
hardcoded

117
117

Task Requirements
Extract the necessary information from automatic macro
variables.

&SYSDATE9=23JAN2011
data orders;
set orion.Orders;
where year(Order_Date)=2011;
Lag=Delivery_Date-Order_Date;
run;

proc means data=orders maxdec=2 min max mean;

class Order_Type;
var Lag;
title "Report based on ORDERS data";
run;
&SYSLAST=WORK.ORDERS

118
118

2.09 Short Answer Poll

In a DATA step, what function would you use to extract
the year portion of a character date?
data new;
date='23JAN2011';
year=???;
run;

119

Macro Functions
Macro functions manipulate macro variables.

SAS function
substr(date,6)

Macro function
%substr(&sysdate9,6)

 Macro functions are executed by the macro

processor.

121
121

Macro Functions
Selected character string manipulation functions:
%UPCASE Translates letters from lowercase to uppercase.
%SUBSTR Extracts a substring from a character string.
%SCAN Extracts a word from a character string.
%INDEX Searches a character string for specified text.

Other macro functions:

%EVAL Performs arithmetic and logical operations.
%SYSFUNC Executes SAS functions.
%STR Masks special characters.
%NRSTR Masks special characters, including macro triggers.

122

%SUBSTR Function
Use the %SUBSTR function to extract the year portion
of the &SYSDATE9 macro variable.
17 %put &=sysdate9;
SYSDATE9=20APR2011
18 %put YEAR=%substr(&sysdate9,6);
YEAR=2011
%SUBSTR(argument, position <,n>)

 The %SUBSTR function returns the portion

of argument beginning at position for a length
of n characters.
 When n is not supplied, the %SUBSTR function
returns the portion of argument beginning
at position to the end of argument.
123

 The values of position and n can also be the result of an arithmetic expression that yields
an integer. For example,
%substr(&var,%length(&var)-1)
returns the last two characters of the value of the macro variable var.

Macro Functions
Arguments to macro string manipulation functions can
be any text or macro triggers, or both, including:

Constant text %put %upcase(angel); ANGEL

Macro variable %let name=angel;
ANGEL
references %put %upcase(&name);
Macro functions %put %upcase(%substr(&name,1,1)); A
Macro calls %put %upcase(%mymacro); unknown

Constant text arguments do not require quotation marks.

124

2.10 Multiple Choice Poll

What is the value of x after %LET statement execution?
a. A
b. B
c. C
d. D

%let x=%substr("ABCD",2,1);

125

%SCAN Function
To extract the data set name from the &SYSLAST macro
variable, use the %SCAN macro function.
SAS Log
26 %put &=syslast;
SYSLAST=WORK.ORDERS
27 %put DSN=%scan(&syslast,2,.);
DSN=ORDERS
%SCAN(argument, n <,delimiters>)

The %SCAN function does the following:

 returns the nth word of argument, where words
are strings of characters separated by delimiters
 uses a default set of delimiters if none are specified
 returns a null string if there are fewer than n words
in argument
127

 The following are default delimiters for the %SCAN function: blank . ( & ! $ * ) ; - / , %
It is not necessary to place argument and delimiters in quotation marks because they are always handled
as character strings by the %SCAN function.

Business Scenario Solution

Step 1 Write and debug a program with hardcoded
constants.
data orders;
set orion.Orders;
where year(Order_Date)=2011;
Lag=Delivery_Date - Order_Date;
run;
proc means data=orders maxdec=2 min max mean;
class Order_Type;
var Lag;
title "Report based on ORDERS data";
run;

m102d05a
128

Business Scenario Solution

Step 2 Use macro functions to extract the necessary
information from automatic macro variables.
data orders;
set orion.Orders;
where year(Order_Date)=%substr(&sysdate9,6);
Lag=Delivery_Date-Order_Date;
run;

proc means data=&syslast maxdec=2 min max mean;

class Order_Type;
var Lag;
title "Report based on %scan(&syslast,2,.) data";
run;
Report based on ORDERS data

Analysis Variable : Lag

Order N
Type Obs Minimum Maximum Mean
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
1 70 0.00 10.00 0.43
2 15 2.00 10.00 4.27
3 31 1.00 7.00 3.87
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ m102d05b
129

Business Scenario
Additional macro functions are available to automate
the generation of SAS code.

%EVAL
%SYSFUNC
%STR
%NRSTR

130
130

%EVAL Function
The %EVAL function performs arithmetic and logical
operations.
28 %put x=2+2;
x=2+2
29 %put x=%eval(2+2);
x=4 %EVAL(expression)

The %EVAL function does the following:

 truncates non-integer results
 returns a text result
 returns 1 (true) or 0 (false) for logical operations
 returns a null value and error message when
non-integer values are used in arithmetic operations
131

%EVAL Function
Example: Compute the first year of a range based
on the current date.
%let thisyr=%substr(&sysdate9,6);
%let lastyr=%eval(&thisyr-1);
proc means data=orion.order_fact maxdec=2 min max mean;
class order_type;
var total_retail_price;
where year(order_date) between &lastyr and &thisyr;
title1 "Orders for &lastyr and &thisyr";
title2 "(as of &sysdate9)";
run;
Orders for 2010 and 2011
(as of 31DEC2011)

The MEANS Procedure

Analysis Variable : Total Retail Price for This Product

Order N
Type Obs Minimum Maximum Mean
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

1 174 3.20 1136.20 126.74

2 62 6.20 1937.20 212.88
3 55 9.60 702.00 172.10
132 ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ m102d06

%SYSFUNC Function
The %SYSFUNC macro function executes SAS functions.
31 %put &=syslast;
SYSLAST=WORK.ORDERS
32 %put DSN=%sysfunc(propcase(&syslast));
DSN=Work.Orders
%SYSFUNC(SAS function(argument(s)) <,format>)

 SAS function(argument(s)) is the name of

a SAS function and its corresponding arguments.
 The second argument is an optional format for
the value returned by the first argument.

133

Most SAS functions can be used with %SYSFUNC. Exceptions include array processing (DIM, HBOUND,
LBOUND), variable information (VNAME, VLABEL, MISSING), macro interface (RESOLVE, SYMGET),
and data conversion functions (INPUT, PUT), as well as the IORCMSG, LAG, and DIF functions.
Because %SYSFUNC is a macro function, do not enclose character arguments in quotation marks as you
do with SAS functions. Use commas to separate all arguments in SAS functions within %SYSFUNC.
You cannot use argument lists preceded by the word OF.

%SYSFUNC executes one SAS function at a time. If you want to execute multiple SAS functions
in a single macro statement or expression, use a separate %SYSFUNC for each SAS function, as shown
below.
%put %sysfunc(year(%sysfunc(today())));

%SYSFUNC Function
The automatic macro variables SYSDATE9 and SYSTIME
can be used in titles.
title1 "&sysdate9";
title2 "&systime";

07MAR2011
13:39

SYSDATE9 and SYSTIME represent the date and time

that the SAS session began.

m102d07
134

%SYSFUNC Function
Example: Generate titles that contain the current date
and time, appropriately formatted.
title1 "%sysfunc(today(),weekdate.)";
title2 "%sysfunc(time(),timeAMPM8.)";

Monday, March 7, 2011

1:39 PM

m102d07
135

2.11 Short Answer Poll

What is the problem with this attempt to store a complete
SAS statement in a macro variable?
%let statement=title "S&P 500";

m102d08
136

%STR Function
Example: Store a SAS statement in a macro variable.
%let statement=%str(title "S&P 500";);
%STR(argument)

m102d08
138

%STR Function
The %STR function masks (removes the normal meaning of)
these special tokens:

Special + - * / , < > = ; ' "

Characters ~ ^ ( ) # blank
Mnemonics LT EQ GT LE GE NE AND
OR NOT IN

139
139

The %STR function does not mask & and % characters.

%STR Function
The following are true for the %STR function:
 masks tokens, so the macro processor does not
interpret them as macro-level syntax
 enables macro triggers to work normally
 preserves leading and trailing blanks in its argument
 masks an unpaired quotation mark or parenthesis
in its argument when the quotation mark or
parenthesis is immediately preceded by a percent
sign (%)

140

%STR Function
The %STR function enables macro triggers to work normally.

SAS Log
3 %let statement=%str(title "S&P 500";);
WARNING: Apparent symbolic reference P not resolved.

m102d08
141

%NRSTR Function
The %NRSTR function works the same as the %STR
function, except it also masks macro triggers.

Example: Use %NRSTR to prevent attempted macro

variable resolution.

%let statement=%nrstr(title "S&P 500";);

%NRSTR(argument)

m102d08
142

In addition to %STR and %NRSTR, several other macro quoting functions are available for specialized
purposes. For further information, see “Macro Quoting” under “Understanding and Using the Macro
Facility” under SAS Macro Language: Reference in Base SAS documentation.

2.12 Multiple Answer Poll

You can use macro functions to do which of the following:
a. manipulate text
b. execute SAS functions
c. manipulate SAS data set variables
d. perform arithmetic calculations

143

Exercises
Level 1
11. Using the %SUBSTR and %SCAN Functions
a. Submit a %LET statement to assign the value Anthony Miller to a macro variable named fullname.
b. Extract the first initial and last name, putting them together into a new macro variable as
A. Miller. Use the %PUT statement to display the results.
12. Using the %SYSFUNC Function
Use the %PUT statement and the %SYSFUNC function to display the current date and time. Format
the date with the MMDDYYP10. format and the time with the TIMEAMPM. format.

Level 2
13. Using Macro Functions
a. Open the m102e13 program shown below into the Editor window. Submit the program
and examine the output that it creates. Verify the titles.
%let d=&sysdate9;
%let t=&systime;
proc print data=orion.product_dim;
where Product_Name contains "Jacket";
var Product_Name Product_ID Supplier_Name;
title1 "Product Names Containing 'Jacket'";
title2 "Report produced &t &d";
run;
b. Submit a %LET statement to assign the value R&D to a macro variable named product. Reference
the new macro variable in the WHERE statement and the TITLE1 statement. Submit the modified
program.
PROC PRINT Output
Product Names Containing 'R&D'
Report produced 14:06 17FEB2009

Obs Product_Name Product_ID Supplier_Name

393 Top Men's R&D Ultimate Jacket 240300300070 Top Sports Inc
395 Top R&D Long Jacket 240300300090 Top Sports Inc

Challenge
14. Manipulating Macro Variables
a. Assign your birthdate to a macro variable.
b. Issue a %PUT statement that writes the day of the week that you were born.

2.7 Solutions
Solutions to Exercises
1. Displaying Automatic Macro Variables
a. The _AUTOMATIC_ argument in the %PUT statement displays the values of all automatic macro
variables in the SAS log. Many of the values shown are dependent on the host system.
%put _automatic_;
b. The value of SYSLAST is _NULL_ unless a data set has been created. Then it is the name
of the last created data set. The value of SYSUSERID is dependent on the host system. The value
of SYSTIME and SYSDATE9 is the initialization time of your SAS session.
c. The values of SYSTIME and SYSDATE do not reflect the current time and date. Instead, they
reflect the initialization time of the SAS session. Therefore, the values are not always accurate.
2. Using Automatic Macro Variables
a. Using the SORT procedure, sort the data set orion.continent by Continent_Name.
proc sort data=orion.continent out=sorted;
by Continent_Name;
run;
b. The SYSLAST automatic macro variable contains the name of the most recently created data set.
proc print data=&syslast;
title "&syslast";
run;
c. Submit the program and examine the results.
SAS Output
WORK.SORTED

Continent_
Obs ID Continent_Name

1 94 Africa
2 95 Asia
3 96 Australia/Pacific
4 93 Europe
5 91 North America

3. Using Automatic Macro Variables

a. The DATA step alters the value of SYSLAST to WORK.NEW.
data new;
set orion.continent;
run;
b. The value of SYSLAST is still WORK.NEW. The PRINT procedure does not create a SAS data
set. Therefore, it does not alter the value of SYSLAST.
proc print data=orion.continent;

run;
4. Using SAS Date Constants
a. Open the program into the Editor window.
b. Modify the program so that it subsets the data to return only the employees hired between
January 1, 2007, and today. Use the automatic macro variable SYSDATE9 to return today’s date.
proc print data=orion.employee_payroll;
format Birth_Date Employee_Hire_Date date9.;
where Employee_Hire_Date between '01jan2007'd and
"&sysdate"d;
run;
5. Defining and Using Macro Variables for Character Substitution
a. Submit the program below.
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "Gold";
title "Gold Customers";
run;

Partial SAS Output

Gold Customers

Customer_ Customer_
Obs Customer_Name Gender Age

2 Sandrina Stephano F 28
3 Cornelia Krahl F 33
7 Markus Sepke M 19
11 Oliver S. Füßling M 43

b. The macro variable type should contain the text string Gold without any surrounding quotation
marks. To resolve the macro variable in the WHERE and TITLE statements, change the single
quotation marks to double quotation marks. It produces the same output as before.
%let type=Gold;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
c. Turn on the SYMBOLGEN option.
options symbolgen;
%let type=Gold;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;

Partial SAS Log

79 proc print data=orion.customer_dim;
80 var customer_name customer_gender customer_age;
81 where customer_group contains "&type";
SYMBOLGEN: Macro variable TYPE resolves to Gold
SYMBOLGEN: Macro variable TYPE resolves to Gold
82 title "&type Customers";
83 run;

d. Modify the value of type to Internet.

%let type=Internet;
proc print data=orion.customer_dim;
var Customer_name Customer_Gender Customer_age;
where Customer_Group contains "&type";
title "&type Customers";
run;
Partial SAS Log
63 %let type=Internet;
64 proc print data=orion.customer_dim;
65 var customer_name customer_gender customer_age;
66 where customer_group contains "&type";
SYMBOLGEN: Macro variable TYPE resolves to Internet
SYMBOLGEN: Macro variable TYPE resolves to Internet
67 title "&type Customers";
68 run;

NOTE: There were 8 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Internet';
NOTE: PROCEDURE PRINT used (Total process time):
real time 5.04 seconds
cpu time 0.01 seconds

e. Turn off the SYMBOLGEN option.

options nosymbolgen;
6. Defining and Using Macro Variables for Numeric Substitution
a. Edit the program to also subset the data for customers between the ages of 30 and 45.
%let type=Gold;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type" and
Customer_Age between 30 and 45;
title "&type Customers between 30 and 45";
run;
b. Modify the program so that the values 30 and 45 are replaced by references to the macro variables
age1 and age2, respectively.
%let type=Gold;
%let age1=30;
%let age2=45;
proc print data=orion.customer_dim;

var Customer_Name Customer_Gender Customer_Ag e;

where Customer_Group contains "&type" and
Customer_Age between &age1 and &age2 ;
title "&type Customers between &age1 and &age2";
run;
c. Include the appropriate system option to display resolved values of macro variables
in the SAS log. Resubmit the program and examine the log.
options symbolgen;
d. Modify the value of age1 and age2. Resubmit the program and examine the log.
Partial SAS Log
88 %let type=Gold;
89 %let age1=20;
90 %let age2=30;
91 proc print data=orion.customer_dim;
92 var customer_name customer_gender customer_age;
93 where customer_group contains "&type" and Customer_Age between &age1 and &age2 ;
SYMBOLGEN: Macro variable TYPE resolves to Gold
SYMBOLGEN: Macro variable AGE1 resolves to 20
SYMBOLGEN: Macro variable AGE2 resolves to 30
SYMBOLGEN: Macro variable TYPE resolves to Gold
SYMBOLGEN: Macro variable AGE1 resolves to 20
SYMBOLGEN: Macro variable AGE2 resolves to 30
94 title "&type Customers between &age1 and &age2";
95 run;

NOTE: There were 7 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Gold' and (Customer_Age>=20 and Customer_Age<=30);

e. Turn off the SYMBOLGEN option.

options nosymbolgen;
7. Deleting Macro Variables
a. Open the program and submit the code. %PUT _USER_ lists all user-defined macro variables
in the SAS log.
%put _user_;
b. The %SYMDEL statement deletes user-defined macro variables.
%symdel type age1 age2 pet1 pet2;

 Your solution could be different. It depends on the macro variables created during
the current SAS session.
c. Use the %PUT statement to verify that the macro variable deletion is successful.
%put _user_;
8. Consecutive Macro Variable References
a. Open the program into the Editor window.
b. Modify the program so that the two occurrences of AUG and 2006 are replaced by references
to the macro variables month and year, respectively.

%let month=AUG;
%let year=2006;
proc print data=orion.organization_dim;
where Employee_Hire_Date="01&month&year"d;
id Employee_ID;
var Employee_Name Employee_Country Employee_Hire_Date;
title "Personal Information for Employees Hired in "
"&month &year";
run;
c. Modify the value of month to JUL and year to 2003. Resubmit the program.
%let month=JUL;
%let year=2003;
proc print data=orion.organization_dim;
where Employee_Hire_Date="01&month&year"d;
id Employee_ID;
var Employee_Name Employee_Country Employee_Hire_Date;
title "Personal Information for Employees Hired in "
"&month &year";
run;
9. Macro Variable References with Delimiters
a. Open the program into the Editor window.
b. Modify the program so that all occurrences of Organization and Employee are replaced with macro
variable references called dsn and var, respectively. Submit the program and verify the output.
%let dsn=Organization;
%let var=Employee;
proc print data=orion.&dsn._dim;
id &var._ID;
var &var._Name &var._Country &var._Gender;
title "Listing of All &var.s From Orion.&dsn ._Dim";
run;
c. Modify the value of dsn to Customer and var to Customer. Resubmit the program
%let dsn=Customer;
%let var=Customer;
proc print data=orion.&dsn._dim;
id &var._ID;
var &var._Name &var._Country &var._Gender;
title "Listing of All &var.s From Orion.&dsn._Dim";
run;
10. Macro Variable References with Multiple Delimiters
a. Open the program into the Editor window.
b. Using a macro variable, modify the program to return the employees with the most or least amount
of seniority.
proc sort data=orion.staff out=staffhires;
by Job_Title Emp_Hire_Date;

run;

%let seniority=First; /* Employees with most seniority */

/* %let seniority=Last; Employees with least seniority */

data &seniority.Hired;
set staffhires;
by Job_Title;
if &seniority..Job_Title;
run;
proc print data=&seniority.Hired;
id Job_Title;
var Employee_ID Emp_Hire_Date;
title "&seniority Employee Hired within Each Job Title";
run;
11. Using the %SUBSTR and %SCAN Functions
a. Submit a %LET statement to assign the value Anthony Miller to a macro variable named fullname.
%let fullname=Anthony Miller;
b. Extract the first initial and last name, putting them together into a new variable as A. Miller.
%let newname=%substr(&fullname,1,1). %scan(&fullname,2);
%put &newname;
Alternate Solution
%let initial=%substr(&fullname,1,1);
%let last=%scan(&fullname,2);
%let newname=&initial.. &last;
%put &newname;
12. Using the %SYSFUNC Function
Use the %PUT statement and the %SYSFUNC function to display the current date and time. Format
the date with the MMDDYYP10. format and the time with the TIMEAMPM. format.
%put Today is %sysfunc(today(), mmddyyp10.) and time is
%sysfunc(time(), timeampm.);
13. Using Macro Functions
a. Open the program into the Editor window.
b. The %NRSTR function is required to prevent macro variable resolution.
%let d=&sysdate9;
%let t=&systime;
%let product=%nrstr(R&D);
proc print data=orion.product_dim;
where Product_Name contains "&product";
var Product_Name Product_ID Supplier_Name;
title1 "Product Names Containing '&product'";
title2 "Report produced &t &d";
run;

14. Manipulating Macro Variables

a. Assign your birthdate to a macro variable.
%let birthdate=06jun1982;
b. Issue a %PUT statement that writes the day of the week that you were born.
%put %sysfunc(putn("&birthdate"d,downame.));

Solutions to Student Activities (Polls/Quizzes)

2.01 Short Answer Poll – Correct Answer

What are the two types of macro variables?
automatic and user-defined

2.02 Poll – Correct Answer

Submit the statement below.
%put _automatic_;

Does SYSTIME match the current time?

 Yes
 No

SYSTIME represents the time of SAS invocation.

It does not necessarily match the current time.

2.03 Short Answer Poll – Correct Answer

SAS Output
Customer Country

Country Frequency Percent

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
AU 8 10.39
CA 15 19.48
DE 10 12.99
IL 5 6.49
TR 7 9.09
US 28 36.36
ZA 4 5.19

Created &systime &sysday, &sysdate9

The word scanner does not tokenize literals enclosed

in single quotation marks, so macro variables do not
resolve.

m102d01
38

2.04 Multiple Choice Poll – Correct Answer

Macro variable references are resolved by which
of the following?
a. SAS compiler
b. macro processor
c. word scanner

2.05 Short Answer Poll – Correct Answer

What is the value of the macro variable z?

%let x=15;
%let y=10;
%let z=&x-&y;

The value of z is 15-10.

2.06 Multiple Choice Poll – Correct Answer

Which WHERE statement is correct?
a. where City=&office;
b. where City='&office';
c. where City="&office";

2.07 Short Answer Poll – Correct Answer

What is the problem with this reference?

Partial SAS Log

1 %let graphics=g;
2 proc &graphicschart data=orion.y&year&month;
-
10
WARNING: Apparent symbolic reference GRAPHICSCHART not resolved.

ERROR 10-205: Expecting the name of the procedure to be executed.

SAS interprets the macro variable’s name as

GRAPHICSCHART because no delimiter separates
the macro variable reference from the trailing text.

107

2.08 Short Answer Poll – Correct Answer

Modify the previous example to include a macro variable
that stores a libref.
libname orion "&path";
proc gchart data=oriony2010jan;
hbar week / sumvar=sale;
run;
proc gplot data=oriony2010jan;
plot sale*day;
run;

missing period
What is the problem this time?
The period after &LIB is interpreted as a delimiter,
so the period does not appear as text.

111

2.09 Short Answer Poll – Correct Answer

In a DATA step, what function would you use to extract
the year portion of a character date?
data new;
date='23JAN2011';
year=substr(date,6);
run;

The SUBSTR function extracts selected characters

from a character string.

120

2.10 Multiple Choice Poll – Correct Answer

What is the value of x after %LET statement execution?
a. A
b. B
c. C
d. D

%let x=%substr("ABCD",2,1);

126

2.11 Short Answer Poll – Correct Answer

What is the problem with this attempt to store a complete
SAS statement in a macro variable?
The semicolon
%let statement=title "S&P 500"; ends the %LET
statement.

The stored value lacks a semicolon.

title "S&P 500" missing semicolon

m102d08
137

2.12 Multiple Answer Poll – Correct Answers

You can use macro functions to do which of the following:
a. manipulate text
b. execute SAS functions
c. manipulate SAS data set variables
d. perform arithmetic calculations

144

3.1 Defining and Calling a Macro .................................................................................... 3-3

Exercises............................................................................................................. 3-28

3.2 Macro Parameters ................................................................................................... 3-30

Demonstration: Macros with Positional Parameters.................................................... 3-35
Demonstration: Macros with Keyword Parameters ..................................................... 3-38

Demonstration: Macros with Mixed Parameter Lists ................................................... 3-42

Exercises............................................................................................................. 3-44

3.3 Solutions ................................................................................................................. 3-46

Solutions to Exercises ........................................................................................... 3-46
Solutions to Student Activities (Polls/Quizzes) ........................................................... 3-53
3-2 Chapter 3 Macro Definitions

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-3

3.1 Defining and Calling a Macro

Objectives
 Define the purpose of a macro definition.
 Specify the steps for creating and storing a macro.
 Identify the steps for calling a macro.
 Investigate the program flow of a macro call.

What Is a Macro Definition?

Like macro variables, macros generate text. However,
macros can contain programming logic to dynamically
control what text is generated and when it is generated.
Macros can also accept parameters.

A macro or macro definition can store the following:

 macro language statements or expressions
 complete or partial SAS statements
 complete or partial SAS steps
 any text
 any combination of the above

Defining a Macro
This code defines the Time macro, which displays
the current time.
%macro time;
%put The current time is %sysfunc
(time(),timeampm.).;
%mend time;
%MACRO macro-name;
macro-text
%MEND <macro-name>;

 Macro names follow SAS naming conventions

and cannot be reserved names such as names
of macro statements or functions (LET and SCAN,
for example).
m103d01
5

Creating and Storing a Macro

Creating a macro requires the following steps:

Step 1 Create the macro definition.

%macro time;
%put The current time is %sysfunc
(time(),timeampm.).;
%mend time;

Step 2 Submit the macro definition.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-5

Creating and Storing a Macro

Submitting a macro definition causes the following actions:

Macro language statements or expressions, if any, are

 checked for syntax errors
 compiled.
SAS statements and other text are
 not checked for syntax errors
 not compiled.

The compiled macro is stored, by default,

in the temporary catalog work.sasmacr,
7 with an entry type of MACRO.

Macro Compilation
To verify macro compilation, specify
the MCOMPILENOTE=ALL option.
OPTIONS MCOMPILENOTE=ALL|NONE;

options mcompilenote=all;
%macro time;
%put The current time is %sysfunc
(time(),timeampm.).;
%mend time;
SAS Log
1 options mcompilenote=all;
2 %macro time;
3 %put The current time is %sysfunc
4 (time(),timeampm.).;
5 %mend time;
NOTE: The macro TIME completed compilation without errors.
3 instructions 76 bytes.

8
 The default value is MCOMPILENOTE=NONE. m103d01

Calling a Macro
To call a macro, precede the macro name with
a percent sign (%).
%time
%macro-name

SAS Log
178 %time
The current time is 2:49:39 PM.

A macro call
 can appear anywhere (similar to a macro variable
reference)
 represents a macro trigger
 is passed to the macro processor
 is not a statement (no semicolon required)
m103d01
9  causes the macro to execute.

3.01 Poll
Does the macro call below require a semicolon?

%Time

 Yes
 No

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-7

Macro Storage
To display a list of compiled macros in a SAS catalog,
use the CONTENTS statement in PROC CATALOG.
proc catalog cat=work.sasmacr;
contents;
title "My Temporary Macros";
quit;

PROC CATALOG Output

My Temporary Macros

Contents of Catalog WORK.SASMACR

# Name Type Create Date Modified Date Description

----------------------------------------------------------------
1 TIME MACRO 11JUN2010:15:55:59 11JUN2010:15:55:59

m103d02
12

Permanent Macro Storage

Storing a macro permanently requires two steps.
Step 1 Set the appropriate system options.
options mstored sasmstore=orion;
MSTORED enables storage of compiled macros
in a permanent library.
SASMSTORE= designates a permanent library to store
compiled macros.
Step 2 Specify the STORE and SOURCE options
in the %MACRO statement.
%macro time / store source;

 The SOURCE option stores the macro source code

13
with the compiled code.

The following code copies macro source code saved with the SOURCE option to the SAS log:
%copy time / source;

Permanent Macro Storage

Store the Time macro permanently in the orion library.
libname orion "&path";
options mstored sasmstore=orion;
%macro time / store source;
%put The current time is %sysfunc
(time(),timeampm.).;
%mend time;

Call the Time macro in a new SAS session.

libname orion "&path";
options mstored sasmstore=orion;
%time

m103d03
14

 The macro processor searches the work.sasmacr catalog first, followed by the SASMSTORE=
catalog.

Business Scenario
To minimize typing, create the Calc macro below.
%macro calc;
proc means data=orion.order_fact;
run;
%mend calc;

The Calc macro

 generates complete SAS statements
and a complete SAS step
 contains no macro language statements.

m103d04a
15

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-9

General Program Flow

When the macro processor receives a macro call, it takes
the following actions.

%calc

Input Stack

Searches the Executes compiled Sends SAS code and

macro catalog for macro language other text to the input
CALC.MACRO statements, if any stack for word scanning

16
continued...

General Program Flow

Pause Resume
Pauses while the word Resumes execution
scanner tokenizes of macro language
inserted text, and SAS statements, if any
code, if any, compiles
and executes

Defining and Calling a Macro

SAS Log
1293 %macro calc;
1294 proc means data=orion.order_fact;
1295 run;
1296 %mend calc;
NOTE: The macro CALC completed compilation without errors.
5 instructions 124 bytes.
1297
1298 %calc

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set
ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.02 seconds
cpu time 0.03 seconds

 PROC MEANS source code does not appear

in the SAS log.
m103d04a
18

Defining and Calling a Macro

To view the text generated by macro execution, specify
the MPRINT option.

SAS Log
OPTIONS MPRINT;
1300 options mprint;
1301 %calc
MPRINT(CALC): proc means data=orion.order_fact;
MPRINT(CALC): run;

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set
ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.02 seconds
cpu time 0.03 seconds

 The default setting is NOMPRINT.

m103d04a
19

 Macro-generated code is treated as a series of tokens. The MPRINT option writes each statement
to a new line without indention.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-11

Business Scenario
For greater flexibility, revise the Calc macro as shown below.

%macro calc;
proc means data=
%mend calc;

 This Calc macro generates a partial

SAS statement and a partial SAS step.

m103d04b
20

Defining and Calling a Macro

Call the revised Calc macro, followed by additional code
to complete the PROC MEANS statement and step.

%macro calc;
proc means data=
%mend calc;

%calc
orion.order_fact;
run;

m103d04b
21

Defining and Calling a Macro

SAS Log
1193 options mprint;
1194 %calc
MPRINT(CALC): proc means data=
1195 orion.order_fact;
1196 run;

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set
ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.04 seconds
cpu time 0.06 seconds

m103d04b
22

Defining and Calling a Macro

Add a semicolon after the macro call.
%macro calc;
proc means data=
%mend calc;

%calc;
orion.order_fact;
run;

A semicolon following a macro call is not

Detail Program Flow

Add a semicolon after the macro call.

Compiler

Word Scanner Macro Processor

Input Stack work.sasmacr

%calc; # Name Type
orion.order_fact; 1 CALC MACRO
run;
2 TIME MACRO

24 ...

Detail Program Flow

The macro call is passed to the macro processor
as usual. The semicolon remains in the input stack.

Compiler

Word Scanner Macro Processor

%calc
Input Stack work.sasmacr
; # Name Type
orion.order_fact; 1 CALC MACRO
run;
2 TIME MACRO

25 ...

Detail Program Flow

The macro processor opens CALC.MACRO.

Compiler

Word Scanner Macro Processor

Input Stack CALC.MACRO

; %macro calc;
orion.order_fact; proc means data=
run; %mend calc;

26 ...

Detail Program Flow

Macro text is placed on the input stack on top
of the semicolon.

Compiler

Word Scanner Macro Processor

Input Stack CALC.MACRO

proc means data= %macro calc;
; proc means data=
orion.order_fact; %mend calc;
run;

27 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-15

Detail Program Flow

The PROC MEANS statement is passed to the compiler,
followed by the semicolon, generating a syntax error.

Compiler
proc means data=;

Word Scanner Macro Processor

Input Stack CALC.MACRO

orion.order_fact; %macro calc;
run; proc means data=
%mend calc;

28 ...

Detail Program Flow

A second syntax error is generated by the subsequent text.

Compiler
proc means data=;

Word Scanner Macro Processor

Input Stack CALC.MACRO

orion.order_fact; %macro calc;
run; proc means data=
%mend calc;

29 ...

Defining and Calling a Macro

SAS Log
1204 %calc;
-
22
ERROR 22-322: Expecting a name.

MPRINT(CALC): proc means data=

1205 orion.order_fact;
ERROR: File WORK.NAME.DATA does not exist.
1205 orion.order_fact;
----------------
180
ERROR 180-322: Statement is not valid or it is used out of proper order.

1206 run;

NOTE: The SAS System stopped processing this step because of errors.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

m103d04b
30

Business Scenario
Revise the Calc macro again to enable easy and flexible
selection of the data set name and variable name(s).
%macro calc;
proc means data= ;
var ;
run;
%mend calc;

m103d04c
31

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-17

Defining the Macro

Use macro variable references within a macro definition
for increased flexibility.
%macro calc;
proc means data=&dsn;
var &vars;
run;
%mend calc;

m103d04c
32

 Macro variable references within a macro definition resolve during macro execution,
not compilation.

Calling the Macro

Calling this macro involves two steps.

Step 1 Precede the call with %LET statements

to populate macro variables referenced
within the macro.
%let dsn=orion.order_fact;
%let vars=quantity;

Step 2 Precede the macro name with a percent

sign (%) to call the macro.
%let dsn=orion.order_fact;
%let vars=quantity;
%calc
m103d04c
33

 Placing a semicolon after a macro call might insert an inappropriate semicolon into the resulting
program, leading to errors during compilation or execution.

Detail Program Flow

Example: Create macro variables and call the Calc macro.

Compiler Symbol Table

Word Scanner Macro Processor

Input Stack work.sasmacr

%let dsn=orion.order_fact; # Name Type
%let vars=quantity; 1 CALC MACRO
%calc 2 TIME MACRO

34 ...

Detail Program Flow

The macro processor executes the %LET statements
and populates the symbol table.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack work.sasmacr

%calc # Name Type
1 CALC MACRO
2 TIME MACRO

35 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-19

Detail Program Flow

When the macro processor receives %Calc, it locates
CALC.MACRO within the work.sasmacr catalog.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
%calc
Input Stack work.sasmacr
# Name Type
1 CALC MACRO
2 TIME MACRO

36 ...

Detail Program Flow

The macro processor opens CALC.MACRO. There
are no macro language statements to execute.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

%macro calc;
proc means data=&dsn;
var &vars;
run;
%mend calc;
37 ...

Detail Program Flow

The macro processor places the macro text
on the input stack.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

proc means data=&dsn; %macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
38 ...

Detail Program Flow

Macro activity pauses while the word scanner tokenizes
SAS code and passes tokens to the compiler.

Compiler Symbol Table

proc means data= DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

&dsn; %macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
39 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-21

Detail Program Flow

Macro variable references are passed to the macro
processor.

Compiler Symbol Table

proc means data= DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
&dsn
Input Stack CALC.MACRO
; %macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
40 ...

Detail Program Flow

Symbolic substitution is performed.

Compiler Symbol Table

proc means data= DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
&stats
Input Stack CALC.MACRO
orion.order_fact; %macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
41 ...

Detail Program Flow

The word scanner tokenizes the resolved value
and passes it to the compiler.

Compiler Symbol Table

proc means data=orion.order_fact; DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

%macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
42 ...

Detail Program Flow

Word scanning continues.

Compiler Symbol Table

proc means data=orion.order_fact;
var DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

%macro calc;
&vars; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
43 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-23

Detail Program Flow

The macro variable reference is passed to the macro
processor.

Compiler Symbol Table

proc means data=orion.order_fact;
var DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
&vars
Input Stack CALC.MACRO
%macro calc;
; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
44 ...

Detail Program Flow

Symbolic substitution is performed.

Compiler Symbol Table

proc means data=orion.order_fact;
var DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
&vars
Input Stack CALC.MACRO
%macro calc;
quantity; proc means data=&dsn;
run; var &vars;
run;
%mend calc;
45 ...

Detail Program Flow

The word scanner tokenizes the resolved value
and passes it to the compiler.

Compiler Symbol Table

proc means data=orion.order_fact;
var quantity; DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

%macro calc;
proc means data=&dsn;
run; var &vars;
run;
%mend calc;
46 ...

Detail Program Flow

When a step boundary is encountered, SAS executes
the compiled step as macro activity remains paused. Macro
activity stops when the %MEND statement is encountered.

Compiler Symbol Table

proc means data=orion.order_fact;
var quantity; DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
run;

Input Stack CALC.MACRO

%macro calc;
proc means data=&dsn;
var &vars;
run;
%mend calc;
47

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-25

Macro Execution
SAS Log
52 %let stats=min max;
53 %let vars=quantity;
54 %calc

NOTE: There were 617 observations read from the data set ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

m103d04c
48

Detail Program Flow

Add a semicolon to the macro call.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack work.sasmacr

%calc; # Name Type
1 CALC MACRO
2 TIME MACRO

50 ...

Detail Program Flow

The macro call is passed to the macro processor
as usual. The semicolon remains in the input stack.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor
%calc

Input Stack work.sasmacr

; # Name Type
1 CALC MACRO
2 TIME MACRO

51 ...

Detail Program Flow

The macro processor opens CALC.MACRO.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

; %macro calc;
proc means data=&dsn;
var &vars;
run;
%mend calc;
52 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-27

Detail Program Flow

The macro text is placed on the input stack on top
of the semicolon.

Compiler Symbol Table

DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

proc means data=&dsn; %macro calc;
var &vars; proc means data=&dsn;
run; var &vars;
; run;
%mend calc;
53 ...

Detail Program Flow

The PROC MEANS step is passed to the compiler.
An extra semicolon follows the RUN statement. The extra
semicolon is harmless, but unnecessary, in this example.

Compiler Symbol Table

proc means data=orion.order_fact;
var quantity; DSN orion.order_fact
VARS quantity
Word Scanner Macro Processor

Input Stack CALC.MACRO

run; %macro calc;
; proc means data=&dsn;
var &vars;
run;
%mend calc;
54 ...

Exercises

Level 1

1. Defining and Calling a Macro

a. Open the m103e01 program shown below into the Editor window.
proc print data=orion.customer_dim;
var Customer_Group Customer_Name Customer_Gender
Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
b. Convert the program into a macro named Customers. Set the appropriate system option
to display a note in the SAS log when a macro definition has compiled. Submit the macro
definition and examine the log.
c. Submit a %LET statement to assign the value Gold to the macro variable type. Call the macro
and examine the log.
d. Change the value of type to Internet.
e. Activate the appropriate system option to display source code received by the SAS compiler.
Call the macro again and examine the log.

Level 2

2. Storing a Macro
a. Open the m103e02 program shown below into the Editor window.
%macro tut;
king tut
%mend tut;
b. Submit the macro definition and check the SAS log.
c. In the SAS Explorer window, navigate to the sasmacr catalog within the work library to locate
the Tut macro. In SAS Enterprise Guide, use PROC CATALOG.

d. Submit a %SYSMACDELETE statement to delete the Tut macro.

 The %SYSMACDELETE statement was introduced in SAS 9.3.

Challenge

3. Calling a Macro from a TITLE Statement

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
3.1 Defining and Calling a Macro 3-29

a. Define a macro that issues the current time of day with the TIMEAMPM. format. Name
the macro Currtime. Submit the macro definition.
b. Open the m103e03 program shown below into the Editor window. Add a TITLE2 statement.
Call the Currtime macro from the TITLE2 statement. Submit the program and examine
the output.
proc print data=orion.customer_dim(obs=10);
var Customer_Name Customer_Group;
title 'Customer List';
run;

3.2 Macro Parameters

Objectives
 Define and call macros with parameters.
 Describe the difference between positional parameters
and keyword parameters.

Review
Example: Notice macro variable references within
the Calc macro.
%macro calc;
proc means data=&dsn;
var &vars;
run;
%mend calc;

m103d04c
59

Business Scenario
Simplify the Calc macro so that it can be called with one
line of code instead of three.

%let dsn=orion.order_fact;
%let vars=quantity; three lines of code
%calc

%calc(orion.order_fact,quantity) one line of code

 You can define a macro with a parameter list.

m103d04c
61

Parameter Lists
A parameter list is a list of macro variables referenced
within the macro. There are three types of parameter lists:
 positional
 keyword
 mixed

Positional Parameters
A positional parameter list defines macro parameters
in a particular order. Parameter names are supplied
when the macro is defined.

%macro calc(dsn,vars);
proc means data=&dsn;
var &vars;
run;
%mend calc;
%MACRO macro-name(parameter-1, … parameter-n);
macro text
%MEND <macro-name>;

m103d05
63

Positional Parameters
Parameter values are supplied when the macro is called.

%macro calc(dsn,vars);
proc means data=&dsn;
var &vars;
run;
%mend calc;

%calc(orion.order_fact,quantity)

%macro-name(value-1, … value-n)

 Parameter values must appear in the same order

as their corresponding parameter names.
m103d05
64

 To assign a null value to one or more positional parameters, use commas as placeholders
for the omitted values.

Local Symbol Tables

When a macro with a parameter list is called, a local
symbol table is created and populated with the macro
parameters.
%calc(orion.order_fact,quantity)

Calc Local Table

DSN orion.order_fact
VARS quantity

Macro variables in a local table exist only during macro

execution and can be referenced only within the macro.

 Local symbol tables are deleted after macro

execution.
65

3.02 Multiple Choice Poll

A %LET statement outside a macro definition creates
a macro variable in which of the following?
a. global symbol table
b. local symbol table

Positional Parameters
Example: Define and call a macro with positional
parameters.
%macro count(opts, start, stop);
proc freq data=orion.orders;
where order_date between
"&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;

%count(nocum,01jan2008,31dec2008)
%count(,01jul2008,31dec2008)

m103d06a
68

Macros with Positional Parameters

m103d06a
Use positional parameters to specify a range of dates and TABLE statement options for
the FREQ procedure.
%macro count(opts, start, stop);
proc freq data=orion.orders;
where order_date between "&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;
%count(nocum,01jan2008,31dec2008)
%count(,01jul2008,31dec2008)

 A null value is passed to the OPTS parameter in the second call.

Partial SAS Log

50 %count(nocum,01jan2008,31dec2008)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan2008"d and "31dec2008"d;
MPRINT(COUNT): table order_type / nocum;
MPRINT(COUNT): title1 "Orders from 01jan2008 to 31dec2008";
MPRINT(COUNT): run;
NOTE: There were 87 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.28 seconds
cpu time 0.11 seconds

51 %count(,01jul2008,31dec2008)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jul2008"d and "31dec2008"d;
MPRINT(COUNT): table order_type / ;
MPRINT(COUNT): title1 "Orders from 01jul2008 to 31dec2008";
MPRINT(COUNT): run;
NOTE: There were 40 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JUL2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

Business Scenario
For additional convenience, assign default values
to macro parameters.

Positional
X
parameters
Keyword
parameters

Keyword Parameters
Keyword parameters are assigned a default value after
an equal sign (=).
%macro count(opts=,start=01jan08,stop=31dec08);
proc freq data=orion.orders;
where order_date between
"&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
%MACRO macro-name(keyword=value, …, keyword=value);
macro text
%MEND <macro-name>;

m103d06b
71

Keyword Parameters
Call a macro with keyword parameters.
%macro count(opts=,start=01jan08,stop=31dec08);
proc freq data=orion.orders;
where order_date between
"&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;
%count(opts=nocum)
%count(stop=01jul08,opts=nocum nopercent)
%count()
%macro-name(k eyword=value, …, keyword=value)
Keyword parameters can appear in any order and can be
omitted from the call without placeholders. If omitted from
the call, a keyword parameter receives its default value.
m103d06b
72

 To omit every keyword parameter from a macro call, specify %macro-name().

 Specifying %macro-name without the parentheses might not execute the macro immediately.

Macros with Keyword Parameters

m103d06b
Alter the previous macro by using keyword parameters. Issue various calls to the macro.
%macro count(opts=,start=01jan08,stop=31dec08);
proc freq data=orion.orders;
where order_date between "&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;
%count(opts=nocum)
%count(stop=01jul08,opts=nocum nopercent)
%count()
Partial SAS Log
64 %count(opts=nocum)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "31dec08"d;
MPRINT(COUNT): table order_type / nocum;
MPRINT(COUNT): title1 "Orders from 01jan08 to 31dec08";
MPRINT(COUNT): run;
NOTE: There were 87 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

65 %count(stop=01jul08,opts=nocum nopercent)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "01jul08"d;
MPRINT(COUNT): table order_type / nocum nopercent;
MPRINT(COUNT): title1 "Orders from 01jan08 to 01jul08";
MPRINT(COUNT): run;
NOTE: There were 47 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='01JUL2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

66 %count()
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "31dec08"d;
MPRINT(COUNT): table order_type / ;
MPRINT(COUNT): title1 "Orders from 01jan08 to 31dec08";
MPRINT(COUNT): run;

NOTE: There were 87 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

3.03 Short Answer Poll

Submit program m103a01.
%macro dog(name=spot);
%put My dog is &name;
%mend dog;

%dog()

Edit the macro call to omit the parentheses.

%dog

Submit the macro call again.

What do you see in the SAS log?

Business Scenario
A parameter list can contain a mix of positional
and keyword parameters.

Positional Keyword Mixed

parameters parameters parameter
list

Mixed Parameter Lists

In a mixed parameter list, positional parameters are listed
before keyword parameters in both the macro definition
and the macro call.
%macro count(opts,start=01jan08,stop=31dec08);
proc freq data=orion.orders;
where order_date between
"&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;
%count(nocum)
%count(stop=30jun08,start=01apr08)
%count(nocum nopercent,stop=30jun08)
%count()

m103d06c
77

Macros with Mixed Parameter Lists

m103d06c
Alter the previous macro by using a mixed parameter list. Issue various calls to the macro.
%macro count(opts,start=01jan08,stop=31dec08);
proc freq data=orion.orders;
where order_date between "&start"d and "&stop"d;
table order_type / &opts;
title1 "Orders from &start to &stop";
run;
%mend count;
options mprint;
%count(nocum)
%count(stop=30jun08,start=01apr08)
%count(nocum nopercent,stop=30jun08)
%count()

Partial SAS Log

76 %count(nocum)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "31dec08"d;
MPRINT(COUNT): table order_type / nocum;
MPRINT(COUNT): title1 "Orders from 01jan08 to 31dec08";
MPRINT(COUNT): run;
NOTE: There were 87 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.04 seconds
cpu time 0.03 seconds

77 %count(stop=30jun08,start=01apr08)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01apr08"d and "30jun08"d;
MPRINT(COUNT): table order_type / ;
MPRINT(COUNT): title1 "Orders from 01apr08 to 30jun08";
MPRINT(COUNT): run;
NOTE: There were 28 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01APR2008'D and order_date<='30JUN2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

78 %count(nocum nopercent,stop=30jun08)
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "30jun08"d;
MPRINT(COUNT): table order_type / nocum nopercent;
MPRINT(COUNT): title1 "Orders from 01jan08 to 30jun08";
MPRINT(COUNT): run;
NOTE: There were 47 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='30JUN2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

79 %count()
MPRINT(COUNT): proc freq data=orion.orders;
MPRINT(COUNT): where order_date between "01jan08"d and "31dec08"d;
MPRINT(COUNT): table order_type / ;
MPRINT(COUNT): title1 "Orders from 01jan08 to 31dec08";
MPRINT(COUNT): run;
NOTE: There were 87 observations read from the data set ORION.ORDERS.
WHERE (order_date>='01JAN2008'D and order_date<='31DEC2008'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

Exercises

Level 1

4. Defining and Using Macro Parameters

a. Open the m103e04 program shown below into the Editor window.
%macro customers;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
%mend customers;
b. Convert this macro without parameters into a macro with one positional parameter. Name
the parameter based on macro variable references within the macro. Set the appropriate system
option to display a note in the SAS log when a macro definition has compiled. Submit the macro
definition to compile the macro.
c. Call the macro defined in the previous step with a value of Gold for the parameter.
d. Call the macro again, but with a parameter value of Catalog.
e. Change the positional parameter to a keyword parameter with a default value of Club. Submit
the revised macro definition to compile the macro.
f. Call the macro defined in the previous step with a value of Internet for the keyword parameter.
g. Call the macro again, but enable the macro to use its default parameter value.

Level 2

5. Using a Macro to Generate PROC MEANS Code

a. Open the m103e05 program shown below into the Editor window.
options nolabel;
title 'Order Stats';
proc means data=orion.order_fact maxdec=2 mean;
var total_retail_price;
class order_type;
run;
title;
b. Create a macro with keyword parameters that generalize the code so that the following attributes
are controlled by macro variables. Choose default values for all parameters so that the code
executes correctly.
 Statistics: any combination of N, NMISS, MIN, MEAN, MAX, RANGE, or a null value
 Decimal places: 0, 1, 2, 3, or 4

 Analysis variables: total_retail_price or costprice_per_unit (or both)

 Class variables: order_type or quantity (or both)
c. Execute the macro using the default parameter values.
d. Call the macro again, but override all default parameter values.
e. Call the macro again, but override only the default parameter values for statistics and decimal
places.

Challenge

6. Using Parameters That Contain Special Characters

a. Open the m103e06 program shown below into the Editor window. Submit the program to compile
the macro.
%macro specialchars(name);
proc print data=orion.employee_addresses;
where Employee_Name="&name";
var Employee_ID Street_Number Street_Name City State
Postal_Code;
title "Data for &name";
run;
%mend specialchars;
b. Execute the macro with a parameter value of Abbott, Ray.
PROC PRINT Output
Data for Abbott, Ray

Employee_ Street_ Postal_

Obs ID Number Street_Name City State Code

1 121044 2267 Edwards Mill Rd Miami-Dade FL 33135

3.3 Solutions
Solutions to Exercises
1. Defining and Calling a Macro
a. Open the program into the Editor window.
b. Convert the program into a macro named Customers. Set the MCOMPILENOTE= option
and add %MACRO and %MEND statements to create a macro definition.
options mcompilenote=all;
%macro customers;
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
%mend customers;
SAS Log
1 options mcompilenote=all;
2 %macro customers;
3 proc print data=orion.customer_dim;
4 var Customer_Name Customer_Gender Customer_Age;
5 where Customer_Group contains "&type";
6 title "&type Customers";
7 run;
8 %mend customers;
NOTE: The macro CUSTOMERS completed compilation without errors.
3 instructions 188 bytes.

c. Submit a %LET statement to assign the value Gold to the macro variable type. Call the macro
by preceding its name with a percent sign.
%let type=Gold;
%customers
SAS Log
10 %let type=Gold;
11 %customers

NOTE: There were 21 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Gold';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.67 seconds
cpu time 0.18 seconds

d. Change the value of type to Internet.

%let type=Internet;
e. Set the appropriate system option to display source code received by the SAS compiler.
Call the macro again and examine the log.

options mprint;
%customers
SAS Log
12 options mprint;
13 %customers
MPRINT(CUSTOMERS): proc print data=orion.customer_dim;
MPRINT(CUSTOMERS): var Customer_Name Customer_Gender Customer_Age;
MPRINT(CUSTOMERS): where Customer_Group contains "Internet";
MPRINT(CUSTOMERS): title "Internet Customers";
MPRINT(CUSTOMERS): run;
NOTE: There were 8 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Internet';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

2. Storing a Macro

a. Open the program into the Editor window.

b. Submit the macro definition and check the SAS log.
c. Use the SAS Explorer window or PROC CATALOG to locate the stored macro.
proc catalog cat=work.sasmacr;
contents;
quit;
d. Submit a %SYSMACDELETE statement.
%sysmacdelete tut;
3. Calling a Macro from a TITLE Statement
a. Define a macro that issues the current time of day with the TIMEAMPM. format. Name
the macro Currtime. Submit the macro definition.
%macro currtime;
%sysfunc(time(),timeAMPM.)
%mend currtime;
b. Open the m103e03 program into the Editor window. Add a TITLE2 statement. Call the macro
from the TITLE2 statement. Submit the program and examine the output.
proc print data=orion.customer_dim(obs=10);
var Customer_Name Customer_Group;
title 'Customer List';
title2 "%currtime";
run;
4. Defining and Using Macro Parameters
a. Open the program into the Editor window.
b. The macro parameter name should be TYPE because the program contains the macro references
&type. When you define positional parameters, enclose the parameter names in parentheses
following the macro name.

options mcompilenote=all;
%macro customers(type);
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
%mend customers;
c. To execute the macro, use a percent sign followed by the name of the macro. To assign a value
to a positional parameter, supply the desired value within parentheses following the macro name.
options mprint;
%customers(Gold)
SAS Log
178 %customers(Gold)
MPRINT(CUSTOMERS): proc print data=orion.customer_dim;
MPRINT(CUSTOMERS): var Customer_Name Customer_Gender Customer_Age;
MPRINT(CUSTOMERS): where Customer_Group contains "Gold";
MPRINT(CUSTOMERS): title "Gold Customers";
MPRINT(CUSTOMERS): run;
NOTE: There were 21 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Gold';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.14 seconds
cpu time 0.00 seconds

PROC PRINT Output

Gold Customers

Customer_ Customer_
Obs Customer_Name Gender Age

2 Sandrina Stephano F 28
3 Cornelia Krahl F 33
7 Markus Sepke M 19
11 Oliver S. Füßling M 43
17 Cynthia Martinez F 48
21 Alphone Greenwald M 23
24 Dianne Patchin F 28
25 Annmarie Leveille F 23
26 Gert-Gunter Mendler M 73
31 Carsten Maestrini M 63
32 James Klisurich M 38
35 Viola Folsom F 38
40 Kyndal Hooks F 43
46 Ramesh Trentholme M 58
48 Avni Umran M 28
53 Sanelisiwe Collier F 19
57 Rita Lotz F 43
58 Bill Cuddy M 21
62 Susan Krasowski F 48
64 Avinoam Tuvia M 23
75 Angel Borwick F 38

d. The macro definition does not need to be resubmitted with each macro call. The macro call does
not end with a semicolon.
%customers(Catalog)
SAS Log
179 %customers(Catalog)
MPRINT(CUSTOMERS): proc print data=orion.customer_dim;
MPRINT(CUSTOMERS): var Customer_Name Customer_Gender Customer_Age;
MPRINT(CUSTOMERS): where Customer_Group contains "Catalog";
MPRINT(CUSTOMERS): title "Catalog Customers";
MPRINT(CUSTOMERS): run;
NOTE: There were 8 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Catalog';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

PROC PRINT Output

Catalog Customers

Customer_ Customer_
Obs Customer_Name Gender Age

8 Ulrich Heyde M 68
13 Tulio Devereaux M 58
14 Robyn Klem F 48
15 Cynthia Mccluney F 38
16 Candy Kinsey F 73
20 Phenix Hill M 43
59 Avinoam Zweig M 48
67 Lauren Marx F 38

e. When you define keyword parameters, an equal sign (=) must follow the name of each parameter.
A default value for each parameter can be specified following the equal sign.
%macro customers(type=Club);
proc print data=orion.customer_dim;
var Customer_Name Customer_Gender Customer_Age;
where Customer_Group contains "&type";
title "&type Customers";
run;
%mend customers;
f. To assign a value to a keyword parameter, specify the name of the parameter followed by an equal
sign (=), followed by the desired value.
%customers(type=Internet)
SAS Log
180 %customers(type=Internet)
MPRINT(CUSTOMERS): proc print data=orion.customer_dim;
MPRINT(CUSTOMERS): var Customer_Name Customer_Gender Customer_Age;
MPRINT(CUSTOMERS): where Customer_Group contains "Internet";
MPRINT(CUSTOMERS): title "Internet Customers";
MPRINT(CUSTOMERS): run;
NOTE: There were 8 observations read from the data set ORION.CUSTOMER_DIM.

WHERE customer_group contains 'Internet';

NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

PROC PRINT Output

Internet Customers

Customer_ Customer_
Obs Customer_Name Gender Age

8 Ulrich Heyde M 68
13 Tulio Devereaux M 58
14 Robyn Klem F 48
15 Cynthia Mccluney F 38
16 Candy Kinsey F 73
20 Phenix Hill M 43
59 Avinoam Zweig M 48
67 Lauren Marx F 38

g. To request that all default parameter values be used, follow the macro call with empty
parentheses.
%customers()
SAS Log
189 %customers()
MPRINT(CUSTOMERS): proc print data=orion.customer_dim;
MPRINT(CUSTOMERS): var Customer_Name Customer_Gender Customer_Age;
MPRINT(CUSTOMERS): where Customer_Group contains "Club";
MPRINT(CUSTOMERS): title "Club Customers";
MPRINT(CUSTOMERS): run;
NOTE: There were 69 observations read from the data set ORION.CUSTOMER_DIM.
WHERE customer_group contains 'Club';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

5. Using a Macro to Generate PROC MEANS Code

a. Open the program into the Editor window.
b. Create a macro with keyword parameters that generalize the code so that the following attributes
are controlled by macro variables. Choose default values for all parameters so that the code
executes correctly.
%macro orderstats

(var=total_retail_price,class=order_type,stats=mean,decimals=2
);
options nolabel;
title 'Order Stats';
proc means data=orion.order_fact maxdec=&decimals &stats;
var &var;
class &class;
run;

title;
%mend orderstats;
c. Execute the macro using the default parameter values.
%orderstats()
SAS Log
85 %orderstats()
MPRINT(ORDERSTATS): options nolabel;
MPRINT(ORDERSTATS): title 'Order Stats';
MPRINT(ORDERSTATS): proc means data=orion.order_fact maxdec=2 mean;
MPRINT(ORDERSTATS): var total_retail_price;
MPRINT(ORDERSTATS): class order_type;
MPRINT(ORDERSTATS): run;

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.04 seconds
cpu time 0.04 seconds

MPRINT(ORDERSTATS): title;

d. Call the macro again, but override all default parameter values.
%orderstats(var=costprice_per_unit, class=quantity, stats=min mean max, decimals=0)

SAS Log
86 %orderstats(var=costprice_per_unit, class=quantity, stats=min mean max, decimals=0)
MPRINT(ORDERSTATS): options nolabel;
MPRINT(ORDERSTATS): title 'Order Stats';
MPRINT(ORDERSTATS): proc means data=orion.order_fact maxdec=0 min mean max;
MPRINT(ORDERSTATS): var costprice_per_unit;
MPRINT(ORDERSTATS): class quantity;
MPRINT(ORDERSTATS): run;

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.04 seconds
cpu time 0.04 seconds

MPRINT(ORDERSTATS): title;

e. Call the macro again, but override only the default parameter values for statistics and decimal
places.

%orderstats(stats=min mean max, decimals=0)

SAS Log
87 %orderstats(stats=min mean max, decimals=0)
MPRINT(ORDERSTATS): options nolabel;
MPRINT(ORDERSTATS): title 'Order Stats';
MPRINT(ORDERSTATS): proc means data=orion.order_fact maxdec=0 min mean max;
MPRINT(ORDERSTATS): var total_retail_price;
MPRINT(ORDERSTATS): class order_type;

MPRINT(ORDERSTATS): run;

NOTE: Multiple concurrent threads will be used to summarize data.

NOTE: There were 617 observations read from the data set ORION.ORDER_FACT.
NOTE: PROCEDURE MEANS used (Total process time):
real time 0.04 seconds
cpu time 0.04 seconds

MPRINT(ORDERSTATS): title;

6. Using Parameters That Contain Special Characters

a. Open the program into the Editor window. Submit the program to compile the macro.
b. The %STR function is required to prevent a special character such as a comma from
being misinterpreted as a parameter delimiter on the macro call.
options mprint;
%specialchars(%str(Abbott, Ray))
SAS Log
63 %specialchars(%str(Abbott, Ray))
MPRINT(SPECIALCHARS): proc print data=orion.employee_addresses;
MPRINT(SPECIALCHARS): where Employee_Name="Abbott, Ray";
MPRINT(SPECIALCHARS): var Employee_ID Street_Number Street_Name City State Postal_Code;
MPRINT(SPECIALCHARS): title "Data for Abbott, Ray";
MPRINT(SPECIALCHARS): run;
NOTE: There were 1 observations read from the data set ORION.EMPLOYEE_ADDRESSES.
WHERE Employee_Name='Abbott, Ray';
NOTE: PROCEDURE PRINT used (Total process time):
real time 2.51 seconds
cpu time 0.00 seconds

Solutions to Student Activities (Polls/Quizzes)

3.01 Poll – Correct Answer

Does the macro call below require a semicolon?

%Time

 Yes
 No

A macro call is not a statement. A semicolon is not

required and can cause problems.

3.02 Multiple Choice Poll – Correct Answer

A %LET statement outside a macro definition creates
a macro variable in which of the following?
a. global symbol table
b. local symbol table

3.03 Short Answer Poll – Correct Answer

Edit the macro call to omit the parentheses.

%dog

Submit the macro call again.

What do you see in the SAS log?

SAS windowing environment: The macro call does

not execute without parentheses.

SAS Enterprise Guide: The macro call executes

without parentheses.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
Chapter 4 DATA Step and SQL
Interfaces

4.1 Creating Macro Variables in the DATA Step.............................................................. 4-3

Demonstration: SYMPUTX Routine ......................................................................... 4-10
Demonstration: SYMPUTX Routine ......................................................................... 4-13

Demonstration: SYMPUTX Routine ......................................................................... 4-15

Demonstration: Passing Values between Steps ......................................................... 4-19

Exercises............................................................................................................. 4-20

4.2 Indirect References to Macro Variables .................................................................. 4-23

Demonstration: Indirect References to Macro Variables .............................................. 4-34

Exercises............................................................................................................. 4-36

4.3 Creating Macro Variables in SQL ............................................................................ 4-39

Exercises............................................................................................................. 4-46

4.4 Solutions ................................................................................................................. 4-49

Solutions to Exercises ........................................................................................... 4-49

Solutions to Student Activities (Polls/Quizzes) ........................................................... 4-54

4-2 Chapter 4 DATA Step and SQL Interfaces

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-3

4.1 Creating Macro Variables in the DATA

Step

Objectives
 Create macro variables during DATA step execution.
 Describe the difference between the SYMPUTX
routine and the %LET statement.

Business Scenario
Automate production of the report below, with a footnote
that indicates whether there are any Internet orders.
Orders for 2-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

1 05FEB2011 1 1 $117.60
2 07FEB2011 1 2 $656.60
3 07FEB2011 1 2 $129.00
4 09FEB2011 1 2 $36.20
5 16FEB2011 1 1 $29.40
6 28FEB2011 1 5 $192.00

No Internet Orders

 Internet orders have an Order_Type value of 3.

Many applications require macro variables to store values based on data, programming logic,
or expressions.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-4 Chapter 4 DATA Step and SQL Interfaces

DATA Step Interface

%let month=2;
%let year=2011;

proc print data=orders;

title "Orders for &month-&year";
footnote "&foot";
run;
5 m104d01a

DATA Step Interface

Why is the footnote incorrect?
Orders for 2-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

1 05FEB2011 1 1 $117.60
2 07FEB2011 1 2 $656.60
3 07FEB2011 1 2 $129.00
4 09FEB2011 1 2 $36.20
5 16FEB2011 1 1 $29.40
6 28FEB2011 1 5 $192.00

Some Internet Orders

m104d01a
6

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-5

DATA Step Interface

Word scanning begins. Macro triggers are encountered.
%let month=2;
%let year=2011;
data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=&year and month(order_date)=&month;
if order_type=3 then Number+1;
if final then do; Symbol Table
put Number=; month 2
if Number=0 then do;
%let foot=No Internet Orders; year 2011
end;
else do;
%let foot=Some Internet Orders;
end;
end;
run;

7 ...

DATA Step Interface

Compiling begins. Macro variable references are resolved.

Symbol Table
month 2
year 2011

8 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-6 Chapter 4 DATA Step and SQL Interfaces

DATA Step Interface

The macro trigger is passed to the macro processor.

data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=2011 and month(order_date)=2;
if order_type=3 then Number+1;
if final then do; Symbol Table
put Number=;
month
month 22
if Number=0 then do;
%let foot=No Internet Orders; year
year
2011
2011
end; foot No Internet Orders
else do; foot No Internet Orders
%let foot=Some Internet Orders;
end;
end;
run;

9 ...

DATA Step Interface

The macro variable foot is assigned.

data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=2011 and month(order_date)=2;
if order_type=3 then Number+1;
if final then do; Symbol Table
put Number=;
month 22
if Number=0 then do; month
year 2011
year 2011
end; foot No Internet Orders
else do; foot No Internet Orders
%let foot=Some Internet Orders;
end;
end;
run;

10 ...

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-7

DATA Step Interface

The macro trigger overwrites the previous value.

data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=2011 and month(order_date)=2;
if order_type=3 then Number+1;
if final then do; Symbol Table
put Number=;
month 2
if Number=0 then do;
year 2011
end; foot Some Internet Orders
else do;
%let foot=Some Internet Orders;
end;
end;
run;

11 ...

 %LET statements execute at word-scanning time. SAS statements are sent to the compiler.

DATA Step Interface

The compilation phase is complete. Ready for execution.

data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=2011 and month(order_date)=2;
if order_type=3 then Number+1;
if final then do; Symbol Table
put Number=;
month 2
if Number=0 then do;
year 2011
end; Nothing in this DATA foot Some Internet Orders
else do; step affects the value
end; of FOOT.
end;
run;

 The value of foot remains Some Internet Orders.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-8 Chapter 4 DATA Step and SQL Interfaces

SYMPUTX Routine
The SYMPUTX routine assigns to a macro variable any
value available to the DATA step during execution time.
It can create macro variables with the following:
 static values
 dynamic (data dependent) values
 dynamic (data dependent) names

Symbol Table

DATA step variables

SYMPUTX
DATA step expressions
Character literals
13

The SYMPUTX routine was introduced in SAS®9. In prior versions of SAS, the SYMPUT routine was
available.

Completed Business Scenario

The SYMPUTX routine is a DATA step statement that assigns
a text value to a macro variable.
%let month=2;
%let year=2011;

data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=&year and month(order_date)=&month;
if order_type=3 then Number+1;
if final then do;
put Number=; CALL SYMPUTX(macro-variable, value);
if Number=0 then do;
call symputx('foot', 'No Internet Orders');
end;
else do;
call symputx('foot', 'Some Internet Orders');
end;
end;
run; fixed m acro fixed m acro no m acro
variable nam e variable value triggers
m104d01b
14

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-9

Completed Business Scenario

The footnote is correct.
Orders for 2-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

1 05FEB2011 1 1 $117.60
2 07FEB2011 1 2 $656.60
3 07FEB2011 1 2 $129.00
4 09FEB2011 1 2 $36.20
5 16FEB2011 1 1 $29.40
6 28FEB2011 1 5 $192.00

No Internet Orders

m104d01b
15

 Macro variables created by the SYMPUTX routine are available following DATA step execution
and can be referenced after a step boundary.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-10 Chapter 4 DATA Step and SQL Interfaces

SYMPUTX Routine

m104d01b
Conditionally assign a text value to a macro variable foot based on DATA step values. Reference this
macro variable later in the program.
%let month=2;
%let year=2011;
data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=&year and month(order_date)=&month;
if order_type=3 then Number+1;
if final then do;
put Number=;
if Number=0 then do;
call symputx('foot', 'No Internet Orders');
end;
else do;
call symputx('foot', 'Some Internet Orders');
end;
end;
run;

proc print data=orders;

title "Orders for &month-&year";
footnote "&foot";
run;
The value assigned to the macro variable foot is set dynamically to No Internet Orders or Some Internet
Orders, based on DATA step execution-time logic.

PROC PRINT Output

Orders for 2-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

1 05FEB2011 1 1 $117.60
2 07FEB2011 1 2 $656.60
3 07FEB2011 1 2 $129.00
4 09FEB2011 1 2 $36.20
5 16FEB2011 1 1 $29.40
6 28FEB2011 1 5 $192.00

No Internet Orders

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-11

4.01 Multiple Choice Poll

What is the value of foot after execution of the DATA
step?
a. No Internet orders
b. Some Internet orders

data _null_;
call symputx('foot','No Internet orders');
%let foot=Some Internet orders;
run;

Business Scenario
Enhance the footnote on the previous report to include
the total number of Internet orders.
Orders for 1-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

1 02JAN2011 3 2 $195.60
2 03JAN2011 1 6 $160.80
3 04JAN2011 1 2 $306.20
4 06JAN2011 3 3 $37.80
5 13JAN2011 1 2 $362.60
6 23JAN2011 1 1 $72.60
7 24JAN2011 1 2 $258.20
8 24JAN2011 1 2 $81.20
9 24JAN2011 1 3 $358.20
10 25JAN2011 3 1 $102.40
11 25JAN2011 3 1 $113.20
12 28JAN2011 3 2 $174.40
13 29JAN2011 2 1 $37.40

5 Internet Orders m104d01c

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-12 Chapter 4 DATA Step and SQL Interfaces

SYMPUTX Routine
Copy the value of a DATA step variable into a macro variable.
%let month=1;
%let year=2011;

proc print data=orders;

title "Orders for &month-&year";
footnote "&num Internet Orders";
run;

 A maximum of 32,767 characters can be assigned to a macro variable.

 Values of numeric variables are converted automatically to character
using the BEST. format, w ith a w idth up to 32 characters.
 Leading and trailing blanks are removed from both SYMPUTX arguments.
20 m104d01c

The SYMPUT routine, which remains available, has syntax and functionality that is similar to that
of the SYMPUTX routine. The SYMPUT routine does not trim leading and trailing blanks automatically
from either argument. Therefore, it is often necessary to use TRIM and LEFT functions within the
SYMPUT routine. When assigning numeric values to macro variables, the SYMPUT routine writes
numeric-to-character conversion notes to the log.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-13

SYMPUTX Routine

m104d01c
%let month=1;
%let year=2011;

proc print data=orders;

title "Orders for &month-&year";
footnote "&num Internet Orders";
run;
PROC PRINT Output
Orders for 1-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

5 Internet Orders

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-14 Chapter 4 DATA Step and SQL Interfaces

Business Scenario
Further enhance the footnotes by adding the average
Internet order amount and last Internet order date.
Orders for 1-2011

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

Average Internet Order: $125

22 Last Internet Order: 01/28/2011 m104d01d

 The data set is sorted by Order_Date.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-15

SYMPUTX Routine

m104d01d
%let month=1;
%let year=2011;
data orders;
keep order_date order_type quantity total_retail_price;
set orion.order_fact end=final;
where year(order_date)=&year and month(order_date)=&month;
if order_type=3 then do;
Number+1;
Amount+total_retail_price;
Date=order_date;
retain date;
end;
if final then do;
if number=0 then do;
call symputx('dat', 'N/A');
call symputx('avg', 'N/A');
end;
else do;
call symputx('dat', put(date,mmddyy10.));
call symputx('avg', put(amount/number,dollar8.));
end;
end;
run;
proc print data=orders;
title "Orders for &month-&year";
footnote1 "Average Internet Order: &avg";
footnote2 "Last Internet Order: &dat";
run;
The PUT function returns a character string by writing a value with a specified format.
You can use the PUT function to do the following:
 format the result of a numeric expression
 perform explicit numeric-to-character conversion

General form of the PUT function:

PUT(source, format)

source is a constant, variable, or expression (numeric or character).

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-16 Chapter 4 DATA Step and SQL Interfaces

format is any SAS format or user-defined format. It determines the width of the resulting string
and whether the string is right-aligned or left-aligned. The type for format must match the type
for source.
PROC PRINT Output
Orders for 1-2007

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

Average Internet Order: $125

Last Internet Order: 01/28/2011

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-17

4.02 Short Answer Poll

This CALL SYMPUTX statement creates a macro variable
named date. Complete the call to assign the value
of the variable current, formatting the result as a date
such as 21NOV2012.
call symputx('date', );

Business Scenario
Based on user-selected time periods, dynamically compute
statistics for automatic inclusion within the graph below.

27 m104d02

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-18 Chapter 4 DATA Step and SQL Interfaces

Passing Values between Steps

%let start=01Jan2011;
%let stop=31Dec2011;
proc means data=orion.order_fact noprint;
where order_date between "&start"d and "&stop"d;
var total_retail_price;
output out=stats n=count mean=avg;
run;
data _null_;
set stats;
call symputx('orders',count); sam e data set
call symputx('average',avg);
run;
proc sgplot data=orion.order_fact;
where order_date between "&start"d and "&stop"d;
vbar order_type / response=total_retail_price stat=mean
group=order_type dataskin=gloss;
refline &average / axis=y ;
inset ("Orders this period:"="&orders"
"Overall Average:"="%sysfunc(putn(&average,dollar4.))")
/border textattrs=(Color=blue Weight=Bold);
xaxis type=discrete;
yaxis values=(0 to 320 by 40);
format total_retail_price dollar4. order_type 3.;
label total_retail_price='Average Order';
title "Report from &start to &stop";
run;

28 m104d02

 The PUTN function returns a character string by writing a value with a numeric format.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-19

Passing Values between Steps

m104d02
%let start=01Jan2011;
%let stop=31Dec2011;

proc means data=orion.order_fact noprint;

where order_date between "&start"d and "&stop"d;
var total_retail_price;
output out=stats n=count mean=avg;
run;

data _null_;
set stats;
call symputx('orders',count);
call symputx('average',avg);
run;

proc sgplot data=orion.order_fact;

where order_date between "&start"d and "&stop"d;
vbar order_type / response=total_retail_price stat=mean
group=order_type dataskin=gloss;
refline &average / axis=y ;
inset ("Orders this period:" = "&orders"
"Overall Average:" = "%sysfunc(putn(&average,dollar4.))")
/border textattrs=(Color=blue Weight=Bold);
xaxis type=discrete;
yaxis values=(0 to 320 by 40);
format total_retail_price dollar4. order_type 3.;
label total_retail_price='Average Order';
title1 "Report from &start to &stop";
run;

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-20 Chapter 4 DATA Step and SQL Interfaces

Exercises

Level 1

1. Creating Macro Variables with the SYMPUTX Routine

a. Open the m104e01 program shown below into the Editor window. Submit the program and
examine the output that it creates.
data staff;
keep employee_ID job_title salary gender;
set orion.staff;
where job_title contains 'Audit';
run;

proc print data=staff;

sum salary;
title 'Audit Staff';
run;
b. Add a %LET statement before the DATA step to create the macro variable job with the value
Audit. Replace hardcoded values of Audit with references to the macro variable job. Resubmit
the program. It should produce the same output as before.
c. Change the value of the macro variable job from Audit to Analyst. Resubmit the program
and examine the new results.
d. Modify the DATA step to create the macro variable avg to store the average salary. Reference
the macro variable avg in a FOOTNOTE statement. Format the average salary with a dollar sign,
comma, and no decimal places.

PROC PRINT Output

Analyst Staff

Obs Employee_ID Job_Title Salary Gender

1 120263 Financial Analyst III $42,605 M

2 120264 Financial Analyst II $37,510 F
3 120710 Business Analyst II $54,840 M
4 120711 Business Analyst III $59,130 F
5 120775 HR Analyst II $41,580 F
6 120779 HR Analyst II $43,690 F
7 121148 Business Analyst II $52,930 M
============
$332,285

Average Salary: $47,469

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.1 Creating Macro Variables in the DATA Step 4-21

Level 2

2. Creating Macro Variables with the SYMPUTX Routine

a. Open the m104e02 program shown below into the Editor window. This program creates
a summary data set named customer_sum that summarizes Total_Retail_Price by
Customer_ID and sorts the data set by descending CustTotalPurchase. Submit the program
(part a) and examine the output that it creates.
proc means data=orion.order_fact nway noprint;
var Total_Retail_Price;
class Customer_ID;
output out=customer_sum sum=CustTotalPurchase;
run;

proc sort data=customer_sum;

by descending CustTotalPurchase;
run;

proc print data=customer_sum(drop=_type_);

run;
b. Create a macro variable named top that contains the ID number for the top customer. Then
modify the program (part b) to print only the orders for Orion’s top customer.
Partial PROC PRINT Output
Orders for Customer 16 - Orion's Top Customer

Order_ Order_ Delivery_

Order_ID Type Date Date

1230450371 2 24MAR2007 26MAR2007

1231305521 2 27AUG2007 04SEP2007
1231305531 2 27AUG2007 29AUG2007
1234538390 2 12JAN2009 14JAN2009
1234588648 2 17JAN2009 19JAN2009

Challenge

3. Creating Macro Variables with the SYMPUTX Routine

a. Open the m104e03 program shown below into the Editor window. Submit the program
and examine the output that it creates.
proc means data=orion.order_fact nway noprint;
var Total_Retail_Price;
class Customer_ID;
output out=customer_sum sum=CustTotalPurchase;
run;

proc sort data=customer_sum ;

by descending CustTotalPurchase;

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-22 Chapter 4 DATA Step and SQL Interfaces

run;

proc print data=customer_sum(drop=_type_);

run;
b. Using the customer_sum data set, create a single macro variable, top3, that contains
the customer IDs of the top three customers by revenue.
 Customer_ID is a numeric variable.
c. Using the orion.customer_dim data set, print a listing of the top three customers.

PROC PRINT Output

Top 3 Customers

Customer_ID Customer_Name Customer_Type

10 Karen Ballinger Orion Club members high activity

16 Ulrich Heyde Internet/Catalog Customers
45 Dianne Patchin Orion Club Gold members low activity

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-23

4.2 Indirect References to Macro

Variables

Objectives
 Reference macro variables indirectly.
 Create a series of macro variables using
the SYMPUTX routine.

Business Scenario
Create an order history for a given customer. Report
titles should display customer name and number.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-24 Chapter 4 DATA Step and SQL Interfaces

Table Lookup Application

Step 1 Hardcode the program, including customer
name and number.
proc print data=orion.order_fact;
where customer_ID=9;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: 9";
title2 "Customer Name: Cornelia Krahl";
run;

m104d03a
35

Table Lookup Application

Step 2 Create and reference a macro variable
for the customer number.
%let custID=9;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: Cornelia Krahl";
run;

How can you reference the customer name in TITLE2

without hardcoding it?

m104d03b
36

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-25

Table Lookup Application

The orion.customer data set contains customer names
and ID numbers. Customer ID numbers are unique.
Orion.Customer

Birth_
Obs Customer_ID Customer_Name Country Gender Date

1 4 James Kvarniq US M 27JUN1974

2 5 Sandrina Stephano US F 09JUL1979
3 9 Cornelia Krahl DE F 27FEB1974
4 10 Karen Ballinger US F 18OCT1984
5 11 Elke Wallstab DE F 16AUG1974
6 12 David Black US M 12APR1969
7 13 Markus Sepke DE M 21JUL1988
8 16 Ulrich Heyde DE M 16JAN1939
9 17 Jimmie Evans US M 17AUG1954
10 18 Tonie Asmussen US M 02FEB1954
11 19 Oliver S. Füßling DE M 23FEB1964
12 20 Michael Dineley US M 17APR1959

Table Lookup Application

Step 3 Add a DATA step to create a macro variable
with the customer’s name.

Reference the macro variable in TITLE2.

%let custID=9;

data _null_;
set orion.customer;
where customer_ID=&custID; same
call symputx('name', Customer_Name); WHERE
run;
statement
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &name";
run;

m104d03c
38

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-26 Chapter 4 DATA Step and SQL Interfaces

4.03 Short Answer Poll

How many rows are selected by the DATA step WHERE
statement in the preceding program, repeated below?

%let custID=9;
data _null_;
set orion.customer;
where customer_ID=&custID;
call symputx('name', Customer_Name);
run;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &name";
run;

m104d03c
39

Table Lookup Application

To select all customers, eliminate the WHERE statement
from the DATA step.
%let custID=9;
data _null_;
set orion.customer;
call symputx('name', Customer_Name);
run;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &name";
Output
run; from Proc Print

What is the problem this time?

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-27

Table Lookup Application

Because only one macro variable is created by the
SYMPUTX routine, its value is overwritten with each
iteration of the DATA step.
%let custID=9;
data _null_;
set orion.customer;
call symputx('name', Customer_Name);
run;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
Output from
title1 Proc Print
"Customer Number: &custID";
title2 "Customer Name: &name";
run;

Unique macro variable names are required.

Creating a Series of Macro Variables

Derive unique macro variable names by appending
the customer ID number to a fixed prefix.

Symbol Table
Variable Value
NAME4 James Kvarniq
NAME5 Sandrina Stephano
NAME9 Cornelia Krahl
.
.
.

prefix Custom er_ID

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-28 Chapter 4 DATA Step and SQL Interfaces

Creating a Series of Macro Variables

Step 4 Create a series of macro variables to store
customer names.
data _null_;
set orion.customer;
call symputx('name'||left(Customer_ID),
Customer_Name);
run;

Generate unique macro variable names with a DATA step

character variable or expression as the first argument
to the SYMPUTX routine.

m104d03d
44

Creating a Series of Macro Variables

Partial SAS Log
639627 %put _user_;
GLOBAL NAME10 Karen Ballinger
GLOBAL NAME1033 Selim Okay
GLOBAL NAME11 Elke Wallstab
GLOBAL NAME1100 Ahmet Canko
GLOBAL NAME111 Karolina Dokter
GLOBAL NAME11171 Bill Cuddy
GLOBAL NAME12 David Black
GLOBAL NAME12386 Avinoam Zweig
GLOBAL NAME13 Markus Sepke
GLOBAL NAME14104 Avinoam Zweig
GLOBAL NAME14703 Eyal Bloch
GLOBAL NAME16 Ulrich Heyde
GLOBAL NAME1684 Carglar Aydemir
GLOBAL NAME17 Jimmie Evans
GLOBAL NAME17023 Susan Krasowski
GLOBAL NAME171 Robert Bowerman
GLOBAL NAME18 Tonie Asmussen
GLOBAL NAME183 Duncan Robertshawe
GLOBAL NAME19 Oliver S. Füßling
45

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-29

Creating a Series of Macro Variables

You can now reference the correct name without
rerunning the DATA step.
Symbol Table
Variable Value
CUSTID 9
NAME4 James Kvarniq
NAME5 Sandrina Stephano
NAME9 Cornelia Krahl
.
. .
.
. .

4.04 Short Answer Poll

Modify program m104a02 to create an order history for
custID 4. How many program changes are required?

m104a02
47

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-30 Chapter 4 DATA Step and SQL Interfaces

Indirect References to Macro Variables

Because the custID value matches the numeric suffix
of another macro variable, custID can indirectly reference
the other macro variable.

Symbol Table
Variable Value
CUSTID 9
NAME4 James Kvarniq
NAME5 Sandrina Stephano
NAME9
.
Cornelia
.
Krahl
.
. .
.

Indirect References to Macro Variables

The Forward Rescan Rule
 Multiple ampersands preceding a name token denote
an indirect reference.
 Two ampersands (&&) resolve to one ampersand (&).
 The macro processor rescans an indirect reference,
left to right, from the point where multiple ampersands
begin.
 Scanning continues until no more references can
be resolved.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-31

Indirect References to Macro Variables

Step 5 Use an indirect reference.
%let custID=9;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &&name&custID";
run;

m104d03f
51

Indirect References to Macro Variables

The indirect reference causes a second scan.
reference &&name&custID

1st scan &name9

2nd scan Cornelia Krahl

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-32 Chapter 4 DATA Step and SQL Interfaces

Indirect References to Macro Variables

The custID macro variable indirectly references
a name macro variable.
Symbol Table
Variable Value
CUSTID 9
NAME4 James Kvarniq
NAME5 Sandrina Stephano
NAME9
.
Cornelia
.
Krahl
.
. .
.

Scan sequence:
&&name&custID &name9 Cornelia Krahl

Completed Business Scenario

PROC PRINT Output
Customer Number: 9
Customer Name: Cornelia Krahl

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

160 15APR2004 3 1 $29.40

273 07JUN2005 3 2 $16.00
288 10AUG2005 3 3 $1,542.60
289 10AUG2005 3 2 $550.20
316 02DEC2005 3 2 $39.20
326 25DEC2005 3 1 $514.20

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-33

4.05 Short Answer Poll

Submit program m104a03.
%let custid=9;
%let name9=Joe;
%put &name&custid;

How many times are the macro variables scanned

in the %PUT statement?

Is this successful?

m104a03
55

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-34 Chapter 4 DATA Step and SQL Interfaces

Indirect References to Macro Variables

m104d04
data _null_;
set orion.customer;
call symputx('name'||left(Customer_ID), customer_Name);
run;

%let custID=9;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &&name&custID";
run;
Partial SAS Log
451 %let custID=9;
452 proc print data=orion.order_fact;
453 where customer_ID=&custID;
SYMBOLGEN: Macro variable CUSTID resolves to 9
454 var order_date order_type quantity total_retail_price;
SYMBOLGEN: Macro variable CUSTID resolves to 9
455 title1 "Customer Number: &custID";
SYMBOLGEN: && resolves to &.
SYMBOLGEN: Macro variable CUSTID resolves to 9
SYMBOLGEN: Macro variable NAME9 resolves to Cornelia Krahl
456 title2 "Customer Name: &&name&custID";
457 run;

NOTE: There were 6 observations read from the data set ORION.ORDER_FACT.
WHERE customer_ID=9;
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.01 seconds
cpu time 0.00 seconds

PROC PRINT Output

Customer Number: 9
Customer Name: Cornelia Krahl

Order_ Order_ Total_Retail_

Obs Date Type Quantity Price

160 15APR2008 3 1 $29.40

273 07JUN2009 3 2 $16.00
288 10AUG2009 3 3 $1,542.60
289 10AUG2009 3 2 $550.20
316 02DEC2009 3 2 $39.20
326 25DEC2009 3 1 $514.20

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4.2 Indirect References to Macro Variables 4-35

SYMGET Function

Create macro Resolve macro

variables variables

Word scanning time %LET &macvar

Execution time CALL SYMPUTX SYMGET(macvar)

Program Data Vector Symbol Table

SYMPUTX

SYMGET

SYMGET Function
The SYMGET function retrieves a macro variable’s
value during DATA step execution.

Symbol Table
Program Data Vector
DATA Step SYMGET
Variables

The SYMGET function is covered in SAS® Macro

Language 2: Advanced Techniques.

Copyright © 2016, SAS Institute Inc., Cary, North Carolina, USA. ALL RIGHTS RESERVED.
4-36 Chapter 4 DATA Step and SQL Interfaces

Exercises

Level 1

4. Creating a Series of Macro Variables with the SYMPUTX Routine

a. Open the m104e04 program shown below into the Editor window.
%macro memberlist(id=1020);
%put _user_;
title "A List of &id";
proc print data=orion.customer;
var Customer_Name Customer_ID Gender;
where Customer_Type_ID=&id;
run;
%mend memberlist;

%memberlist()
b. The orion.customer_type data set contains the variable Customer_Type_ID, which uniquely
identifies the customer membership level and activity level. Add a DATA step to create a series
of macro variables named typexxxx, where xxxx is the value of Customer_Type_ID. The value
of each type macro variable should be the value of Customer_Type. Place the DATA step before
the macro definition.

Listing of orion.customer_type

c. Modify the TITLE statement so that it displays the appropriate customer type. Use an indirect
macro variable reference to one of the type variables based on the current value of ID. Submit the
modified program.

Partial PROC PRINT Output

A List of Orion Club members low activity

Obs Customer_Name Customer_ID Gender

1 James Kvarniq 4 M
10 Tonie Asmussen 18 M
19 Alvan Goheen 34 M

d. Call the macro again, but with a parameter value of 2030.

Partial PROC PRINT Output
A List of Orion Club Gold members high activity

Obs Customer_Name Customer_ID Gender

11 Oliver S. Füßling 19 M
21 Alphone Greenwald 39 M
25 Annmarie Leveille 49 F
26 Gert-Gunter Mendler 50 M

Level 2

5. Using Indirect References in a Macro Call

a. Open the m104e05 program shown below into the Editor window. Submit the program and
examine the results.
data _null_;
set orion.customer_type;
call symputx('type'||left(Customer_Type_ID),
Customer_Type);
run;

%put _user_;

%macro memberlist(custtype);
proc print data=orion.customer_dim;
var Customer_Name Customer_ID Customer_Age_Group;
where Customer_Type="&custtype";
title "A List of &custtype";
run;
%mend memberlist;
b. Create a macro variable named num with the value 2010. Call the Memberlist macro. Pass the
appropriate parameter to the Memberlist macro, such that custtype resolves to Orion Club Gold
members low activity on the macro call.

PROC PRINT Output

A List of Orion Club Gold members low activity

Customer_
Obs Customer_Name Customer_ID Age_Group

7 Markus Sepke 13 15-30 years

24 Dianne Patchin 45 15-30 years
53 Sanelisiwe Collier 2550 15-30 years
58 Bill Cuddy 11171 15-30 years
75 Angel Borwick 70201 31-45 years

Challenge

6. Using a Table Lookup Application

b. Open the m104e06 program shown below into the Editor window.
%let code=AU;
proc print data=Orion.Employee_Addresses;
var Employee_Name City;
where Country="&code";
title "A List of xxxxx Employees";
run;
c. Use indirect macro variable referencing to replace the xxxxx with the appropriate country name.

Partial PROC PRINT Output

A List of Australia Employees

Obs Employee_Name City

2 Aisbitt, Sandy Melbourne

17 Bahlman, Sharon Sydney
18 Baker, Gabriele Sydney
22 Baran, Shanmuganathan Sydney
23 Barbis, Viney Sydney
24 Barcoe, Selina Melbourne
25 Barreto, Geok-Seng Sydney
31 Billington, Kareen Sydney
34 Blanton, Brig Melbourne
37 Body, Meera Sydney
48 Buddery, Jeannette Sydney
52 Cantatore, Lorian Sydney

4.3 Creating Macro Variables in SQL

Objectives
 Create macro variables during PROC SQL execution.
 Store several values in one macro variable using
the SQL procedure.

Business Scenario
Create a macro variable that contains the total price
of all 2011 Internet orders.

TOTAL= $6,731

INTO Clause
The INTO clause creates macro variables from query
results.

Symbol Table

INTO clause
Query Results

INTO Clause
This INTO clause creates a single macro variable
named total.
proc sql noprint;
select sum(total_retail_price) format=dollar8.
into :total
from orion.order_fact
where year(order_date)=2011 and order_type=3;
quit;
SELECT col1, col2, ...
%put &=total; INTO :mvar1, :mvar2,...
FROM table-expression
Partial SAS Log WHERE where-expression
ORDER BY col1, col2, . . . ;
TOTAL= $6,731

This form of the INTO clause does not trim leading

or trailing blanks.
m104d05a
70

 A %LET statement can be used to trim leading and trailing blanks created by the INTO clause:

%let macrovariable=&macrovariable;

INTO Clause
The INTO clause can create multiple macro variables.

Example: Create macro variables with the date and

amount of the top three sales from 2011.
title 'Top 2011 Sales';
proc sql outobs=3 double;
select total_retail_price, order_date format=mmddyy10.
into :price1-:price3, :date1-:date3
from orion.order_fact
where year(order_date)=2011
order by total_retail_price desc;
quit;

m104d05b
71

INTO Clause
SQL Result
Top 2011 Sales

Total_Retail_
Price Order_Date
Macro variables ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ Macro variables
$1,937.20 06/20/2011
price1 date1
price2 $1,066.40 11/01/2011 date2
price3 $760.80 12/12/2011 date3
Partial SAS Log
1529 %put &price1 &date1, &price2 &date2, &price3 &date3;
$1,937.20 06/20/2011, $1,066.40 11/01/2011, $760.80 12/12/2011

m104d05b
72

INTO Clause
Example: Create a macro variable with a list of all
customer countries. Delimit the country
codes with a comma and a space.
proc sql noprint;
select distinct country
into :countries separated by ', '
from orion.customer;
quit;
INTO : macro-variable SEPARATED BY 'delimiters'
%put &=Countries;

Partial SAS Log

COUNTRIES=AU, CA, DE, IL, TR, US, ZA

The SEPARATED BY argument stores multiple values

in a single macro variable.
m104d05c
73

The DATA step below creates the same macro variable.

proc sort data=orion.customer(keep=country) nodupkey
out=allcountries;
by country;
run;

data _null_;
set allcountries end=eof;
length countries $ 50;
retain countries;
countries=catx(', ', countries, country);
if eof then call symputx('countries', countries);
run;

4.06 Multiple Choice Poll

Which technique creates macro variables during execution
time?
a. %LET statement
b. SYMPUTX routine
c. INTO clause
d. both b and c

SQL Procedure
Example: Display user-defined macro variables in a report.

proc sql flow;

select name, value
from dictionary.macros
where scope='GLOBAL'
order by name;
quit; Partial Output
Macro Variable
Name Macro Variable Value
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

AVERAGE 157.49094595
AVG $125
COUNTRIES AU, CA, DE, IL, TR, US, ZA
CUSTID 9
DAT 01/28/2011
DATE1 06/20/2011
DATE2 11/01/2011
DATE3 12/12/2011
MONTH 1
NAME Cornelia Krahl
ORDERS 148
...
m104d05d
76

 The dictionary.macros table is one of several PROC SQL read-only DICTIONARY tables that
store SAS metadata. For additional information, see “Accessing SAS System Information Using
DICTIONARY Tables” under “Programming with the SQL Procedure” in the SQL Procedure
User’s Guide.

An alternative to the SQL query above is the two-step program below that uses sashelp.vmacro in place
of dictionary.macros.
proc sort data=sashelp.vmacro
(keep=name value scope

where=(scope='GLOBAL'))
out=mymacrovariables(keep=name value)
sortseq=linguistic(numeric_collation=on);
by name;
run;

proc print data=mymacrovariables;

title 'Sorted list of global macro variables';
run;

SQL Procedure
Example: Create a utility macro to display user-defined
macro variables in a report.
%macro putALL;
proc sql flow;
select name, value
from dictionary.macros
where scope='GLOBAL'
order by name;
quit;
%mend putALL;

Call the macro:

%putALL

m104d05d
77

INTO Clause
Example: Create a macro variable with a list of all
user-defined macro variable names. Delimit
the names with spaces.
proc sql noprint;
select name into: vars separated by ' '
from dictionary.macros
where scope='GLOBAL';
quit;

Partial SAS Log

705 %put &vars;
SQLOBS SQLOOPS PRICE1 MONTH CUSTID SYS_SQL_IP_ALL DATE1 COUNTRIES
DAT DATE2 YEAR TOTAL DATE3 NAME ORDERS START SQLXOBS SQLRC AVG STOP
AVERAGE SQLEXITCODE PRICE2 PRICE3

INTO Clause
Example: Create a utility macro that deletes all
user-defined macro variables.
%macro deleteALL;

proc sql noprint;

select name into: vars separated by ' '
from dictionary.macros
where scope='GLOBAL';
quit;

%symdel &vars;

%mend deleteALL;

%deleteALL

m104d05e
79

Exercises

Level 1

7. Creating Macro Variables Using SQL

a. Open the m104e07 program shown below into the Editor window.
%let start=01Jan2011;
%let stop=31Jan2011;

proc means data=orion.order_fact noprint;

where order_date between "&start"d and "&stop"d;
var Quantity Total_Retail_Price;
output out=stats mean=Avg_Quant Avg_Price;
run;

data _null_;
set stats;
call symputx('Quant',put(Avg_Quant,4.2));
call symputx('Price',put(Avg_Price,dollar7.2));
run;

proc print data=orion.order_fact noobs n;

where order_date between "&start"d and "&stop"d;
var Order_ID Order_Date Quantity Total_Retail_Price;
sum Quantity Total_Retail_Price;
format Total_Retail_Price dollar6.;
title1 "Report from &start to &stop";
title3 "Average Quantity: &quant";
title4 "Average Price: &price";
run;
b. Submit the program and view the results.
Report from 01Jan2011 to 31Jan2011

Average Quantity: 2.15

Average Price: $173.89

Total_
Order_ Retail_
Order_ID Date Quantity Price

1241054779 02JAN2011 2 $196

1241063739 03JAN2011 6 $161
1241066216 04JAN2011 2 $306
1241086052 06JAN2011 3 $38
1241147641 13JAN2011 2 $363
1241235281 23JAN2011 1 $73

1241244297 24JAN2011 2 $258

1241244297 24JAN2011 2 $81
1241244297 24JAN2011 3 $358
1241263172 25JAN2011 1 $102
1241263172 25JAN2011 1 $113
1241286432 28JAN2011 2 $174
1241298131 29JAN2011 1 $37
======== =======
28 $2,261

N = 13

c. Delete the macro variables quant and price from the symbol table.

d. Replace the PROC MEANS step and the DATA step with a PROC SQL step.
e. Resubmit the PROC PRINT step and verify that the output is the same.
Level 2
8. Creating a List of Values in a Macro Variable Using SQL
a. Open the m104e08 program into the Editor window. Create a macro variable named top3 with
the customer ID numbers of the top three customers by Total_Retail_Price. Separate the ID
numbers with a comma and a blank. Use the OUTOBS= option.
proc sql;
select customer_id, sum(Total_Retail_Price) as total
from orion.order_fact
group by Customer_ID
order by total descending;
quit;

 The GROUP BY clause summarizes the data by customer ID number.

The ORDER BY clause sorts the data in descending order.

b. Submit the program and review the results, which are shown below.
Top 3 Customers

Customer_ID Customer_Name Customer_Type

10 Karen Ballinger Orion Club members high activity

16 Ulrich Heyde Internet/Catalog Customers
45 Dianne Patchin Orion Club Gold members low activity

Challenge

9. Creating Multiple Macro Variables Using SQL

a. The orion.customer_type data set contains the variable Customer_Type_ID, which holds
the unique customer type codes. Use the SQL procedure to create a series of macro variables
named CTYPE1 through CTYPExx, where xx resolves to the number of rows that the query
returns.

 You need two queries, one to return the number of rows that the query returns and the
other to create CTYPE1 through CTYPExx.

b. Open the program m104e09 to display only the macro variables that begin with CTYPE.
proc sql;
select name, value
from dictionary.macros
where name like "CTYPE%";
quit;
PROC SQL Output
Macro Variable beginning with CTYPE

Macro Variable Name Macro Variable Value

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
CTYPE1 1010
CTYPE2 1020
CTYPE3 1030
CTYPE8 3010
CTYPE4 1040
CTYPE5 2010
CTYPE6 2020
CTYPE7 2030

4.4 Solutions
Solutions to Exercises
1. Creating Macro Variables with the SYMPUTX Routine
a. Open the program into the Editor window.

b. Add a %LET statement before the DATA step to create the macro variable job with the value
Audit. Replace hardcoded values of Audit with references to the macro variable job. Resubmit
the program. It should produce the same output as before.

c. Change the value of job to Analyst. Resubmit the program and examine the new results.

d. Modify the DATA step to create the macro variable avg to store the average salary. Reference
the macro variable avg in a FOOTNOTE statement. Format the average salary with a dollar sign,
comma, and no decimal places.
%let job=Analyst;

data staff;
keep employee_ID job_title salary gender;
set orion.staff end=last;
where job_title contains "&job";
total+salary;
count+1;
if last then call symputx('avg',
put(total/count,dollar9.));
run;

proc print data=staff;

sum salary;
title "&job Staff";
footnote "Average Salary: &avg";
run;
2. Creating Macro Variables with the SYMPUTX Routine
a. Open the program into the Editor window.
b. Create a macro variable named top that contains the ID number of the top customer.
Then modify the program (part b) to print only the orders for Orion’s top customer.
data _null_;
set customer_sum (obs=1);
call symputx('top', Customer_ID);
run;

proc print data=orion.orders noobs;

where Customer_ID =&top;
var Order_ID Order_Type Order_Date Delivery_Date;
title "Orders for Customer &top - Orion's Top Customer";

run;
3. Creating Macro Variables with the SYMPUTX Routine
a. Open the program into the Editor window.
b. Using the customer_sum data set, create a single macro variable, top3, that contains
the customer IDs of the top three customers by revenue.
data _null_;
set customer_sum(obs=3) end=last;
length top3 $50;
retain top3;
top3=catx(' ',top3, Customer_ID);
/* Alternative Solution for the CATX Function */
/* top3=trim(top3)||' '||left(Customer_ID); */
if last then call symputx('top3', top3);
run;
c. Using the orion.customer_dim data set, print a listing of the top three customers.
proc print data=orion.customer_dim noobs;
where Customer_ID in (&top3);
var Customer_ID Customer_Name Customer_Type;
title 'Top 3 Customers';
run;
4. Creating a Series of Macro Variables with the SYMPUTX Routine
a. Open the program into the Editor window.
b. Concatenating the character value type with the value of the Customer_Type_ID variable
specifies the name of each macro variable. Because the Customer_Type_ID variable is numeric,
the LEFT function is required to remove the leading blanks introduced by the automatic numeric -
to-character conversion that occurs as part of the SYMPUTX routine. The %PUT statement
displays the names and values of all user-created macro variables.

c. Because each macro variable that contains the customer type has a common root at the start
of its name (type) and a suffix that corresponds to the value of the ID macro variable, two
ampersands are used in front of the complete reference.
data _null_;
set orion.customer_type;
call symputx('type'||left(Customer_Type_ID),
Customer_Type);
*Alternative solution using the CATS function;
*call symputx(cats('type',Customer_Type_ID),
Customer_Type);
run;

%put _user_;

%macro memberlist(id=1020);
title "A List of &&type&id";

proc print data=orion.customer;

var Customer_Name Customer_ID Gender;
where Customer_Type_ID=&id;
run;
%mend memberlist;

%memberlist()
SAS Log
39 %put _user_;
GLOBAL TYPE1010 Orion Club members inactive
GLOBAL TYPE1020 Orion Club members low activity
GLOBAL TYPE2010 Orion Club Gold members low activity
GLOBAL TYPE1030 Orion Club members medium activity
GLOBAL TYPE2020 Orion Club Gold members medium activity
GLOBAL TYPE3010 Internet/Catalog Customers
GLOBAL TYPE1040 Orion Club members high activity
GLOBAL TYPE2030 Orion Club Gold members high activity

40 options symbolgen;
41 %let id=1020;
42 proc print data=orion.customer;
43 var Customer_Name Customer_ID Gender;
44 where Customer_Type_ID=&id;
SYMBOLGEN: Macro variable ID resolves to 1020
SYMBOLGEN: && resolves to &.
SYMBOLGEN: Macro variable ID resolves to 1020
SYMBOLGEN: Macro variable TYPE1020 resolves to Orion Club members low activity
45 title "A List of &&type&id";
46 run;

NOTE: There were 17 observations read from the data set ORION.CUSTOMER.
WHERE Customer_Type_ID=1020;
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.17 seconds
cpu time 0.00 seconds

d. Call the macro again, but with a parameter value of 2030.

%memberlist(id=2030)

Partial SAS Log

SYMBOLGEN: && resolves to &.
SYMBOLGEN: Macro variable ID resolves to 2030
SYMBOLGEN: Macro variable TYPE2030 resolves to Orion Club Gold members high activity
SYMBOLGEN: Macro variable ID resolves to 2030

NOTE: There were 10 observations read from the data set ORION.CUSTOMER.
WHERE Customer_Type_ID=2030;
NOTE: PROCEDURE PRINT used (Total process time):
real time 1.43 seconds
cpu time 0.00 seconds

5. Using Indirect References in a Macro Call

a. Open the program into the Editor window.

b. Create a macro variable named num with the value of 2010. Execute the macro so that the value
of custtype resolves to Orion Club members low activity in the macro call.

%let num=2010;

%memberlist(&&type&num)

Partial SAS Log

18 %let num=2010;
19 %memberlist(&&type&num)
SYMBOLGEN: && resolves to &.
SYMBOLGEN: Macro variable NUM resolves to 2010
SYMBOLGEN: Macro variable TYPE2010 resolves to Orion Club Gold members low activity
SYMBOLGEN: Macro variable CUSTTYPE resolves to Orion Club Gold members low activity
SYMBOLGEN: Macro variable CUSTTYPE resolves to Orion Club Gold members low activity
NOTE: There were 5 observations read from the data set ORION.CUSTOMER_DIM.
WHERE Customer_Type='Orion Club Gold members low activity';
NOTE: PROCEDURE PRINT used (Total process time):
real time 2.24 seconds
cpu time 0.00 seconds

6. Using a Table Lookup Application

a. Using orion.country, create a series of macro variables in which the name of the macro variable
is the country abbreviation (Country) and the value of the macro variable is the country name
(Country_Name).
data _null_;
set orion.country;
call symputx(Country, Country_Name);
run;
%put _user_;
Partial SAS Log
639645 %put _user_;
GLOBAL AU Australia
GLOBAL CA Canada
GLOBAL DE Germany
GLOBAL IL Israel
GLOBAL TR Turkey
GLOBAL US United States
GLOBAL ZA South Africa

b. Open the program into the Editor window.

c. Use indirect macro variable referencing to replace the xxxxx with the appropriate country name.
%let code=AU;
proc print data=orion.employee_addresses;
var Employee_Name City;
where Country="&code";
title "A List of &&&code Employees";
run;
7. Creating Macro Variables Using SQL

a. Open the program into the Editor window.

b. Submit the program and view the results.
c. Delete the macro variables quant and price.

%symdel quant price;

d. Replace the PROC MEANS step and the DATA step with a PROC SQL step.
proc sql noprint;
select mean(Quantity) format=4.2,
mean(Total_Retail_Price) format=dollar7.2
into :quant, :price
from orion.order_fact
where order_date between "&start"d and "&stop"d;
quit;
e. Resubmit the PROC PRINT step and verify that the output is the same.
8. Creating a List of Values in a Macro Variable Using SQL
a. Open the m104e08 program into the Editor window and modify the SQL procedure to create
a macro variable named top3 that contains the customer ID numbers of the top three customers
by Total_Retail_Price in the orion.order_fact data set. Separate each of the values with
a comma and a blank. Use the OUTOBS= option to limit the number of output rows.
proc sql noprint outobs=3;
select customer_id, sum(Total_Retail_Price) as total
into :top3 separated by ', '
from orion.order_fact
group by Customer_ID
order by total descending;
b. Submit the program and review the results, which are shown in part b of the exercise.
9. Creating Multiple Macro Variables Using SQL
a. The first query creates the numobs macro variable that stores how many records are returned
by the query. This is the same as the number of macro variables in each series.

A special form of the INTO clause is useful for creating a series of macro variables from multiple
rows of an SQL query.
proc sql noprint;
select count(*) into :numobs
from orion.customer_type;
%let numobs=&numobs;
select Customer_Type_ID into :ctype1-:ctype&numobs
from orion.customer_type;
quit;
b. Submit the program and review the results, which are shown in part b of the exercise.

Solutions to Student Activities (Polls/Quizzes)

4.01 Multiple Choice Poll – Correct Answer

What is the value of foot after execution of the DATA
step?
a. No Internet orders
b. Some Internet orders

data _null_;
call symputx('foot','No Internet orders');
%let foot=Some Internet orders;
run;

1. Word scanning begins. DATA step compilation

begins.
2. %LET encountered. Macro trigger executes.
3. Step boundary. DATA step executes. SYMPUTX
18 executes.

4.02 Short Answer Poll – Correct Answer

4.03 Short Answer Poll – Correct Answer

How many rows are selected by the DATA step WHERE
statement in the preceding program, repeated below?
one row
%let custID=9;
data _null_;
set orion.customer;
where customer_ID=&custID;
call symputx('name', Customer_Name);
run;
proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &name";
run;

Each time you select a customer number, the DATA

step rereads the entire customer data set to select
a subset of one customer. m104d03c
40

4.04 Short Answer Poll – Correct Answer

Modify program m104a02 to create an order history for
custID 4. How many program changes are required?

%let custID=9; Change

proc print data=orion.order_fact;
where customer_ID=&custID;
var order_date order_type quantity total_retail_price;
title1 "Customer Number: &custID";
title2 "Customer Name: &name9"; Change
run;

Two program changes are required.

m104a02
48

4.05 Short Answer Poll – Correct Answer

Submit program m104a03.
%let custid=9;
%let name9=Joe;
%put &name&custid;
How many times are the macro variables scanned
in the %PUT statement?
One time

Is this successful?
No. It generates the following warning:
WARNING: Apparent symbolic reference NAME not resolved.

m104a03
56

4.06 Multiple Choice Poll – Correct Answer

Which technique creates macro variables during execution
time?
a. %LET statement
b. SYMPUTX routine
c. INTO clause
d. both b and c

5.1 Conditional Processing............................................................................................. 5-3

Exercises............................................................................................................. 5-15

5.2 Parameter Validation ............................................................................................... 5-17

Exercises............................................................................................................. 5-22

5.3 Iterative Processing ................................................................................................ 5-25

Exercises............................................................................................................. 5-36

5.4 Global and Local Symbol Tables ............................................................................ 5-39

Exercises............................................................................................................. 5-53

5.5 Solutions ................................................................................................................. 5-56

Solutions to Exercises ........................................................................................... 5-56

Solutions to Student Activities (Polls/Quizzes) ........................................................... 5-67

5-2 Chapter 5 Macro Programs

5.1 Conditional Processing

Objectives
 Conditionally process SAS code within a macro
program.
 Monitor macro execution.
 Insert entire steps, entire statements, and partial
statements into a SAS program.

Business Scenario
A daily sales report is generated every night. Every
Friday, a weekly report is generated. Determine the best
method to automate these reports.

Every Night
PROC PRINT

Every Friday

PROC MEANS

Conditional Processing

proc print data=orion.order_fact;

Always print where order_date="&sysdate9"d;
the daily var product_id total_retail_price;
report. title "Daily sales: &sysdate9";
run;

proc means data=orion.order_fact n sum mean;

where order_date between
Yes "&sysdate9"d - 6 and "&sysdate9"d;
Is it
Friday? var quantity total_retail_price;
title "Weekly sales: &sysdate9";
run;

Solutions
Three methods:
Method 1 Create multiple macros, including a driver macro.
Method 2 Create a single macro with %DO and %END statements.
Method 3 Create a single macro with %INCLUDE statements.

1? 3?

6
6

Processing Complete Steps

Method 1 Create separate macros for the Daily and Weekly
programs.
%macro daily;
proc print data=orion.order_fact;
where order_date="&sysdate9"d;
var product_id total_retail_price;
title "Daily sales: &sysdate9";
run;
%mend daily;

%macro weekly;
proc means data=orion.order_fact n sum mean;
where order_date between
"&sysdate9"d-6 and "&sysdate9"d;
var quantity total_retail_price;
title "Weekly sales: &sysdate9";
run;
%mend weekly;
m105d01a
7 continued...

Processing Complete Steps

Method 1 Create a driver macro that always calls the Daily
macro and conditionally calls the Weekly macro.

%macro reports;
%daily
%if &sysday=Friday %then %weekly;
%mend reports;
%IF expression %THEN action;
%ELSE action;

 Character constants are

• not quoted
• case sensitive.
 The %ELSE statement is optional.
 %IF-%THEN and %ELSE statements can
be used inside a macro definition only.
m105d01a
8

Macro Expressions
%IF expression

Macro SAS
Expressions Expressions

Arithmetic operators  
Logical operators
 
(do not precede AND or OR with %)
Comparison operators
 
(symbols and mnemonics)
Case sensitivity  
Special WHERE operators
Quotation marks 
Ranges such as 1<=x<=10 
IN operator  
9 IN operator: parentheses required 

 Special WHERE operators include CONTAINS, IS NULL, IS MISSING, LIKE,

BETWEEN-AND, SAME-AND, and =* (sounds like).

Conditional Processing
%IF ... %THEN action;
%ELSE action;

These actions can follow keywords %THEN and %ELSE:

 a macro language statement
 a macro variable reference
 a macro call
 any text

Monitoring Macro Execution

The MLOGIC system option displays macro execution
messages in the SAS log.

Partial SAS Log

494 %macro reports;
495 %daily
496 %if &sysday=Friday %then %weekly;
497 %mend reports;
498
499 options mlogic;
OPTIONS MLOGIC;
500 %reports
MLOGIC(REPORTS): Beginning execution.
MLOGIC(DAILY): Beginning execution.
MLOGIC(DAILY): Ending execution.
MLOGIC(REPORTS): %IF condition &sysday=Friday is TRUE
MLOGIC(WEEKLY): Beginning execution.
MLOGIC(WEEKLY): Ending execution.
MLOGIC(REPORTS): Ending execution.

 The default setting is NOMLOGIC. m105d01a

 The SYMBOLGEN option can be used to debug %IF expressions.

5.01 Short Answer Poll

Submit the program m105a01. What error do you see
in the log?
%macro reports;
%daily
%if &sysday=Friday then %weekly;
%mend reports;

Processing Complete Steps

Method 2 Create a single macro with %DO and %END statements
to generate text that contains semicolons.
%macro reports;
proc print data=orion.order_fact;
where order_date="&sysdate9"d;
var product_id total_retail_price;
title "Daily sales: &sysdate9";
run;
%if &sysday=Friday %then %do;
proc means data=orion.order_fact n sum mean;
where order_date between
%IF expression
"&sysdate9"d %THEN
- 6 and %DO;
"&sysdate9"d;
var quantitystatement; statement;...
total_retail_price;
title "Weekly
%END;sales: &sysdate9";
run; %ELSE %DO;
%end;
%mend reports; statement; statement;...
%END; m105d01b
14

Processing Complete Steps

Method 3 Create a single macro with %INCLUDE statements.

%INCLUDE file-specification < / SOURCE2 >;

%macro reports;
%include "&path\daily.sas";
%if &sysday=Friday %then %do;
%include "&path\weekly.sas";
%end;
%mend reports;
The %INCLUDE statement
 retrieves SAS source code from an external file and places
it on the input stack
 is a global SAS statement, not a macro language statement.

The SOURCE2 option displays inserted SAS statements

in the SAS log. m105d01c
15

Business Scenario
Generate frequency reports for all order types
or a specific order type.

01Jan2011
To
31Dec2011

orion.order_fact

Processing Complete Statements

Example: Insert individual statements within a PROC step.
%macro count(type=,start=01jan2011,stop=31dec2011);
proc freq data=orion.order_fact;
where order_date between "&start"d and "&stop"d;
tables quantity;
title1 "Orders from &start to &stop";
%if &type= %then %do;
title2 "For All Order Types";
%end;
%else %do;
title2 "For Order Type &type Only";
where same and order_type=&type;
%end;
run;
%mend count;
options mprint mlogic;
%count()
%count(type=3)

m105d02
17

Processing Complete Statements

SAS Log
802 %count()
MLOGIC(COUNT): Beginning execution.
MLOGIC(COUNT): Parameter TYPE has value
MLOGIC(COUNT): Parameter START has value 01jan2011
MLOGIC(COUNT): Parameter STOP has value 31dec2011
MPRINT(COUNT): proc freq data=orion.order_fact;
MPRINT(COUNT): where order_date between "01jan2011"d and "31dec2011"d;
MPRINT(COUNT): tables quantity;
MPRINT(COUNT): title1 "Orders from 01jan2011 to 31dec2011";
MLOGIC(COUNT): %IF condition &type= is TRUE
MPRINT(COUNT): title2 "For All Order Types";
MPRINT(COUNT): run;

NOTE: There were 148 observations read from the data set ORION.ORDER_FACT.
WHERE (order_date>='01JAN2011'D and order_date<='31DEC2011'D);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

MLOGIC(COUNT): Ending execution.

18 m105d02

Processing Complete Statements

SAS Log
803 %count(type=3)
MLOGIC(COUNT): Beginning execution.
MLOGIC(COUNT): Parameter TYPE has value 3
MLOGIC(COUNT): Parameter START has value 01jan2011
MLOGIC(COUNT): Parameter STOP has value 31dec2011
MPRINT(COUNT): proc freq data=orion.order_fact;
MPRINT(COUNT): where order_date between "01jan2011"d and "31dec2011"d;
MPRINT(COUNT): tables quantity;
MPRINT(COUNT): title1 "Orders from 01jan2011 to 31dec2011";
MLOGIC(COUNT): %IF condition &type= is FALSE
MPRINT(COUNT): title2 "For Order Type 3 only";
MPRINT(COUNT): where same and order_type=3;
NOTE: Where clause has been augmented.
MPRINT(COUNT): run;

NOTE: There were 40 observations read from the data set ORION.ORDER_FACT.
WHERE (order_date>='01JAN2011'D and order_date<='31DEC2011'D) and
(order_type=3);
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

MLOGIC(COUNT): Ending execution.

19 m105d02

SAME-AND is a WHERE statement operator that supplements an existing WHERE statement without
respecifying the original WHERE statement.

Processing Complete Statements

Example: Insert individual statements within a DATA step.
%macro cust(place);
%let place=%upcase(&place);
data customers;
set orion.customer;
%if &place=US %then %do;
where country='US';
keep customer_name customer_address country;
%end;
%else %do;
where country ne 'US';
keep customer_name customer_address country location;
length location $ 12;
if country="AU" then location='Australia';
else if country="CA" then location='Canada';
else if country="DE" then location='Germany';
else if country="IL" then location='Israel';
else if country="TR" then location='Turkey';
else if country="ZA" then location='South Africa';
%end;
run;
%mend cust;
%cust(us)
%cust(international)

20 m105d03

The %UPCASE function converts values to uppercase.

%UPCASE(argument)

Because macro comparisons are case sensitive, %UPCASE can eliminate case sensitivity when
a macro program checks a parameter value.

Processing Complete Statements

SAS Log
828 %cust(us)
MLOGIC(CUST): Beginning execution.
MLOGIC(CUST): Parameter PLACE has value us
MLOGIC(CUST): %LET (variable name is PLACE)
MPRINT(CUST): data customers;
MPRINT(CUST): set orion.customer;
MLOGIC(CUST): %IF condition &place=US is TRUE
MPRINT(CUST): where country='US';
MPRINT(CUST): keep customer_name customer_address country;
MPRINT(CUST): run;

NOTE: There were 28 observations read from the data set ORION.CUSTOMER.
WHERE country='US';
NOTE: The data set WORK.CUSTOMERS has 28 observations and 3 variables.
NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

MLOGIC(CUST): Ending execution.

21 m105d03

Processing Complete Statements

SAS Log
829 %cust(international)
MLOGIC(CUST): Beginning execution.
MLOGIC(CUST): Parameter PLACE has value international
MLOGIC(CUST): %LET (variable name is PLACE)
MPRINT(CUST): data customers;
MPRINT(CUST): set orion.customer;
MLOGIC(CUST): %IF condition &place=US is FALSE
MPRINT(CUST): where country ne 'US';
MPRINT(CUST): keep customer_name customer_address country location;
MPRINT(CUST): length location $ 12;
MPRINT(CUST): if country="AU" then location='Australia';
MPRINT(CUST): else if country="CA" then location='Canada';
MPRINT(CUST): else if country="DE" then location='Germany';
MPRINT(CUST): else if country="IL" then location='Israel';
MPRINT(CUST): else if country="TR" then location='Turkey';
MPRINT(CUST): else if country="ZA" then location='South Africa';
MPRINT(CUST): run;

NOTE: There were 49 observations read from the data set ORION.CUSTOMER.
WHERE country not = 'US';
NOTE: The data set WORK.CUSTOMERS has 49 observations and 4 variables.
NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

MLOGIC(CUST): Ending execution.

22 m105d03

Idea Exchange
What is the difference between %IF-%THEN
and IF-THEN?

23
23

Business Scenario
Create a macro that conditionally generates a one-way
frequency report or a two-way frequency report.

tables customer_gender;

orion.customer_dim
tables customer_age_group*customer_gender;

25
25

Processing Partial Statements

Conditionally insert text into the middle of a statement.

Example: Generate either a one-way or two-way

frequency table, depending on parameter values.

%macro counts(rows);
title 'Customer Counts by Gender';
proc freq data=orion.customer_dim;
tables
%if &rows ne %then &rows *;
customer_gender;
run;
%mend counts;

%counts()
%counts(customer_age_group)
m105d04
26

Processing Partial Statements

SAS Log
1798 %counts()
MPRINT(COUNTS): title 'Customer Counts by Gender';
MPRINT(COUNTS): proc freq data=orion.customer_dim;
MPRINT(COUNTS): tables customer_gender ;
MPRINT(COUNTS): run;

NOTE: There were 77 observations read from the data set ORION.CUSTOMER_DIM.
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

1799 %counts(customer_age_group)
MPRINT(COUNTS): title 'Customer Counts by Gender';
MPRINT(COUNTS): proc freq data=orion.customer_dim;
MPRINT(COUNTS): tables customer_age_group * customer_gender ;
MPRINT(COUNTS): run;

NOTE: There were 77 observations read from the data set ORION.CUSTOMER_DIM.
NOTE: PROCEDURE FREQ used (Total process time):
real time 0.03 seconds
cpu time 0.03 seconds

m105d04
27

Exercises

Level 1

1. Conditionally Processing Complete Statements

a. Open the m105e01 program shown below into the Editor window.
%macro listing(custtype);
proc print data=orion.customer noobs;
run;
%mend listing;
%listing(2010)
b. Modify the macro to test the CUSTTYPE parameter. If the value is null, insert the following
statements into the PROC PRINT step:
var Customer_ID Customer_Name Customer_Type_ID;
title "All Customers";
If the value is not null, insert the following statements into the PROC PRINT step:
where Customer_Type_ID=&custtype;
var Customer_ID Customer_Name;
title "Customer Type: &custtype";
c. Resubmit the macro definition and call the macro using a null value and a valid value
of CUSTTYPE.

Level 2

2. Debugging a Macro
a. Open the m105e02 program shown below into the Editor window.
%macro day;
%if &sysday=SATURDAY
%then %put Yes;
%else %put Sorry;
%mend day;

options nomlogic nosymbolgen ;

%day
b. Change SATURDAY to today’s value and submit the program.
c. Did the log say Yes? If not, take the following steps, in sequence, until it does.
1) Activate the MLOGIC option, resubmit the program, check the log, and change the day
as necessary.

2) Activate the SYMBOLGEN option, resubmit the program, check the log, and change the day
as necessary.

Challenge

3. Debugging a Macro
a. Open the m105e03 program shown below into the Editor window.
%macro where(state);
%if &state=OR
%then %put Oregon;
%else %put Wherever;
%mend where;

%where(CA)
b. Submit the program. Examine the log and correct the error.

5.2 Parameter Validation

Objectives
 Perform parameter validation with the OR operator.
 Perform parameter validation with the IN operator.
 Perform data-driven parameter validation.

Business Scenario
Validate a parameter against a list of valid COUNTRY
values. Explore three possible methods.

AU
CA
DE ? DE
IL
TR
US
ZA

32
32

Parameter Validation
Method 1 Use OR operators for parameter validation.
%macro customers(place);
%let place=%upcase(&place);
%if &place=AU
or &place=CA
or &place=DE
or &place=IL
or &place=TR
or &place=US
or &place=ZA %then %do;
proc print data=orion.customer;
var customer_name customer_address country;
where upcase(country)="&place";
title "Customers from &place";
run;
%end;
%else %put ERROR: No customers from &place..;
%mend customers;
%customers(de)
%customers(aa)

m105d05a
33

Parameter Validation
SAS Log
955 %customers(de)

NOTE: There were 10 observations read from the data set ORION.CUSTOMER.
WHERE UPCASE(country)='DE';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

956 %customers(aa)
ERROR: No customers from AA.

m105d05a
34

5.02 Short Answer Poll

Instead of using multiple OR operators, which SAS
comparison operator compares a value to a list of values?

if Country='FR'
or Country='CA'
or Country='DE'
or Country='ZA' then do;

Parameter Validation
Method 2 Use the IN operator for parameter validation.

%macro customers(place) / minoperator;

%let place=%upcase(&place);
%if &place in AU CA DE IL TR US ZA %then %do;
proc print data=orion.customer;
var customer_name customer_address country;
where upcase(country)="&place";
title "Customers from &place";
run;
%end;
%else %put Sorry, no customers from &place..;
%mend customers;
%customers(de)
%customers(aa)

 The MINOPERATOR option is required.

m105d05b
37

 The macro IN operator does not require parentheses.

 When using NOT with the IN operator, NOT must precede the IN expression. Parentheses
are required, as shown below.
%if not(&macvar in &valuelist) %then … ;

5.03 Short Answer Poll

What can be done if the value list is extremely long
or changes frequently, or both?
%macro customers(place) / minoperator;
%let place=%upcase(&place);
%if &place in AU CA DE IL TR US ZA %then %do;
proc print data=orion.customer;
var customer_name customer_address country;
where upcase(country)="&place";
title "Customers from &place";
run;
%end;
%else %put Sorry, no customers from &place..;
%mend customers;
%customers(de)
%customers(aa)

m105d05b
38

Data-Driven Validation
Method 3 Use data-driven parameter validation.
%macro customers(place) / minoperator;
%let place=%upcase(&place);
proc sql noprint;
select distinct country into :list separated by ' '
from orion.customer;
quit;
%if &place in &list %then %do;
proc print data=orion.customer;
var customer_name customer_address country;
where upcase(country)="&place";
title "Customers from &place";
run;
%end;
%else %do;
%put ERROR: No customers from &place..;
%put ERROR- Valid countries are: &list..;
%end;
%mend customers;
40 m105d05c

Data-Driven Validation
SAS Log
1246 %customers(de)
NOTE: PROCEDURE SQL used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

NOTE: There were 10 observations read from the data set ORION.CUSTOMER.
WHERE UPCASE(country)='DE';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

1247 %customers(aa)
NOTE: PROCEDURE SQL used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

ERROR: No customers from AA.

Valid countries are: AU CA DE IL TR US ZA.

m105d05c
41

Exercises

Level 1

4. Validating a Macro Parameter

a. Open the m105e04 program shown below into the Editor window and submit it.
%macro custtype(type);
%let type=%upcase(&type);
proc print data=orion.customer_dim;
var Customer_Group Customer_Name Customer_Gender
Customer_Age;
where upcase(Customer_Group) contains "& type";
title "&type Customers";
run;
%mend custtype;
%custtype(internet)
b. Modify the macro to use %IF-%THEN and %ELSE statements to validate the TYPE parameter.
The macro should submit the PROC PRINT step only if the TYPE parameter is GOLD or
INTERNET. If the TYPE parameter is not correct, the macro should write this message to the
SAS log:
ERROR: Invalid TYPE: xxxx.
ERROR: Valid TYPE values are INTERNET or GOLD.
The value xxxx is the TYPE parameter.
 Be sure to set the appropriate option to activate the IN operator.
c. Resubmit the macro definition and call the macro using valid and invalid parameter values.

d. Modify the macro to test whether TYPE is null. If so, do not execute PROC PRINT. Instead,
the macro should write this message to the SAS log:
ERROR: Missing TYPE.
ERROR: Valid values are INTERNET or GOLD.

The macro should first check for a null parameter value. If TYPE is not null, the macro should
convert TYPE to uppercase and test for valid values of GOLD or INTERNET.

e. Resubmit the macro definition with a null parameter value, valid values in uppercase, lowercase,
and mixed case, and an invalid value.

Level 2
5. Validating a Macro Parameter Using Data-Driven TechniquesOpen the m105e05 program shown
below into the Editor window and submit it.
%macro listing(custtype);
%if &custtype= %then %do;
proc print data=orion.customer noobs;
var Customer_ID Customer_Name Customer_Type_ID;

title "All Customers";

run;
%end;
%else %do;
proc print data=orion.customer noobs;
where Customer_Type_ID=&custtype;
var Customer_ID Customer_Name;
title "Customer Type: &custtype";
run;
%end;
%mend listing;

%listing(1020)
%listing()
b. Modify the macro definition to validate CUSTTYPE against a data-driven list.
1) Use the SQL procedure to create the macro variable idlist that contains the unique values
of the variable Customer_Type_ID in the orion.customer_type data set.
2) If the CUSTTYPE parameter is not missing, validate it against idlist.

 The macro IN operator cannot check for a null value. Check that a value is not null
before using the IN operator to check whether the value is valid.

3) If CUSTTYPE is in idlist, execute PROC PRINT. Otherwise, do not execute PROC PRINT.
Instead, the macro should write this message to the SAS log:

Partial SAS Log

ERROR: Value for CUSTTYPE is invalid.
Valid values are 1010 1020 1030 1040 2010 2020 2030 3010

c. Resubmit the macro definition and call the macro with a null parameter value, a valid value,
and an invalid value.

Challenge

6. Validating a Macro Parameter

a. Open the m105e06 program shown below into the Editor window and submit it.
%macro salarystats(decimals=2,order=internal);
options nolabel;
title 'Salary Stats';
proc means data=orion.staff maxdec=&decimals order=&order;
where job_title contains 'Sales';
var salary;
class job_title;
run;
title;
%mend salarystats;

b. Modify the macro to validate parameters according to the following requirements:

Macro Variable Possible Values
decimals 0 to 4
order INTERNAL, FREQ

1) Create an additional macro variable named numerrors that accumulates the number
of parameter errors.
2) Execute PROC MEANS only if numerrors is zero.

 To express NOT IN, use the syntax below.

%IF NOT(&macvar IN &value-list) %THEN … ;

c. The macro should write the following messages to the SAS log when parameters are invalid:

Partial SAS Log

211 %salarystats(decimals=5,order=fudge)
ERROR: Invalid DECIMALS parameter: 5.
Valid DECIMALS values are 0 to 4.
ERROR: Invalid ORDER parameter: FUDGE.
Valid ORDER values are INTERNAL or FREQ.
ERROR: 2 errors. Code not submitted.

5.3 Iterative Processing

Objectives
 Execute macro language statements iteratively.
 Generate SAS code iteratively.

Business Scenario
Macro applications might require iterative processing.

The iterative %DO statement can execute macro

language statements and generate SAS code.

%do;
…
%end;

SYMPUTX Routine (Review)

Example: Create a numbered series of macro variables.
data _null_;
set orion.country end=no_more;
call symputx('Country'||left(_n_),country_name);
if no_more then call symputx('numrows',_n_);
run;

SAS Log
639680 %put _user_;
GLOBAL COUNTRY1 Australia
GLOBAL COUNTRY2 Canada
GLOBAL COUNTRY3 Germany
GLOBAL COUNTRY4 Israel
GLOBAL COUNTRY5 Turkey
GLOBAL COUNTRY6 United States
GLOBAL COUNTRY7 South Africa
GLOBAL NUMROWS 7

m105d06
47

The PROC SQL step below creates the same series of macro variables. In SAS 9.3, you can use a hyphen
in the INTO clause to specify a range without an upper bound.
proc sql noprint;
select country_name into :country1-
from orion.country;
%let numrows=&sqlobs;
quit;

 The automatic macro variable SQLOBS is populated with the number of rows processed
by the SELECT statement.

Simple Loops
Example: Display a series of macro variables in the SAS log
by repeatedly executing %PUT within a macro loop.
%macro putloop;
%do i=1 %to &numrows;
%put Country&i is &&country&i;
%end;
%mend; %DO index-variable=start %TO stop <%BY increment>;
text
%END;

SAS Log
200 %putloop
Country1 is Australia
Country2 is Canada
Country3 is Germany
Country4 is Israel
Country5 is Turkey
Country6 is United States
Country7 is South Africa
m105d07
48

 No code is sent to the compiler when the macro executes. The %PUT statements are executed
by the macro processor.

Simple Loops
%DO index-variable=start %TO stop <%BY increment>;
text
%END;

 %DO and %END are valid only inside a macro definition.

 index-variable is a macro variable created in the local symbol
table if it does not already exist in another symbol table.
 start, stop, and increment values can be any valid macro
expressions that resolve to integers.
 The %BY clause is optional. (The default increment is 1.)
 text can be any of the following:
– constant text
– macro variables or expressions
– macro statements
49 – macro calls

5.04 Multiple Choice Poll

Which statement correctly creates an index variable
named i?
a. %do &i=1 %to 10;
b. %do &i=1 to 10;
c. %do i=1 %to 10;
d. %do i=1 to 10;

Business Scenario
Import a series of text files into corresponding SAS data sets.

orders2007.dat year2007

orders2008.dat year2008

orders2009.dat year2009

52
52

Generating Repetitive Code

Example: Iteratively generate complete SAS steps.
%macro readraw(first=2007,last=2011);
%do year=&first %to &last;
data year&year;
infile "&path\orders&year..dat";
input order_ID order_type order_date : date9.;
run;
%end;
%mend readraw;
options mlogic mprint;

%readraw(first=2008,last=2010)

m105d08
53

Generating Repetitive Code

Partial SAS Log
MLOGIC(READRAW): %DO loop index variable YEAR is now 2009; loop will iterate again.
MPRINT(READRAW): data year2009;
MPRINT(READRAW): infile "s:\workshop\orders2009.dat";
MPRINT(READRAW): input order_ID order_type order_date : date9.;
MPRINT(READRAW): run;

NOTE: The infile "s:\workshop\orders2009.dat" is:

File Name=s:\workshop\orders2009.dat,
RECFM=V,LRECL=256

NOTE: 70 records were read from the infile "s:\workshop\orders2009.dat".

The minimum record length was 22.
The maximum record length was 22.
NOTE: The data set WORK.YEAR2009 has 70 observations and 3 variables.
NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

MLOGIC(READRAW): %DO loop index variable YEAR is now 2010; loop will iterate again.
MPRINT(READRAW): data year2010;
MPRINT(READRAW): infile "s:\workshop\orders2010.dat";
MPRINT(READRAW): input order_ID order_type order_date : date9.;
MPRINT(READRAW): run;
54 m105d08

Generating Data-Dependent Code

Example: Create a separate data set for each value
of a selected variable in a selected data set.
%split(data=orion.customer, var=country)

Partial SAS Log

MPRINT(SPLIT): data AU CA DE IL TR US ZA ;
MPRINT(SPLIT): set orion.customer;
MPRINT(SPLIT): select(country);
MPRINT(SPLIT): when("AU") output AU;
MPRINT(SPLIT): when("CA") output CA;
MPRINT(SPLIT): when("DE") output DE;
MPRINT(SPLIT): when("IL") output IL;
MPRINT(SPLIT): when("TR") output TR;
MPRINT(SPLIT): when("US") output US;
MPRINT(SPLIT): when("ZA") output ZA;
MPRINT(SPLIT): otherwise;
MPRINT(SPLIT): end;
MPRINT(SPLIT): run;

Generating Data-Dependent Code

Step 1 Store unique data values into macro variables.
%macro split (data=, var=);
proc sort data=&data(keep=&var) out=values nodupkey;
by &var;
run;
data _null_;
set values end=last;
call symputx('val'||left(_n_),&var);
if last then call symputx('count',_n_);
run;
%put _local_;
%mend split;
%split(data=orion.customer, var=country)

m105d09a
56

 Values of the selected variable must represent valid SAS names. The NOTNAME function
detects the position of invalid SAS name characters within a character string.

 The statement below writes a macro’s local symbol table to the SAS log.

%put _local_;

Generating Data-Dependent Code

Partial SAS Log
SPLIT COUNT 7
SPLIT DATA orion.customer
SPLIT VAL1 AU
SPLIT VAL2 CA
SPLIT VAL3 DE
SPLIT VAL4 IL
SPLIT VAL5 TR
SPLIT VAL6 US
SPLIT VAL7 ZA
SPLIT VAR country

m105d09a
57

Generating Data-Dependent Code

Step 2 Use loops to generate the DATA step.
%macro split (data=, var=);
proc sort data=&data(keep=&var) out=values nodupkey;
by &var;
run;
data _null_;
set values end=last;
call symputx('val'||left(_n_),&var);
if last then call symputx('count',_n_);
run;
data
%do i=1 %to &count;
&&val&i
%end;
;
set &data;
select(&var);
%do i=1 %to &count;
when("&&val&i") output &&val&i;
%end;
otherwise;
end;
run;
%mend split; m105d09b
58

Generating Data-Dependent Code

Partial SAS Log
MPRINT(SPLIT): data AU CA DE IL TR US ZA ;
MPRINT(SPLIT): set orion.customer;
MPRINT(SPLIT): select(country);
MPRINT(SPLIT): when("AU") output AU;
MPRINT(SPLIT): when("CA") output CA;
MPRINT(SPLIT): when("DE") output DE;
MPRINT(SPLIT): when("IL") output IL;
MPRINT(SPLIT): when("TR") output TR;
MPRINT(SPLIT): when("US") output US;
MPRINT(SPLIT): when("ZA") output ZA;
MPRINT(SPLIT): otherwise;
MPRINT(SPLIT): end;
MPRINT(SPLIT): run;

NOTE: There were 77 observations read from the data set ORION.CUSTOMER.
NOTE: The data set WORK.AU has 8 observations and 12 variables.
NOTE: The data set WORK.CA has 15 observations and 12 variables.
NOTE: The data set WORK.DE has 10 observations and 12 variables.
NOTE: The data set WORK.IL has 5 observations and 12 variables.
NOTE: The data set WORK.TR has 7 observations and 12 variables.
NOTE: The data set WORK.US has 28 observations and 12 variables.
NOTE: The data set WORK.ZA has 4 observations and 12 variables.
NOTE: DATA statement used (Total process time):
real time 0.10 seconds
cpu time 0.10 seconds

m105d09b
59

5.05 Short Answer Poll

Given the symbol table below, what is the value
of &&VAL&COUNT?
SPLIT COUNT 7
SPLIT DATA orion.customer
SPLIT VAL1 AU
SPLIT VAL2 CA
SPLIT VAL3 DE
SPLIT VAL4 IL
SPLIT VAL5 TR
SPLIT VAL6 US
SPLIT VAL7 ZA
SPLIT VAR country

Business Scenario
Print every data set in a library.
%printlib(lib=orion)

sashelp.vstabvw

Generating Data-Dependent Code

SAS metadata is available in the sashelp.vstabvw
dynamic view.
proc print data=sashelp.vstabvw;
where libname="ORION";
title "SASHELP.VSTABVW";
run;
Partial PROC PRINT Output
SASHELP.VSTABVW

Obs libname memname memtype

336 ORION CITY DATA

337 ORION CONTINENT DATA
338 ORION COUNTRY DATA
339 ORION COUNTY DATA
340 ORION CUSTOMER DATA
341 ORION CUSTOMER_DIM DATA
m105d10
63

Generating Data-Dependent Code

Step 1 Store data set names into macro variables.
%macro printlib(lib=WORK);
%let lib=%upcase(&lib);
data _null_;
set sashelp.vstabvw end=final;
where libname="&lib";
call symputx('dsn'||left(_n_),memname);
if final then call symputx('totaldsn',_n_);
run;
%put _local_;
%mend printlib;

%printlib(lib=orion)

m105d11a
64

 The statement below writes a macro’s local symbol table to the SAS log.

%put _local_;

Generating Data-Dependent Code

Partial SAS Log
PRINTLIB DSN1 AGESMOD
PRINTLIB DSN10 COUPONS
PRINTLIB DSN11 CUSTOMER
PRINTLIB DSN12 CUSTOMERDIM
PRINTLIB DSN13 CUSTOMERDIMMORE
PRINTLIB DSN14 CUSTOMERTYPE
PRINTLIB DSN15 CUSTOMER_DIM
PRINTLIB DSN16 CUSTOMER_TYPE
PRINTLIB DSN17 DAILY_SALES
PRINTLIB DSN18 DISCOUNT
PRINTLIB DSN19 EMPLOYEEADDRESSES
PRINTLIB DSN2 BIZLIST
PRINTLIB DSN20 EMPLOYEEDONATIONS
PRINTLIB DSN21 EMPLOYEEORGANIZATION
PRINTLIB DSN22 EMPLOYEEPAYROLL
PRINTLIB DSN23 EMPLOYEEPHONES
PRINTLIB DSN24 EMPLOYEE_PAYROLL
PRINTLIB DSN25 EUROPECUSTOMERS

m105d11a
65

Generating Data-Dependent Code

Step 2 Use a macro loop to print every data set
in the library.
%macro printlib(lib=WORK,obs=5);
%let lib=%upcase(&lib);
data _null_;
set sashelp.vstabvw end=final;
where libname="&lib";
call symputx('dsn'||left(_n_),memname);
if final then call symputx('totaldsn',_n_);
run;
%do i=1 %to &totaldsn;
proc print data=&lib..&&dsn&i(obs=&obs);
title "&lib..&&dsn&i Data Set";
run;
%end;
%mend printlib;
%printlib(lib=orion)
m105d11b
66

Generating Data-Dependent Code

Partial SAS Log
MPRINT(PRINTLIB): proc print data=ORION.CUSTOMER(obs=5);
MPRINT(PRINTLIB): title "ORION.CUSTOMER Data Set";
MPRINT(PRINTLIB): run;
NOTE: There were 5 observations read from the data set ORION.CUSTOMER.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

MPRINT(PRINTLIB): proc print data=ORION.CUSTOMER_DIM(obs=5);

MPRINT(PRINTLIB): title "ORION.CUSTOMER_DIM Data Set";
MPRINT(PRINTLIB): run;
NOTE: There were 5 observations read from the data set ORION.CUSTOMER_DIM.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

MPRINT(PRINTLIB): proc print data=ORION.CUSTOMER_TYPE(obs=5);

MPRINT(PRINTLIB): title "ORION.CUSTOMER_TYPE Data Set";
MPRINT(PRINTLIB): run;
NOTE: There were 5 observations read from the data set ORION.CUSTOMER_TYPE.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds
67 m105d11b

Exercises

Level 1

7. Using Macro Loops and Indirect References

a. Open the m105e07 program shown below into the Editor window.
proc means data=orion.order_fact sum mean maxdec=2;
where Order_Type=1;
var Total_Retail_Price CostPrice_Per_Unit;
title "Order Type: 1";
run;
b. Define a macro that generates a separate PROC MEANS step for each of the order types
in the orion.order_fact data set. The values of Order_Type are 1, 2, and 3.
c. The orion.lookup_order_type data set contains the variable START, which represents each
order type, and the variable LABEL, which describes each order type.

Partial Listing of orion.lookup_order_type

START LABEL

1 Retail Sale
2 Catalog Sale
3 Internet Sale

Modify the following DATA step to create a series of macro variables named type1-typen, where
n is the number of observations in the orion.lookup_order_type data set. Each macro variable
should receive the value of the data set variable LABEL. Place the modified DATA step into the
macro definition.
data _null_;
set orion.lookup_order_type;
run;
c. Modify the macro to do the following:
 create a macro variable endloop and populate it with the number of observations
in the orion.lookup_order_type data set
 use the endloop macro variable as the stop value for the macro DO loop
 use an indirect reference to typex in the TITLE statement

Level 2

8. Generating Data-Dependent Steps

a. Open the m105e08 program shown below into the Editor window. This program creates
a summary data set named customer_freq that summarizes the variable Total_Retail_Price
by Customer_ID, and then sorts by descending sum. CALL SYMPUTX creates a series
of macro variables named top1 through topx, where x is the value of the OBS parameter.

%macro tops(obs=3);
proc means data=orion.order_fact sum nway noprint;
var Total_Retail_Price;
class Customer_ID;
output out=customer_freq sum=sum;
run;

proc sort data=customer_freq;

by descending sum;
run;

data _null_;
set customer_freq(obs=&obs);
call symputx('top'||left(_n_), Customer_ID);
run;
%mend tops;

%tops()
%tops(obs=5)
b. Modify the macro to print a listing of the top x customers from the orion.customer_dim data set.
Display the variables Customer_ID, Customer_Name, and Customer_Type. Use a macro loop
to dynamically generate values for the WHERE statement based on the macro variables top1
through topx.

Partial SAS Log

MPRINT(TOPS): proc print data=orion.customer_dim noobs;
MPRINT(TOPS): where Customer_ID in ( 16 10 45);
MPRINT(TOPS): var Customer_ID Customer_Name Customer_Type;
MPRINT(TOPS): title "Top 3 Customers";
MPRINT(TOPS): run;

MPRINT(TOPS): proc print data=orion.customer_dim noobs;

MPRINT(TOPS): where Customer_ID in ( 16 10 45 2806 195);
MPRINT(TOPS): var Customer_ID Customer_Name Customer_Type;
MPRINT(TOPS): title "Top 5 Customers";
MPRINT(TOPS): run;

Challenge

9. Generating Multiple Macro Calls

a. Open the m105e09 program shown below into the Editor window. Submit the program and
review the result.
%macro memberlist(custtype);
proc print data=Orion.Customer_dim;
var Customer_Name Customer_ID Customer_Age_Group;
where Customer_Type="&custtype";
title "A List of &custtype";
run;
%mend memberlist;

%macro listall;
data _null_;
set orion.customer_type end=final;
call symputx('type'||left(_n_), Customer_Type);
if final then call symputx('n',_n_);
run;
%put _user_;
%mend listall;

%listall
b. Modify the Listall macro to call the Memberlist macro. The result of the macro call should
create a PROC PRINT step for each customer type. Use a macro loop and indirect references
to generate the appropriate macro calls.

5.4 Global and Local Symbol Tables

Objectives
 Explain the difference between global and local
symbol tables.
 Describe how the macro processor decides which
symbol table to use.
 Describe the concept of nested macros and the
hierarchy of symbol tables.

Global Symbol Table (Review)

The global symbol table is
 created during SAS initialization
 initialized with automatic macro variables
 deleted at the end of the session.

Global Symbol Table

Global macro variables can be created by any
of the following:
 %LET statement
 DATA step SYMPUTX routine
 PROC SQL SELECT statement INTO clause
 %GLOBAL statement

%GLOBAL Statement
General form of the %GLOBAL statement:

%GLOBAL macro-variable1 macro-variable2 . . . ;

 The %GLOBAL statement adds one or more macro

variables to the global symbol table with null values.
 It has no effect on variables already in the global table.
 It can be used anywhere in a SAS program.

Local Symbol Table

A local symbol table is
 created when a macro with a parameter list
is called or a local macro variable is created
during macro execution
 deleted when the macro finishes execution.

Macros that do not create local variables do not have

a local table.

Local Symbol Table

Local macro variables can be
 created at macro invocation (parameters)
 created during macro execution
 updated during macro execution
 referenced anywhere within the macro.

Local Symbol Table

Use local variables instead of global variables whenever
possible. Memory used by a local table can be reused
when the table is deleted following macro execution.
Local Symbol Table
Variable Value
parameter1 value1
parameter2 value2
. .
. .
. .
uservar1 value1
uservar2 value2

Local Symbol Table

Local macro variables can be created by the following
within a macro definition:
 %LET statement
 DATA step SYMPUTX routine
 PROC SQL SELECT statement INTO clause
 %LOCAL statement

 The DATA step SYMPUT routine can also create local macro variables, but it does not remove
leading and trailing blanks from the macro variable's value.

%LOCAL Statement
General form of the %LOCAL statement:

%LOCAL macro-variable1 macro-variable2 . . . ;

 The %LOCAL statement adds one or more macro

variables to the local symbol table with null values.
 It has no effect on variables already in the local table.
 It can appear only inside a macro definition.

%LOCAL Statement
Declare the index variable of a macro loop
as a local variable to prevent accidental contamination
of a like-named macro variable in the global table
or another local table.
%macro putloop;
%local i;
%do i=1 %to &numrows;
%put Country&i is &&country&i;
%end;
%mend putloop;

m105d13
80

Rules for Creating and Updating Variables

When the macro processor receives a request to create
or update a macro variable during macro execution,
the macro processor follows these rules:
%let macvar=value;

Does MACVAR already exist Update MACVAR in the local

in the local table? Yes table.
No
Does MACVAR already exist Update MACVAR in the global
in the global table? Yes table.
No
Create MACVAR in the local table.

Rules for Resolving Variables

To resolve a macro variable reference during macro
execution, the macro processor follows these rules:

&macvar

Does MACVAR exist in the

local table? Yes Retrieve from local table.
No
Does MACVAR exist in the
Yes Retrieve from global table.
global table?
No
Return tokens to word scanner. Issue a warning to SAS log:
Apparent symbolic reference MACVAR not resolved.

5.06 Short Answer Poll

How many local symbol tables are created when
macro A is called and begins to execute?
%macro a(value=);
%b
%mend a;
%macro b;
%put The value to write is: &value.;
%put _user_;
%mend b;

%a(value=Today is Monday)

Multiple Local Tables

Multiple local tables can exist concurrently during macro
execution.

Example: Define two macros. One calls the other.

%macro outer;
%local x;
%let x=1;
%inner
%mend outer;
%macro inner;
%local y;
%let y=&x;
%mend inner;

Multiple Local Tables

Create global macro variable X.
Global Table

%let x=0; X 0

%macro outer;
%local x;
%let x=1;
%inner
%mend outer;
%macro inner;
%local y;
%let y=&x;
%mend inner;

Multiple Local Tables

Call the Outer macro. When the %LOCAL statement
executes, a local table is created.
Global Table
%outer
X 0

%macro outer; Outer Local

Outer Table
Local Table
%local x;
%let x=1; X
%inner
%mend outer;
%macro inner;
%local y;
%let y=&x;
%mend inner;

Multiple Local Tables

The %LET statement updates local macro variable X.
Access to global macro variable X is blocked.
Global TableGlobal Table
X 0

%macro outer; Outer Local

Outer Table
Local Table
%local x;
%let x=1; X 1
%inner
%mend outer;
%macro inner;
%local y;
%let y=&x;
%mend inner;

What happens if the %LOCAL statement in the Outer

macro is omitted?
88

Multiple Local Tables

A nested macro call can create its own local symbol table
in addition to any other tables that might currently exist.

Global TableGlobal Table

%macro outer; X 0
%local x;
%let x=1;
%inner Outer
OuterLocal
LocalTable
Table
%mend outer;
%macro inner; X 1
%local y;
%let y=&x; Inner Local Table
%mend inner;
Y

Multiple Local Tables

The macro processor resolves a macro variable reference
by searching symbol tables in the reverse order in which
they were created, as follows:
1. current local table
2. previously created local tables
3. global table
Global Table
%macro outer; X 0
%local x;
%let x=1; Outer Local Table
%inner
%mend outer; X 1
%macro inner;
%local y; Inner Local Table
%let y=&x;
%mend inner;
Y 1

The global variable X is not available to the Inner macro.

Multiple Local Tables

When the Inner macro finishes execution, its local table
is deleted. Control returns to the Outer macro.

Global Table
%macro outer; X 0
%local x;
%let x=1; Outer Local Table
%inner
%mend outer; X 1
%macro inner;
%local y;
%let y=&x;
%mend inner;

Multiple Local Tables

When the Outer macro finishes execution, its local table
is deleted. Only the global table remains.

Global Table
%macro outer; X 0
%local x;
%let x=1;
%inner
%mend outer;
%macro inner;
%local y;
%let y=&x;
%mend inner;

Setup for the Poll

The Numobs macro creates a macro variable with
the number of observations in a SAS data set.
%macro numobs(dsn);
options nonotes;
data _null_;
call symputx('num', number);
stop;
set &dsn nobs=number;
run;
options notes;
%mend numobs;

m105d14
93

5.07 Short Answer Poll

What is wrong with this attempt?
%macro numobs(dsn);
options nonotes;
data _null_;
call symputx('num', number);
stop;
set &dsn nobs=number;
run;
options notes;
%mend numobs;

%numobs(orion.order_fact)
%put ---> &num observations;

m105d14
94

Multiple Local Tables

The SYMPUTX routine accepts an optional scope argument,
which specifies where to store the macro variable.
%macro numobs(dsn);
options nonotes;
data _null_;
call symputx('num', number,'G');
stop;
set &dsn CALL
nobs=number;
SYMPUTX(macro-variable, text <,scope>);
run;
options notes;
%mend numobs;

SAS Log
1831 %numobs(orion.orders)
1832 %put ---> &num observations;
---> 490 observations

m105d14
96

 The scope argument is recommended any time that the SYMPUTX routine is used within a macro
definition.

 The STOP statement stops DATA step execution. The special variable NUMBER is populated
during compile time.

Multiple Local Tables

Example: The Printds macro calls the Numobs macro.

m105d14
97

Multiple Local Tables

SAS Output
First 10 of 490 observations from ORION.ORDERS

Order_ Order_ Delivery_

Obs Order_ID Type Employee_ID Customer_ID Date Date

1 1230058123 1 121039 63 11JAN2007 11JAN2007

2 1230080101 2 99999999 5 15JAN2007 19JAN2007
3 1230106883 2 99999999 45 20JAN2007 22JAN2007
4 1230147441 1 120174 41 28JAN2007 28JAN2007
5 1230315085 1 120134 183 27FEB2007 27FEB2007
6 1230333319 2 99999999 79 02MAR2007 03MAR2007
7 1230338566 2 99999999 23 03MAR2007 08MAR2007
8 1230371142 2 99999999 45 09MAR2007 11MAR2007
9 1230404278 1 121059 56 15MAR2007 15MAR2007
10 1230440481 1 120149 183 22MAR2007 22MAR2007

m105d14
98

Multiple Local Tables

SAS Output
All 7 observations from ORION.COUNTRY

Country_ Continent_ Country_Former

Obs Country Country_Name Population ID ID Name

1 AU Australia 20,000,000 160 96

2 CA Canada . 260 91
3 DE Germany 80,000,000 394 93 East/West Germany
4 IL Israel 5,000,000 475 95
5 TR Turkey 70,000,000 905 95
6 US United States 280,000,000 926 91
7 ZA South Africa 43,000,000 801 94

m105d14
99

Exercises

Level 1

10. Understanding Symbol Tables

Without submitting the programs, identify in which symbol table the macro variable dog is located.

 Assume that each example is submitted in a new SAS session.

a. ______________________________________________________________________________
%let dog=Paisley;
%macro whereisit;
%put My dog is &dog;
%mend whereisit;

%whereisit
b. ______________________________________________________________________________
%macro whereisit;
%let dog=Paisley;
%put My dog is &dog;
%mend whereisit;

%whereisit
c. ______________________________________________________________________________
%macro whereisit(dog);
%put My dog is &dog;
%mend whereisit;

%whereisit(Paisley)

Level 2

11. Controlling Macro Variable Storage

a. Open the m105e11 program shown below into the Editor window.
%macro varscope;
data _null_;
set orion.customer_type end=final;
call symputx('localtype'||left(_n_), Customer_Type);
if final then call symputx('localn',_n_);
run;
%put _user_;
%mend varscope;

%varscope
b. Modify the program so that all macro variables that are created in the DATA step are stored in the
local symbol table.

c. Modify the program by adding the following statement before the DATA step and removing the
scope specification in the SYMPUTX routine:
%local x;
In which symbol table are the macro variables stored?
____________________________________

d. Modify the program so that all macro variables that are created in the DATA step are stored in the
global symbol table.

Challenge

12. Creating Multiple Symbol Tables

a. Open the cleanup program and submit the macro to delete all global macro variables.
b. Open the m105e12 program shown below into the Editor window.
%macro createmacvar;
data _null_;
set orion.lookup_order_type end=last;
call symputx('type'||left(start), label,'L');
if last then call symputx('endloop', _n_,'L');
run;
%mend createmacvar;

%macro sumreport;
%createmacvar
%do num=1 %to &endloop;
proc means data=orion.order_fact sum mean maxdec=2;
where Order_Type = &num;
var Total_Retail_Price CostPrice_Per_Unit;
title "Summary Report for &&type&num";
run;
%end;
%mend sumreport;

%sumreport
Submit the program. Review and describe the results.
___________________________________________________________________________
___________________________________________________________________________
___________________________________________________________________________

c. Correct the program so that the Sumreport macro executes correctly and does not create any
global macro variables. Verify that the title resolves properly. In addition, add an s to the end
of the type description in the title.
PROC MEANS Output
Summary Report for Retail Sales

The MEANS Procedure

Variable Label Sum Mean
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Total_Retail_Price Total Retail Price for This Product 44654.56 137.82
CostPrice_Per_Unit Cost Price Per Unit 11730.73 36.21
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

Summary Report for Catalog Sales

The MEANS Procedure

Variable Label Sum Mean

ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Total_Retail_Price Total Retail Price for This Product 33931.35 199.60
CostPrice_Per_Unit Cost Price Per Unit 8718.45 51.29
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

Summary Report for Internet Sales

The MEANS Procedure

Variable Label Sum Mean
ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ
Total_Retail_Price Total Retail Price for This Product 21491.55 174.73
CostPrice_Per_Unit Cost Price Per Unit 5356.95 43.55
Ƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒƒ

5.5 Solutions
Solutions to Exercises
1. Conditionally Processing Complete Statements
a. Open the program into the Editor window.
b. Modify the macro to test the CUSTTYPE parameter. If the value is null, insert VAR and TITLE
statements into the PROC PRINT step. If the value is not null, insert WHERE, VAR, and TITLE
statements into the PROC PRINT step.
%macro listing(custtype);
proc print data=orion.customer noobs;
%if &custtype= %then %do;
var Customer_ID Customer_Name Customer_Type_ID;
title "All Customers";
%end;
%else %do;
where Customer_Type_ID=&custtype;
var Customer_ID Customer_Name;
title "Customer Type: &custtype";
%end;
run;
%mend listing;
c. Resubmit the macro definition using a null value and a valid value for CUSTTYPE.
Partial SAS Log
36 %listing()
MPRINT(LISTING): proc print data=orion.customer noobs;
MPRINT(LISTING): var Customer_ID Customer_Name Customer_Type_ID;
MPRINT(LISTING): title "All Customers";
MPRINT(LISTING): run;

NOTE: There were 77 observations read from the data set ORION.CUSTOMER.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.23 seconds
cpu time 0.01 seconds

37 %listing(2010)
MPRINT(LISTING): proc print data= orion.customer noobs;
MPRINT(LISTING): where Customer_Type_ID=2010;
MPRINT(LISTING): var Customer_ID Customer_Name;
MPRINT(LISTING): title "Customer Type: 2010";
MPRINT(LISTING): run;

NOTE: There were 5 observations read from the data set ORION.CUSTOMER.
WHERE Customer_Type_ID=2010;
NOTE: PROCEDURE PRINT used (Total process time):
real time 1.13 seconds
cpu time 0.01 seconds

2. Debugging a Macro
a. Open the program into the Editor window.
b. Change SATURDAY to today’s value and submit the program.
c. Change today’s value to the day that you launched SAS. Enter the day in mixed case, such
as Tuesday.

3. Debugging a Macro
a. Open the program into the Editor window.
b. The %STR function is required to interpret OR as plain text instead of a logical operator.
%macro where(state);
%if &state=%str(OR)
%then %put Oregon;
%else %put Wherever;
%mend where;
4. Validating a Macro Parameter
a. Open the program into the Editor window and submit it.
b. Use %IF to validate the TYPE parameter. The MINOPERATOR option in the %MACRO
statement activates the IN operator.
%macro custtype(type)/minoperator;
%let type=%upcase(&type);
%if &type in GOLD INTERNET %then %do;
proc print data=orion.customer_dim;
var Customer_Group Customer_Name Customer_Gender
Customer_Age;
where upcase(Customer_Group) contains "&type";
title "&type Customers";
run;
%end;
%else %do;
%put ERROR: Invalid TYPE: &type..;
%put ERROR: Valid TYPE values are INTERNET or GOLD.;
%end;
%mend custtype;
c. Resubmit the macro definition and call the macro using valid and invalid parameter values.
Partial SAS Log
248 %custtype(internet)

NOTE: There were 8 observations read from the data set ORION.CUSTOMER_DIM.
WHERE UPCASE(Customer_Group) contains 'INTERNET';
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

249 %custtype(silver)

ERROR: Invalid TYPE: SILVER.

d. Modify the macro to test whether TYPE is null. If so, do not execute PROC PRINT. Instead,
the macro should write messages to the SAS log.
%macro custtype(type)/minoperator;
%if &type= %then %do;
%put ERROR: Missing TYPE.;
%put ERROR: Valid TYPE values are INTERNET or GOLD.;
%end;
%else %do;
%let type=%upcase(&type);
%if &type in GOLD INTERNET %then %do;
proc print data=orion.customer_dim;
var Customer_Group Customer_Name Customer_Gender
Customer_Age;
where upcase(Customer_Group) contains "&type";
title "&type Customers";
run;
%end;
%else %do;
%put ERROR: Invalid TYPE: &type..;
%put ERROR: Valid TYPE values are INTERNET or GOLD.;
%end;
%end;
%mend custtype;
e. Resubmit the macro definition with a null parameter value, a valid value, and an invalid value.
Partial SAS Log
272 %custtype()
ERROR: Missing TYPE.
ERROR: Valid TYPE values are INTERNET or GOLD.
273 %custtype(internet)

274 %custtype(silver)
ERROR: Invalid TYPE: SILVER.
ERROR: Valid TYPE values are INTERNET or GOLD.

5. Validating a Macro Parameter Using Data-Driven Techniques

a. Open the program into the Editor window and submit it.
b. Modify the macro definition to validate CUSTTYPE against a data-driven list.
1) Use the SQL procedure to create the macro variable IDLIST that contains the unique values
of the variable Customer_Type_ID in the orion.customer_type data set.
2) Validate the CUSTTYPE parameter against IDLIST.

3) If CUSTTYPE is in IDLIST, execute PROC PRINT. Otherwise, do not execute PROC PRINT.
Instead, the macro should write this message to the SAS log:
Partial SAS Log
ERROR: Invalid CUSTTYPE.
Valid CUSTTYPEs: 1010 1020 1030 1040 2010 2020 2030 3010.

%macro listing(custtype) / minoperator;

%if &custtype= %then %do;
proc print data= orion.customer noobs;
var Customer_ID Customer_Name Customer_Type_ID;
title "All Customers";
run;
%end;
%else %do;
proc sql noprint;
select distinct Customer_Type_ID
into :IDlist separated by ' '
from orion.customer_type;
quit;
%if &custtype in &idlist %then %do;
proc print data= orion.customer noobs;
where Customer_Type_ID =&custtype;
var Customer_ID Customer_Name;
title "Customer Type: &custtype";
run;
%end;
%else %do;
%put ERROR: Invalid CUSTTYPE.;
%put ERROR- Valid CUSTTYPEs: &IDlist..;
%end;
%end;
%mend listing;
c. Resubmit the macro definition and call the macro with a null parameter value, a valid value,
and an invalid value.
Partial SAS Log
411 %listing()
MPRINT(LISTING): proc print data= orion.customer noobs;
MPRINT(LISTING): var Customer_ID Customer_Name Customer_Type_ID;
MPRINT(LISTING): title "All Customers";
MPRINT(LISTING): run;

NOTE: There were 77 observations read from the data set ORION.CUSTOMER.
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

412 %listing(1020)
MPRINT(LISTING): proc sql noprint;

MPRINT(LISTING): select distinct Customer_Type_ID into :IDlist separated by ' ' from
orion.customer_type;
MPRINT(LISTING): quit;
NOTE: PROCEDURE SQL used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

MPRINT(LISTING): proc print data= orion.customer noobs;

MPRINT(LISTING): where Customer_Type_ID =1020;
MPRINT(LISTING): var Customer_ID Customer_Name;
MPRINT(LISTING): title "Customer Type: 1020";
MPRINT(LISTING): run;
NOTE: There were 17 observations read from the data set ORION.CUSTOMER.
WHERE Customer_Type_ID=1020;
NOTE: PROCEDURE PRINT used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

413 %listing(9999)
MPRINT(LISTING): proc sql noprint;
MPRINT(LISTING): select distinct Customer_Type_ID into :IDlist separated by ' ' from
orion.customer_type;
MPRINT(LISTING): quit;
NOTE: PROCEDURE SQL used (Total process time):
real time 0.00 seconds
cpu time 0.00 seconds

ERROR: Invalid CUSTTYPE.

Valid CUSTTYPEs: 1010 1020 1030 1040 2010 2020 2030 3010.

6. Validating a Macro Parameter

a. Open the program into the Editor window and submit it.
b. Modify the macro to validate parameters according to the following requirements:
Macro Variable Possible Values
decimals 0 to 4
order INTERNAL, FREQ

c. Create an additional macro variable named numerrors that accumulates the number
of parameter errors.
1) Execute PROC MEANS only if numerrors is zero.
2) The macro should write messages to the SAS log when parameters are invalid.
%macro salarystats(decimals=2,order=internal)/minoperator;
%let numerrors=0;
%if not (&decimals in 0 1 2 3 4) %then %do;
%let numerrors=%eval(&numerrors+1);
%put ERROR: Invalid DECIMALS parameter: &decimals..;
%put ERROR- Valid DECIMALS values are 0 to 4.;
%end;
%let order=%upcase(&order);
%if not (&order in INTERNAL FREQ) %then %do;

%let numerrors=%eval(&numerrors+1);
%put ERROR: Invalid ORDER parameter: &order..;
%put ERROR- Valid ORDER values are INTERNAL or FREQ.;
%end;
%if &numerrors=0 %then %do;
options nolabel;
title 'Salary Stats';
proc means data=orion.staff maxdec=&decimals
order=&order;
where job_title contains 'Sales';
var salary;
class job_title;
run;
title;
%end;
%else %put ERROR: &numerrors errors. Code not submitted.;
%mend salarystats;

%salarystats()
%salarystats(decimals=5,order=fudge)
7. Using Macro Loops and Indirect References
a. Open the program into the Editor window.
b. Define a macro that generates a separate PROC MEANS step for each of the order types
in the orion.order_fact data set. The values of Order_Type are 1, 2, and 3.
%macro sumreport;
%do num=1 %to 3;
proc means data=orion.order_fact sum mean maxdec=2;
where Order_Type=&num;
var Total_Retail_Price CostPrice_Per_Unit;
title "Order Type: &num";
run;
%end;
%mend sumreport;

%sumreport
c. Insert the provided DATA step to create the series of macro variables.
d. Modify the macro to use the following:
 the endloop macro variable as the stop value for the iterative DO loop
 an indirect reference to typex in the TITLE statement
%macro sumreport;
data _null_;
set orion.lookup_order_type end=last;
call symputx('type'||left(_n_), label);
if last then call symputx('endloop', _n_);
run;

%do num=1 %to &endloop;

proc means data=orion.order_fact sum mean maxdec=2;
where Order_Type=&num;
var Total_Retail_Price CostPrice_Per_Unit;
title "Order Type: &&type&num";
run;
%end;
%mend sumreport;

%sumreport
8. Generating Data-Dependent Steps
a. Open the program into the Editor window.
b. Modify the macro to print a listing of the topx customers from the orion.customer_dim data set.
Display the variables Customer_ID, Customer_Name, and Customer_Type.
Use a macro loop to dynamically generate values for the WHERE statement based on the macro
variables top1 through topx.
%macro tops(obs=3);
proc means data=orion.order_fact sum nway noprint;
var Total_Retail_Price;
class Customer_ID;
output out=customer_freq sum=sum;
run;

proc sort data=customer_freq;

by descending sum;
run;

data _null_;
set customer_freq(obs=&obs);
call symputx('top'||left(_n_), Customer_ID);
run;

proc print data=orion.customer_dim noobs;

where Customer_ID in (%do num=1 %to &obs; &&top&num
%end;);
var Customer_ID Customer_Name Customer_Type;
title "Top &obs Customers";
run;
%mend tops;

%tops()
%tops(obs=5)

9. Generating Multiple Macro Calls

a. Review the output generated by the Listall macro in program m105e09.
GLOBAL N 8
GLOBAL TYPE1 Orion Club members inactive
GLOBAL TYPE2 Orion Club members low activity
GLOBAL TYPE3 Orion Club members medium activity
GLOBAL TYPE4 Orion Club members high activity
GLOBAL TYPE5 Orion Club Gold members low activity
GLOBAL TYPE6 Orion Club Gold members medium activity
GLOBAL TYPE7 Orion Club Gold members high activity
GLOBAL TYPE8 Internet/Catalog Customers

b. Modify the Listall macro to call the Memberlist macro. The result of the macro call should
create a PROC PRINT step for each customer type. Use a macro loop and indirect references
to generate the appropriate macro calls.
%macro memberlist(custtype);
proc print data=Orion.Customer_dim;
var Customer_Name Customer_ID Customer_Age_Group;
where Customer_Type="&custtype";
title "A List of &custtype";
run;
%mend memberlist;

%macro listall;
data _null_;
set orion.customer_type end=final;
call symputx('type'||left(_n_), Customer_Type);
if final then call symputx('n',_n_);
run;
%do num=1 %to &n;
%memberlist(&&type&num)
%end;
%mend listall;

%listall
10. Understanding Symbol Tables
Without submitting the programs, identify in which symbol table the macro variable dog is located.

 Assume that each example is submitted in a new SAS session.

a. Because the %LET statement is outside the macro definition, the macro variable dog is stored
in the global symbol table.
%let dog=Paisley;
%macro whereisit;
%put My dog is &dog;
%mend whereisit;
%whereisit

b. Because the %LET statement is inside the macro definition, the macro variable dog is stored
in the local symbol table.
%macro whereisit;
%let dog=Paisley;
%put My dog is &dog;
%mend whereisit;
%whereisit
c. Because DOG is a macro parameter, it is stored in the local symbol table.
%macro whereisit(dog);
%put My dog is &dog;
%mend whereisit;
%whereisit(Paisley)
11. Controlling Macro Variable Storage
a. Open the program into the Editor window.
b. Specifying L as the third argument of the SYMPUTX routine stores the macro variable in the
local symbol table.
%macro varscope;
data _null_;
set orion.customer_type end=final;
call symputx('localtype'||left(_n_), Customer_Type,'L');
if final then call symputx('localn',_n_,'L');
run;
%put _user_;
%mend varscope;
%varscope
c. By adding the %LOCAL statement and removing the scope specification, the SYMPUTX routine
creates local macro variables.
2 %macro varscope;
3 %local x;
4 data _null_;
5 set orion.customer_type end=final;
6 call symputx('localtype'||left(_n_), Customer_Type);
7 if final then call symputx('localn',_n_);
8 run;
9 %put _user_;
10 %mend varscope;
11
12 %varscope

NOTE: Numeric values have been converted to character values at the places given by:
(Line):(Column).
1:94
NOTE: There were 8 observations read from the data set ORION.CUSTOMER_TYPE.
NOTE: DATA statement used (Total process time):
real time 0.01 seconds
cpu time 0.01 seconds

VARSCOPE LOCALN 8
VARSCOPE LOCALTYPE1 Orion Club members inactive
VARSCOPE LOCALTYPE2 Orion Club members low activity
VARSCOPE LOCALTYPE3 Orion Club members medium activity
VARSCOPE LOCALTYPE4 Orion Club members high activity
VARSCOPE LOCALTYPE5 Orion Club Gold members low activity
VARSCOPE LOCALTYPE6 Orion Club Gold members medium activity
VARSCOPE LOCALTYPE7 Orion Club Gold members high activity
VARSCOPE LOCALTYPE8 Internet/Catalog Customers
VARSCOPE X

d. Specifying G as the third argument of the SYMPUTX routine stores the macro variables
in the global symbol table.
%macro varscope;
%local x;
data _null_;
set orion.customer_type end=final;
call symputx('localtype'||left(_n_), Customer_Type,'G');
if final then call symputx('localn',_n_,'G');
run;
%put _user_;
%mend varscope;
%varscope
12. Creating Multiple Symbol Tables
a. Open the cleanup program and submit the macro.
b. Open the m105e12 program into the Editor window.
c. Correct the program so that the Sumreport macro executes correctly and does not create any
global macro variables. Verify that the title resolves properly. In addition, add an s to the end
of the type description in the title.
The macro variables endloop and typex are stored in the local symbol table for the
Createmacvar macro and are not available to the Sumreport macro. Delete the scope argument
from the SYMPUTX routine and add the %LOCAL statement to the Sumreport macro to force
the macro variables into the local symbol table for the Sumreport macro.
%macro createmacvar;
data _null_;
set orion.lookup_order_type end=last;
call symputx('type'||left(start), label);
if last then call symputx('endloop', _n_);
run;
%mend createmacvar;

%macro sumreport;
%local type1 type2 type3 endloop num;
%createmacvar
%do num=1 %to &endloop;
proc means data=orion.order_fact sum mean maxdec=2;
where Order_Type = &num;
var Total_Retail_Price CostPrice_Per_Unit;
title "Summary Report for &&type&num..s";
run;
%end;
%mend sumreport;

%sumreport

Solutions to Student Activities (Polls/Quizzes)

5.01 Short Answer Poll – Correct Answer

Submit the program m105a01. What error do you see
in the log?

Partial SAS Log

514 %macro reports;
515 %daily
516 %if &sysday=Friday then %weekly;
ERROR: Expected %THEN statement not found. A dummy macro will be
compiled.
517 %mend reports;

If a macro definition contains macro language syntax

errors, error messages are written to the SAS log
and a dummy (nonexecutable) macro is created.

Idea Exchange
What is the difference between %IF-%THEN
and IF-THEN?

24
24

5.02 Short Answer Poll – Correct Answer

Instead of using multiple OR operators, which SAS
comparison operator compares a value to a list of values?

if Country in ('FR' 'CA' 'DE' 'ZA') then do;

The IN operator is a SAS comparison operator.

5.03 Short Answer Poll – Correct Answer

5.04 Multiple Choice Poll – Correct Answer

Which statement correctly creates an index variable
named i?
a. %do &i=1 %to 10;
b. %do &i=1 to 10;
c. %do i=1 %to 10;
d. %do i=1 to 10;

5.05 Short Answer Poll – Correct Answer

&&VAL&COUNT  &VAL7  ZA

5.06 Short Answer Poll – Correct Answer

How many local symbol tables are created when
macro A is called and begins to execute?
%macro a(value=);
%b
%mend a;
%macro b;
%put The value to write is: &value.;
%put _user_;
%mend b;

%a(value=Today is Monday)
One. The parameter list in macro A causes a local
symbol table to be created. No local symbol table
is created when macro B is called because macro
84 B does not create local variables.

5.07 Short Answer Poll – Correct Answer

What is wrong with this attempt?
352 %macro numobs(dsn);
353 options nonotes;
354 data _null_;
355 call symputx('num', number);
356 stop;
357 set &dsn nobs=number;
358 run;
359 options notes;
360 %mend numobs;
NOTE: The macro NUMOBS completed compilation without errors.
9 instructions 324 bytes.
361
362 %numobs(orion.order_fact)
363 %put ---> &num observations;
WARNING: Apparent symbolic reference NUM not resolved.
---> &num observations

The macro variable num was placed in the Numobs

macro’s local table and deleted when the Numobs
95 macro finished execution. m105d14

6.1 SAS Resources.......................................................................................................... 6-3

6.2 Beyond This Course.................................................................................................. 6-6

6-2 Chapter 6 Learning More

6.1 SAS Resources

Objectives
 Identify areas of support that SAS offers.

Education
Comprehensive training to deliver greater value
to your organization

 More than 200 course offerings

 World-class instructors
 Multiple delivery methods: instructor-led
and self-paced
 Training centers around the world

http://support.sas.com/training/

SAS Books
Convenient. Practical. Enlightening.
Valuable insight with solid results.
Available in a variety of formats
to best meet your needs:
 hard-copy books
 e-books
 PDF

5 www.sas.com/store/books

SAS Global Certification Program

SAS offers several globally recognized certifications.

 Computer-based
certification exams –
typically 60-70 questions and
2-3 hours in length
 Preparation materials and
practice exams available
 Worldwide directory of
SAS Certified Professionals

http://support.sas.com/certify/

Support
SAS provides a variety of self-help and assisted-help
resources.

 SAS Knowledge Base

 Downloads and hot fixes
 License assistance
 SAS discussion forums
 SAS Technical Support

http://support.sas.com/techsup/

User Groups
SAS supports many local, regional, international,
and special-interest SAS user groups.

 SAS Global Forum

 Online SAS Community:
http://www.sascommunity.org/wiki/

http://support.sas.com/usergroups/

6.2 Beyond This Course

Objectives
 Identify the next course that follows this course.

Next Steps
To learn advanced macro techniques, enroll in the following
course:

SAS® Macro Language 2: Advanced Techniques

Next Steps
In addition, there are prerecorded short technical
discussions and demonstrations called e-lectures.

http://support.sas.com/training/

7.1 Program Flow ............................................................................................................ 7-3

7-2 Chapter 7 Supplemental Materials

7.1 Program Flow

Program Flow
DATA Step Procedure Global Statement SQL
Compiler Parser Parser Compiler

Symbol Table
W ord Scanner Macro Processor

Macro Library

Input Stack

Getting Started With Sharepoint Framework Development Using Typescript PNP Js
No ratings yet
Getting Started With Sharepoint Framework Development Using Typescript PNP Js
160 pages
A Complete Tutorial On SAS Macros For Faster Data Manipulation
No ratings yet
A Complete Tutorial On SAS Macros For Faster Data Manipulation
36 pages
SAS Programming 1: Essentials: Course Notes
No ratings yet
SAS Programming 1: Essentials: Course Notes
10 pages
LWPRG1
100% (2)
LWPRG1
548 pages
LWMC1V2 001
100% (2)
LWMC1V2 001
296 pages
Macros Básicas de SAS
No ratings yet
Macros Básicas de SAS
22 pages
Why Use Macros
No ratings yet
Why Use Macros
38 pages
Complete Tutorial Sas Macros Faster Data Manipulation
No ratings yet
Complete Tutorial Sas Macros Faster Data Manipulation
15 pages
SAS+Programming+Fast+Track+ Cource+Notes
No ratings yet
SAS+Programming+Fast+Track+ Cource+Notes
406 pages
Intro To SAS Macro Language Programming Seminar Description Aug 2012
No ratings yet
Intro To SAS Macro Language Programming Seminar Description Aug 2012
1 page
Mcrolref
No ratings yet
Mcrolref
420 pages
SAS Macro
No ratings yet
SAS Macro
19 pages
SAS Macro Language Reference PDF
No ratings yet
SAS Macro Language Reference PDF
385 pages
SAS Macro Language Reference
No ratings yet
SAS Macro Language Reference
385 pages
Week 12 - Chapter 25
No ratings yet
Week 12 - Chapter 25
77 pages
Day5 PPT
No ratings yet
Day5 PPT
25 pages
Adv. SAS (Macro)
No ratings yet
Adv. SAS (Macro)
25 pages
SAS Training - Macro
No ratings yet
SAS Training - Macro
25 pages
SAS Macro Programming Made Easy 2nd Edition Michele Burlew - Quickly download the ebook to read anytime, anywhere
100% (1)
SAS Macro Programming Made Easy 2nd Edition Michele Burlew - Quickly download the ebook to read anytime, anywhere
58 pages
SAS Macro Programming Made Easy 2nd Edition Michele Burlew download
No ratings yet
SAS Macro Programming Made Easy 2nd Edition Michele Burlew download
46 pages
Sas Macros
No ratings yet
Sas Macros
247 pages
Hands-On SAS Macro Programming Tips and Techniques
No ratings yet
Hands-On SAS Macro Programming Tips and Techniques
9 pages
Macros
No ratings yet
Macros
348 pages
SAS Macro Basics
No ratings yet
SAS Macro Basics
10 pages
Lisa Sanbonmatsu, Consultant, Somerville, MA: Moving From Macro Variables To Macros
No ratings yet
Lisa Sanbonmatsu, Consultant, Somerville, MA: Moving From Macro Variables To Macros
9 pages
SAS Macro Programming Tips, Tricks and Techniques
No ratings yet
SAS Macro Programming Tips, Tricks and Techniques
9 pages
SAS Interview Questions:macros
No ratings yet
SAS Interview Questions:macros
6 pages
Writing Cleaner and More Powerful SAS Code Using Macros: Patrick Breheny
100% (1)
Writing Cleaner and More Powerful SAS Code Using Macros: Patrick Breheny
42 pages
SAS Macros Interview Questions
No ratings yet
SAS Macros Interview Questions
13 pages
SAS Macro Programming Made Easy 2nd Edition Michele Burlew download
No ratings yet
SAS Macro Programming Made Easy 2nd Edition Michele Burlew download
49 pages
SAS Macro Tutorial Imp
No ratings yet
SAS Macro Tutorial Imp
27 pages
SAS Macro Workshop
No ratings yet
SAS Macro Workshop
22 pages
Sas Macro Interview Question
No ratings yet
Sas Macro Interview Question
7 pages
SAS Macro Workshop Notes
No ratings yet
SAS Macro Workshop Notes
17 pages
SAS Macro Programming Tips and Techniques: Kirk Paul Lafler, Software Intelligence Corporation, Spring Valley, California
No ratings yet
SAS Macro Programming Tips and Techniques: Kirk Paul Lafler, Software Intelligence Corporation, Spring Valley, California
8 pages
SAS Tipsheet Macro
No ratings yet
SAS Tipsheet Macro
2 pages
BaseSas and Macros Interview Quetions
No ratings yet
BaseSas and Macros Interview Quetions
14 pages
Macros Ejemplos
No ratings yet
Macros Ejemplos
10 pages
Sas Macros Final
No ratings yet
Sas Macros Final
75 pages
Unlocking Statistics for the Social Sciences
From Everand
Unlocking Statistics for the Social Sciences
Norma Sinclair
No ratings yet
SAS Macro Language: Reference: Getting Started With The Macro Facility
No ratings yet
SAS Macro Language: Reference: Getting Started With The Macro Facility
8 pages
Macro
No ratings yet
Macro
6 pages
Software Patterns Made Easy
From Everand
Software Patterns Made Easy
Justice Nanhou
No ratings yet
Blog Smarter, Not Harder: SEO, Blogging, and AI Strategies to Skyrocket Your Traffic
From Everand
Blog Smarter, Not Harder: SEO, Blogging, and AI Strategies to Skyrocket Your Traffic
Jay Nans
No ratings yet
ChatGPT for Business: Strategies for Success
From Everand
ChatGPT for Business: Strategies for Success
Matthew C. Smith
1/5 (1)
Breaking Barriers: S.T.E.M Mentorship in Business
From Everand
Breaking Barriers: S.T.E.M Mentorship in Business
Matthew C. Smith
No ratings yet
Macro Basic
No ratings yet
Macro Basic
6 pages
Epoch Forumotions in t67 General Sas Macro Interview Questio
No ratings yet
Epoch Forumotions in t67 General Sas Macro Interview Questio
8 pages
SAS Macro
No ratings yet
SAS Macro
7 pages
When Best To Use The %LET Statement, The SYMPUT Routine, or The INTO Clause To Create Macro Variables
No ratings yet
When Best To Use The %LET Statement, The SYMPUT Routine, or The INTO Clause To Create Macro Variables
18 pages
Base SAS 9.3 Utilities: Reference
No ratings yet
Base SAS 9.3 Utilities: Reference
54 pages
Gray Hat Hacking the Ethical Hacker's
From Everand
Gray Hat Hacking the Ethical Hacker's
Çağatay Şanlı
5/5 (1)
10K Blueprint
From Everand
10K Blueprint
Cian O Farrell
5/5 (2)
Using Macro in A Spreadsheet
No ratings yet
Using Macro in A Spreadsheet
1 page
SAS Macro Tutorial
100% (1)
SAS Macro Tutorial
6 pages
Macro Chapter2
No ratings yet
Macro Chapter2
42 pages
Advanced Programming Sas94
No ratings yet
Advanced Programming Sas94
4 pages
Betancourt SAS Macro Basics
No ratings yet
Betancourt SAS Macro Basics
31 pages
Macro Variables
No ratings yet
Macro Variables
4 pages
Mastering Python Advanced Concepts and Practical Applications
From Everand
Mastering Python Advanced Concepts and Practical Applications
Aissa Younes
No ratings yet
5 Ways To Create Macro Variables
No ratings yet
5 Ways To Create Macro Variables
12 pages
TLE 6 Activity Sheet 2 Agriculture Orchard Growers
No ratings yet
TLE 6 Activity Sheet 2 Agriculture Orchard Growers
3 pages
Programming in C, 2/e: March 2020
No ratings yet
Programming in C, 2/e: March 2020
3 pages
Grade 11: Fundamentals OF Communication Oral Communication in Context
No ratings yet
Grade 11: Fundamentals OF Communication Oral Communication in Context
14 pages
Rapport2020with Doker18
No ratings yet
Rapport2020with Doker18
78 pages
Book 7
No ratings yet
Book 7
241 pages
Articles
No ratings yet
Articles
2 pages
Full A Treatise On Universal Algebra With Applications Alfred North Whitehead PDF All Chapters
100% (13)
Full A Treatise On Universal Algebra With Applications Alfred North Whitehead PDF All Chapters
60 pages
Unit V
No ratings yet
Unit V
42 pages
Biography of Hadrat Syed Dawood Hussain Shirazi
No ratings yet
Biography of Hadrat Syed Dawood Hussain Shirazi
17 pages
Avocent AutoView AV2108 - AV2216 - AV3108 - AV3216 InstallerUser Manual
No ratings yet
Avocent AutoView AV2108 - AV2216 - AV3108 - AV3216 InstallerUser Manual
68 pages
5.hindi - (Language - 1st Sem) - 23-24
No ratings yet
5.hindi - (Language - 1st Sem) - 23-24
5 pages
The Tragic Hero: Poetics
No ratings yet
The Tragic Hero: Poetics
2 pages
G11 Recollection Draft
No ratings yet
G11 Recollection Draft
10 pages
DOC-20240815-WA0002.
No ratings yet
DOC-20240815-WA0002.
39 pages
Online Applicants List and Schedule For English Test
No ratings yet
Online Applicants List and Schedule For English Test
86 pages
DE Community ? ?? ?ipad
No ratings yet
DE Community ? ?? ?ipad
1 page
Kumbabishegam Arikkai 2025
No ratings yet
Kumbabishegam Arikkai 2025
4 pages
MIS 370 Syllabus
No ratings yet
MIS 370 Syllabus
3 pages
Função de Determinação de Preço
No ratings yet
Função de Determinação de Preço
18 pages
Group Theory Mcqs 1 To 20
No ratings yet
Group Theory Mcqs 1 To 20
32 pages
The Boys S03E08 WEBRip x264-ION10 SRT
No ratings yet
The Boys S03E08 WEBRip x264-ION10 SRT
72 pages
Parables-of-Jesus
No ratings yet
Parables-of-Jesus
22 pages
LM Fusion251 Lick9 Tab
0% (1)
LM Fusion251 Lick9 Tab
3 pages
12393856063 (1)
No ratings yet
12393856063 (1)
2 pages
WINSEMFRE2024-25_STS1009_TH_AP2024255000503_2025-01-30_Reference-Material-I
No ratings yet
WINSEMFRE2024-25_STS1009_TH_AP2024255000503_2025-01-30_Reference-Material-I
20 pages
Control System University Project
No ratings yet
Control System University Project
29 pages
The Drama of the End Time - Oral Roberts
No ratings yet
The Drama of the End Time - Oral Roberts
98 pages
1 Thessalonians 2
No ratings yet
1 Thessalonians 2
3 pages
18TH Class - Adjectives of Personality Flashcards
No ratings yet
18TH Class - Adjectives of Personality Flashcards
70 pages