Front cover

Developing Applications with IBM FileNet P8 APIs

Content Engine basic and advanced APIs Process Engine basic and advanced APIs REST API, ECM Widgets, sample applications, and more

Wei-Dong Zhu Bill Carpenter Tim Lai Wei Liao Michael Oland James S Pagadala Juan Saad

ibm.com/redbooks

International Technical Support Organization Developing Applications with IBM FileNet P8 APIs December 2009

SG24-7743-00

Note: Before using this information and the product it supports, read the information in Notices on page xiii.

First Edition (December 2009) This edition applies to Version 4, Release 5, of IBM FileNet Content Manager (product number 5724-R81) and Version 4, Release 5, of IBM FileNet Business Process Manager (product number 5724-R76)

Copyright International Business Machines Corporation 2009. All rights reserved. Note to U.S. Government Users Restricted Rights -- Use, duplication or disclosure restricted by GSA ADP Schedule Contract with IBM Corp.

Contents
Figures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . ix Notices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiii Trademarks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xiv Preface . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xv The team who wrote this book . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xvi Become a published author . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii Comments welcome. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . xviii Chapter 1. Platform and API overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 1 1.1 Platform overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1.2 IBM FileNet API overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1.2.1 Content Engine APIs. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4 1.2.2 Process Engine APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 7 1.2.3 Records Manager Java API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.2.4 The eForms APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.2.5 Capture APIs . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8 1.2.6 Image Services Resource Adapter . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 1.3 Communication between applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9 Chapter 2. Setting up development environments. . . . . . . . . . . . . . . . . . . 13 2.1 Content Engine Java development setup . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1.1 Required libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1.2 Transport protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 14 2.1.3 Thick client versus thin client requirements . . . . . . . . . . . . . . . . . . . 15 2.1.4 Sample Content Engine Java API application setup in Eclipse . . . . 22 2.2 Process Engine Java development setup . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.2.1 Required libraries . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27 2.3 .NET environment setup for CE and PE . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.3.1 Prerequisites . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 29 2.3.2 Running the sample application supplied by IBM . . . . . . . . . . . . . . . 29 2.3.3 Configuring VisualStudio.NET 2005 . . . . . . . . . . . . . . . . . . . . . . . . . 30 2.4 PE REST API sample code development setup . . . . . . . . . . . . . . . . . . . . 33 2.5 ECM Widgets development setup . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 Chapter 3. Introduction to Content Engine API programming . . . . . . . . . 39 3.1 Content Engine API class overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 40 3.1.1 Content Engine API class model. . . . . . . . . . . . . . . . . . . . . . . . . . . . 40

Copyright IBM Corp. 2009. All rights reserved.

iii

3.2 Making the initial connection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43 3.2.1 User authentication . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 44 3.2.2 Java. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45 3.2.3 .NET . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48 3.3 Exception handling . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 49 3.4 Creating, retrieving, updating, and deleting objects . . . . . . . . . . . . . . . . . 52 3.4.1 Pending actions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.4.2 Creating objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 52 3.4.3 Working with properties. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 53 3.4.4 Retrieving objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 58 3.4.5 Deleting objects. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 60 3.4.6 Retrieving content . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 61 3.4.7 Working with property filters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 66 3.5 getInstance() versus fetchInstance() . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 3.6 Querying . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 71 3.6.1 SearchSQL . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 72 3.6.2 Search scope . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 74 3.6.3 Content searches . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.6.4 Paging support . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 76 3.7 Viewing documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 78 3.7.1 User tokens . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 80 3.8 Batching and batch execution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 81 Chapter 4. Advanced Content Engine API programming . . . . . . . . . . . . . 87 4.1 Permissions and authorization . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 88 4.2 Versioning . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 90 4.3 Relationships. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.3.1 Object-valued properties . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 94 4.3.2 Filing in a folder . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 95 4.3.3 Compound documents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 97 4.4 Annotations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 99 4.5 Subscriptions and event actions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 102 4.6 Workflow subscriptions and workflow event actions . . . . . . . . . . . . . . . . 105 4.7 Metadata discovery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 105 4.8 Dynamic security inheritance . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 108 Chapter 5. Introduction to Process Engine API programming . . . . . . . . 111 5.1 Process Engine API overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 112 5.1.1 Functional groups . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5.1.2 Available API functionality . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 114 5.1.3 Naming conventions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 115 5.1.4 Core classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 116 5.1.5 Functional relationship . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 118

iv

Developing Applications with IBM FileNet P8 APIs

5.2 Establishing a Process Engine session. . . . . . . . . . . . . . . . . . . . . . . . . . 119 5.2.1 Java API scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 119 5.2.2 PEWS API scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 121 5.2.3 REST API scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 5.3 Handling API exceptions . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 123 5.3.1 VWException object . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 5.3.2 Steps to handle an exception . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 124 5.4 Launching a workflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 125 5.4.1 REST API scenario . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 127 5.5 Search work items. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 129 5.5.1 Query a roster . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 131 5.5.2 Workflow queues. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134 5.5.3 Query event log . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137 5.6 Process work items . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 5.6.1 Retrieve step element . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140 5.6.2 Get step element parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 143 5.6.3 Set step element parameter values . . . . . . . . . . . . . . . . . . . . . . . . 148 5.6.4 Complete work items. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 156 5.7 Work with process status . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 5.7.1 Retrieve process history . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 162 5.7.2 Retrieve process milestones . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 165 Chapter 6. Advanced Process Engine API programming . . . . . . . . . . . . 167 6.1 Component Integrator . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 6.1.1 CE_Operations component. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 168 6.1.2 Implementing a custom Java component . . . . . . . . . . . . . . . . . . . . 170 6.2 Application space, role, and workbasket . . . . . . . . . . . . . . . . . . . . . . . . . 187 6.2.1 Retrieve role list . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 188 6.2.2 Retrieve role description and attributes. . . . . . . . . . . . . . . . . . . . . . 189 6.2.3 Retrieve workbasket . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 192 6.2.4 Query work items from workBasket . . . . . . . . . . . . . . . . . . . . . . . . 194 6.3 Resource navigation in Process Engine REST API . . . . . . . . . . . . . . . . 199 6.4 ECM Widgets overview . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 201 6.4.1 ECM Widgets concepts. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 202 6.4.2 ECM Widgets system architecture . . . . . . . . . . . . . . . . . . . . . . . . . 205 6.5 Building a custom Get Next In-basket widget . . . . . . . . . . . . . . . . . . . . . 206 6.5.1 Use case for the Get Next widget . . . . . . . . . . . . . . . . . . . . . . . . . . 207 6.5.2 Setup development environment . . . . . . . . . . . . . . . . . . . . . . . . . . 209 6.5.3 Code structure of the Get Next widget . . . . . . . . . . . . . . . . . . . . . . 210 6.5.4 Defining the Get Next widget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 6.5.5 Code skeleton for GetNext.js . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213 6.5.6 Rendering the Get Next widget user interface . . . . . . . . . . . . . . . . 215 6.5.7 Invoking PE REST service to fetch queue element . . . . . . . . . . . . 215

Contents

6.5.8 Invoking PE REST service to fetch role and In-basket list . . . . . . . 217 6.5.9 Deploying the widget. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 217 6.5.10 Building the solution . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 Chapter 7. Sample applications for Fictional Auto Rental Company A . 223 7.1 Introduction to sample applications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 224 7.2 Business use cases. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 225 7.3 User view of the sample applications . . . . . . . . . . . . . . . . . . . . . . . . . . . 226 7.3.1 User view: Reservation Web application. . . . . . . . . . . . . . . . . . . . . 226 7.3.2 User view: Kiosk application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 230 7.3.3 User view: Agent handheld application . . . . . . . . . . . . . . . . . . . . . . 232 7.3.4 User view: Fleet Status Manager Web application . . . . . . . . . . . . . 232 7.3.5 User view: Billing Report application . . . . . . . . . . . . . . . . . . . . . . . . 237 7.4 Data model . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 239 7.4.1 Base classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 243 7.4.2 ITSOVechicle . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 244 7.4.3 ITSOPhotoGallery . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 7.4.4 ITSOThumbnail . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 247 7.4.5 ITSOVehicleActivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 248 7.4.6 ITSOSingleton. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 249 7.4.7 ITSORentalActivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 250 7.4.8 ITSOMaintenanceActivity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 251 7.4.9 ITSODisposalActivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253 7.4.10 ITSOIdleActivity. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 7.4.11 ITSOCustomer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 254 7.4.12 ITSOCommentary . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 255 7.4.13 ITSOFranchise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 257 7.4.14 ITSORole . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 258 7.5 Security model. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 260 7.6 Workflows . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 262 7.6.1 Isolated region configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 7.6.2 Component queues . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 263 7.6.3 Vehicle reservation workflow. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 264 7.6.4 Vehicle maintenance workflow . . . . . . . . . . . . . . . . . . . . . . . . . . . . 266 7.7 Internal architecture of sample applications . . . . . . . . . . . . . . . . . . . . . . 268 7.7.1 Architecture: Reservation Web application . . . . . . . . . . . . . . . . . . . 268 7.7.2 Architecture: Kiosk application . . . . . . . . . . . . . . . . . . . . . . . . . . . . 270 7.7.3 Architecture: Fleet Status Manager Web application . . . . . . . . . . . 272 7.7.4 Architecture: Billing Report application . . . . . . . . . . . . . . . . . . . . . . 274 7.8 Deployment instructions for sample applications . . . . . . . . . . . . . . . . . . 276 7.8.1 Application package structure . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 278 7.8.2 Content Engine artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 279 7.8.3 Process Engine artifacts . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285

vi

Developing Applications with IBM FileNet P8 APIs

7.8.4 7.8.5 7.8.6 7.8.7

Deployment: Reservation Web application . . . . . . . . . . . . . . . . . . . 292 Deployment: Kiosk application . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295 Deployment: Fleet Status Manager Web application . . . . . . . . . . . 296 Deployment: Billing Report application . . . . . . . . . . . . . . . . . . . . . . 299

Chapter 8. Logging and troubleshooting . . . . . . . . . . . . . . . . . . . . . . . . . 305 8.1 Logging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 8.1.1 CE Java API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 306 8.1.2 CE .NET API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 8.1.3 Content Engine Web Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . 310 8.1.4 PE Java API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 8.1.5 Process Engine Web Services . . . . . . . . . . . . . . . . . . . . . . . . . . . . 311 8.1.6 PE REST Service . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 8.2 Troubleshooting. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 8.2.1 log4j debugging . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 312 8.2.2 Content Engine troubleshooting techniques . . . . . . . . . . . . . . . . . . 313 8.2.3 Process Engine troubleshooting techniques . . . . . . . . . . . . . . . . . . 313 8.2.4 Data must be gathered for troubleshooting . . . . . . . . . . . . . . . . . . . 314 Appendix A. Additional material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Locating the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 315 Using the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 System requirements for downloading the Web material . . . . . . . . . . . . . 316 How to use the Web material . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 316 Related publications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 IBM Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 Online resources . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 317 How to get Redbooks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 Help from IBM . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 318 Index . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 319

Contents

vii

viii

Developing Applications with IBM FileNet P8 APIs

Figures
1-1 Core engines and data storage. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2 1-2 Core engines with additional applications . . . . . . . . . . . . . . . . . . . . . . . . . . 3 1-3 Application communication between core engines and available applications. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1-4 .NET application communication. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10 1-5 Typical Java application communication . . . . . . . . . . . . . . . . . . . . . . . . . . 11 2-1 Content Engine Client libraries installation directory . . . . . . . . . . . . . . . . . 14 2-2 WebSphere 6.1 Application Client installation directory . . . . . . . . . . . . . . 16 2-3 Eclipse runtime configuration for WebSphere EJB transport JVM parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 17 2-4 Sample WebSphere sas.client.props entries . . . . . . . . . . . . . . . . . . . . . . 17 2-5 CE 4.5.0 CEWS transport installation directory . . . . . . . . . . . . . . . . . . . . 19 2-6 Eclipse runtime configuration for CEWS transport JVM parameters . . . . 20 2-7 Required .jar files for CE and CEWS transport . . . . . . . . . . . . . . . . . . . . . 23 2-8 ContentEngineDemo create cesample package . . . . . . . . . . . . . . . . . . . . 23 2-9 Imported demo Java classes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 24 2-10 Java Application runtime profile for MainFrame.java . . . . . . . . . . . . . . . 24 2-11 Runtime JVM parameters for MainFrame.java . . . . . . . . . . . . . . . . . . . . 25 2-12 Demo application main page. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 25 2-13 Demo application: Test Server Connection. . . . . . . . . . . . . . . . . . . . . . . 26 2-14 Process Engine Client libraries installation directory . . . . . . . . . . . . . . . 27 2-15 Eclipse jar files for Process Engine development . . . . . . . . . . . . . . . . . . 28 2-16 Process Engine JVM parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 28 2-17 Adding a reference to the CE API DLL . . . . . . . . . . . . . . . . . . . . . . . . . . 31 2-18 Adding a Content Engine Web Service .NET Web Reference . . . . . . . . 32 2-19 Adding a Process Engine Web Service .NET Web Reference . . . . . . . . 33 2-20 Create New Web Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 34 2-21 Specify the Project Name . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35 2-22 Code structure after you copy Dojo library . . . . . . . . . . . . . . . . . . . . . . . 35 2-23 Test page for PE REST API . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 36 2-24 Export the project as WAR file . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38 3-1 Class relationships and general object flow . . . . . . . . . . . . . . . . . . . . . . . 43 3-2 Objects and properties returned . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 70 3-3 Compiled SQL statement from Content Engine Query Builder . . . . . . . . . 73 4-1 VersionSeries object with multiple IndependentObjects . . . . . . . . . . . . . . 91 5-1 Primary Runtime API objects . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 113 5-2 Runtime API core classes calling sequences . . . . . . . . . . . . . . . . . . . . . 118 6-1 Java component association. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 171

Copyright IBM Corp. 2009. All rights reserved.

ix

6-2 Run configuration for PEComponentQueueHelper . . . . . . . . . . . . . . . . . 183 6-3 Example for PEComponentQueueHelper parameters in Eclipse . . . . . . 184 6-4 Run configuration for Component Manager START . . . . . . . . . . . . . . . . 186 6-5 Example for Component Manager START parameters in Eclipse . . . . . 186 6-6 Example for Component Manager STOP parameters in Eclipse . . . . . . 187 6-7 My Work page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 6-8 Step Processor Page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203 6-9 Single Page application. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 204 6-10 ECM Widgets system architecture . . . . . . . . . . . . . . . . . . . . . . . . . . . . 206 6-11 Get Next In-basket Widget working with the other ECM Widgets . . . . . 207 6-12 Use case for the Get Next widget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 208 6-13 Create a new Web application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 209 6-14 Specify the project name. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 210 6-15 Code Structure of Get Next Widget. . . . . . . . . . . . . . . . . . . . . . . . . . . . 211 6-16 The Get Next widget in the view mode . . . . . . . . . . . . . . . . . . . . . . . . . 215 6-17 The Get Next widget in the edit mode. . . . . . . . . . . . . . . . . . . . . . . . . . 215 6-18 Export sample Web application. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 218 6-19 The Get Next widget in Business Space toolbox . . . . . . . . . . . . . . . . . 219 6-20 Create mashup page. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 219 6-21 Place the widgets in the mashup page . . . . . . . . . . . . . . . . . . . . . . . . . 220 6-22 Wire Get Next and Step Completion by Send Work Item Id and Receive Work Item Id events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 6-23 Wire Step Completion and Get Next by Send Work Item Id and Receive Work Item Id events . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 220 6-24 Configure the Get Next widget . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 221 7-1 Reservation Web application login panel . . . . . . . . . . . . . . . . . . . . . . . . 227 7-2 Reservation Web application registration panel . . . . . . . . . . . . . . . . . . . 227 7-3 Reservation Web application main data panel . . . . . . . . . . . . . . . . . . . . 228 7-4 Reservation Web application vehicle selection panel . . . . . . . . . . . . . . . 228 7-5 Reservation Web application credit card information collection panel . . 229 7-6 Reservation Web application rental reservation confirmation panel . . . . 230 7-7 Entering confirmation number . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 7-8 Rental information dialog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 231 7-9 Vehicle return dialog . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 232 7-10 Fleet Status Dashboard . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 234 7-11 Fleet Manager page to send vehicle for maintenance . . . . . . . . . . . . . 235 7-12 Maintenance queue page . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 235 7-13 Maintenance worker page: route work item to workers work queue . . 236 7-14 Maintenance worker page: process work item . . . . . . . . . . . . . . . . . . . 236 7-15 Maintenance queue page for a different user . . . . . . . . . . . . . . . . . . . . 237 7-16 Process Administrator showing work items in roster. . . . . . . . . . . . . . . 237 7-17 Login dialog for Billing Report application . . . . . . . . . . . . . . . . . . . . . . . 238 7-18 Login failed . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 238

Developing Applications with IBM FileNet P8 APIs

7-19 7-20 7-21 7-22 7-23 7-24 7-25 7-26 7-27 7-28 7-29 7-30 7-31 7-32 7-33 7-34 7-35 7-36 7-37 7-38 7-39 7-40 7-41 7-42 7-43 7-44 7-45

Main window for Billing Report application . . . . . . . . . . . . . . . . . . . . . . 239 Model for ITSODocument and subclasses . . . . . . . . . . . . . . . . . . . . . . 240 Models for ITSOFolder, ITSOAnnotation, and subclasses . . . . . . . . . . 241 Model for ITSOCustomObject and subclasses . . . . . . . . . . . . . . . . . . . 242 ITSO vehicle reservation main workflow diagram . . . . . . . . . . . . . . . . . 265 ITSO vehicle maintenance main workflow diagram . . . . . . . . . . . . . . . 267 Program flow for Reservation Web application . . . . . . . . . . . . . . . . . . . 269 Program flow for Kiosk application . . . . . . . . . . . . . . . . . . . . . . . . . . . . 271 Program flow for Fleet Status Manager application . . . . . . . . . . . . . . . 273 Program flow for Billing Report application . . . . . . . . . . . . . . . . . . . . . . 276 Directory structure of sg247743-sample . . . . . . . . . . . . . . . . . . . . . . . . 278 CE import task. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 281 CE import helper panel Import Options. . . . . . . . . . . . . . . . . . . . . . . . . 282 CE import helper panel Import Items for metadata . . . . . . . . . . . . . . . . 283 CE import helper panel Import Items for instance data. . . . . . . . . . . . . 284 ITSOAutoRental and subfolders . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 285 Stop all component manager instances . . . . . . . . . . . . . . . . . . . . . . . . 286 Ensure correct notification settings on Process Task Manager . . . . . . 286 Import isolated region configuration . . . . . . . . . . . . . . . . . . . . . . . . . . . 287 Set JAAS credentials on component queues . . . . . . . . . . . . . . . . . . . . 288 ITSOOperations.jar required library . . . . . . . . . . . . . . . . . . . . . . . . . . . 291 ITSO_Operations JRE parameters sample. . . . . . . . . . . . . . . . . . . . . . 291 Component Manager started with no errors . . . . . . . . . . . . . . . . . . . . . 292 Importing the BillingReportApplication project into Eclipse . . . . . . . . . . 301 BillingReportApplication structure in Eclipse . . . . . . . . . . . . . . . . . . . . . 302 Run As Java Application . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303 Billing application JVM requirements . . . . . . . . . . . . . . . . . . . . . . . . . . 304

Figures

xi

xii

Developing Applications with IBM FileNet P8 APIs

Notices
This information was developed for products and services offered in the U.S.A. IBM may not offer the products, services, or features discussed in this document in other countries. Consult your local IBM representative for information on the products and services currently available in your area. Any reference to an IBM product, program, or service is not intended to state or imply that only that IBM product, program, or service may be used. Any functionally equivalent product, program, or service that does not infringe any IBM intellectual property right may be used instead. However, it is the user's responsibility to evaluate and verify the operation of any non-IBM product, program, or service. IBM may have patents or pending patent applications covering subject matter described in this document. The furnishing of this document does not give you any license to these patents. You can send license inquiries, in writing, to: IBM Director of Licensing, IBM Corporation, North Castle Drive, Armonk, NY 10504-1785 U.S.A. The following paragraph does not apply to the United Kingdom or any other country where such provisions are inconsistent with local law: INTERNATIONAL BUSINESS MACHINES CORPORATION PROVIDES THIS PUBLICATION "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESS OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF NON-INFRINGEMENT, MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. Some states do not allow disclaimer of express or implied warranties in certain transactions, therefore, this statement may not apply to you. This information could include technical inaccuracies or typographical errors. Changes are periodically made to the information herein; these changes will be incorporated in new editions of the publication. IBM may make improvements and/or changes in the product(s) and/or the program(s) described in this publication at any time without notice. Any references in this information to non-IBM Web sites are provided for convenience only and do not in any manner serve as an endorsement of those Web sites. The materials at those Web sites are not part of the materials for this IBM product and use of those Web sites is at your own risk. IBM may use or distribute any of the information you supply in any way it believes appropriate without incurring any obligation to you. Information concerning non-IBM products was obtained from the suppliers of those products, their published announcements or other publicly available sources. IBM has not tested those products and cannot confirm the accuracy of performance, compatibility or any other claims related to non-IBM products. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. This information contains examples of data and reports used in daily business operations. To illustrate them as completely as possible, the examples include the names of individuals, companies, brands, and products. All of these names are fictitious and any similarity to the names and addresses used by an actual business enterprise is entirely coincidental. COPYRIGHT LICENSE: This information contains sample application programs in source language, which illustrate programming techniques on various operating platforms. You may copy, modify, and distribute these sample programs in any form without payment to IBM, for the purposes of developing, using, marketing or distributing application programs conforming to the application programming interface for the operating platform for which the sample programs are written. These examples have not been thoroughly tested under all conditions. IBM, therefore, cannot guarantee or imply reliability, serviceability, or function of these programs.

Copyright IBM Corp. 2009. All rights reserved.

xiii

Trademarks
IBM, the IBM logo, and ibm.com are trademarks or registered trademarks of International Business Machines Corporation in the United States, other countries, or both. These and other IBM trademarked terms are marked on their first occurrence in this information with the appropriate symbol ( or ), indicating US registered or common law trademarks owned by IBM at the time this information was published. Such trademarks may also be registered or common law trademarks in other countries. A current list of IBM trademarks is available on the Web at http://www.ibm.com/legal/copytrade.shtml The following terms are trademarks of the International Business Machines Corporation in the United States, other countries, or both: FileNet IBM Rational Redbooks Redpaper Redbooks (logo) WebSphere

The following terms are trademarks of other companies: FileNet, and the FileNet logo are registered trademarks of FileNet Corporation in the United States, other countries or both. JBoss, and the Shadowman logo are trademarks or registered trademarks of Red Hat, Inc. in the U.S. and other countries. Java, and all Java-based trademarks are trademarks of Sun Microsystems, Inc. in the United States, other countries, or both. Microsoft, Windows, and the Windows logo are trademarks of Microsoft Corporation in the United States, other countries, or both. Intel, Intel logo, Intel Inside logo, and Intel Centrino logo are trademarks or registered trademarks of Intel Corporation or its subsidiaries in the United States and other countries. Other company, product, or service names may be trademarks or service marks of others.

xiv

Developing Applications with IBM FileNet P8 APIs

Preface
This IBM Redbooks publication can help you develop content and process management applications with IBM FileNet APIs. The IBM FileNet P8 suite of products contains a set of robust APIs that range from core platform APIs to supporting application APIs. This book focuses specifically on Content Engine and Process Engine APIs. The two core components of IBM FileNet P8 Platform are Content Engine and Process Engine. Content Engine handles content storage, indexing, and retrieval. In addition, it offers services for enabling active content through events and subscriptions, allowing code to be executed on an event-driven basis. Process Engine handles all workflow and process routing. The Content Engine and Process Engine APIs enable you to develop applications that can perform the Content Engine and Process Engine functions. In this book, we show you how to set up programming environments for Java, .NET, REST API, and ECM Widget development. We provide a class overview, a description of how to make connections to the engines, and examples of exception handling. Content Engine API topics that we discuss include creating, retrieving, updating, and deleting objects; querying and viewing documents; and batching and batch execution. We also explore more complex topics, including permissions and authorization, versioning, relationships, annotations, workflow subscriptions and event actions, metadata discovery, and dynamic security inheritance. Process Engine API topics that we discuss include launching a workflow, searching for and processing work items, and working with process status. The more complex topics we cover include, Component Integrator application space, role, workbasket, resource navigation in Process Engine REST API, ECM Widgets, and building a custom Get Next In-basket widget. To help you better understand programming with IBM FileNet APIs, we provide a sample application implemented for a fictional company. We include the data model, security model, workflows, and various applications developed for the sample. You can download them for your reference. In this book, we also teach you how to enable and interpret the various logging options for the IBM FileNet P8 Platform APIs. We point you to the technical troubleshooting articles and provide important data-gathering points when troubleshooting IBM FileNet programming issues.

Copyright IBM Corp. 2009. All rights reserved.

xv

This book is intended for IBM FileNet P8 application developers. We recommend using this book in conjunction with the online ECM help.

The team who wrote this book

This book was produced by a team of specialists from around the world working at the International Technical Support Organization, Rochester Center. Wei-Dong Zhu (Jackie) is an Enterprise Content Management Project Leader with the International Technical Support Organization in Los Angeles, California. She has more than 10 years of software development experience in accounting, image workflow processing, and digital media distribution. Jackie holds a Master of Science degree in Computer Science from the University of the Southern California. Jackie joined IBM in 1996. She is a Certified Solution Designer for IBM Content Manager and has managed and lead the production of many enterprise content management books. Bill Carpenter is an Enterprise Content Management Architect with the IBM Software Group in Seattle, Washington. Bill has 10 years of experience in the IBM enterprise content management business, as a Developer, Development Manager, and Architect. He is co-author of the book IBM FileNet Content Manager Implementation Best Practices and Recommendations, SG24-7547. He has extensive experience in building large software systems for Fortune 50 companies and has served as the CTO of an Internet startup. He has been a contributor to several open source projects. Bill holds degrees in Mathematics and Computer Science from Rensselaer Polytechnic Institute in Troy, New York. Tim Lai is a Team Lead for the IBM FileNet Software Development Support Group in Costa Mesa, California. He has 12 years of experience in programming, consulting, and supporting ECM, E-Commerce, ERP, CRM, and Data Warehouse solutions worldwide. Tim holds a Bachelors of Science degree in Industrial and Systems Engineering from the University of Southern California. He joined IBM FileNet in 2003. Wei Liao is a Enterprise Content Management Software Developer in the IBM China Development Lab. He has four years of IT industry experience in programming, J2EE development, Windows .NET development, Web development and enterprise content management. He holds a Master of Science degree in Computer Science from the Beijing University of Technology. His core competency is Web 2.0 and Enterprise Mashup. He is proficient in building rich Internet applications, based on Web standards.

xvi

Developing Applications with IBM FileNet P8 APIs

Michael Oland is an IT Advisory Specialist with IBM Enterprise Content Management Lab Services in Washington, D.C. He has 10 years experience in enterprise content management, which includes Web content delivery and management, document and information management, and business process management. For the last six years, he has focused primarily on content migration, and developing tools for content migration. He holds a degree in Broadcasting from the University of Tennessee, Knoxville. James S Pagadala is a J2EE Enterprise Application Consultant from IBM India. He has more than seven years of experience in developing J2EE applications. His primary focus is supply chain management. He has managed and lead many J2EE applications in this domain. James has a Master of Computer Applications degree from the Karnataka Regional Engineering College, Surathkal, India. He joined IBM in 2001. Juan Saad is a Senior Architect for Enterprise Content Management solutions with an IBM Business Partner in Melbourne, Australia. Juan has nine years of experience with organizations, designing, implementing, and maintaining Business Process Management and Content Management solutions based on the IBM FileNet P8 Platform. Juan has been involved in all phases of a system development life cycle, both functionally and technically, including client interface, business process analysis and modeling, application design and development, and system testing and research methods. He is a Certified Administrator and Developer for several IBM FileNet P8 products such as Business Process Manager, Content Manager, Records Manager, Email Manager, and Forms Manager. Thanks to the following people and groups for their contributions to this project: Patrick Doonan William Lobig Mike Winter Quynh Dang Evangeline Fink Robert Mariano Mike Marin Lauren Mayes Himanshu Shah Darik Siegfried IBM Software Development Lab in Costa Mesa, California Dave Perman IBM Software Development Lab in Vancouver, B.C., Canada

Preface

xvii

Become a published author

Join us for a two- to six-week residency program! Help write a book dealing with specific products or solutions, while getting hands-on experience with leading-edge technologies. You will have the opportunity to team with IBM technical professionals, Business Partners, and Clients. Your efforts will help increase product acceptance and customer satisfaction. As a bonus, you will develop a network of contacts in IBM development labs, and increase your productivity and marketability. Find out more about the residency program, browse the residency index, and apply online at: ibm.com/redbooks/residencies.html

Comments welcome
Your comments are important to us! We want our books to be as helpful as possible. Send us your comments about this book or other IBM Redbooks publications in one of the following ways: Use the online Contact us review Redbooks form found at: ibm.com/redbooks Send your comments in an e-mail to: [email protected] Mail your comments to: IBM Corporation, International Technical Support Organization Dept. HYTD Mail Station P099 2455 South Road Poughkeepsie, NY 12601-5400

xviii

Developing Applications with IBM FileNet P8 APIs

Chapter 1.

Platform and API overview

Before beginning development for the IBM FileNet P8 Platform, it is important to review the platform itself, the core engines that comprise a IBM FileNet P8 deployment, the application programming interface (API) available for each engine, and how they communicate with each other. This chapter discusses the following topics: Platform overview IBM FileNet API overview Communication between applications

Copyright IBM Corp. 2009. All rights reserved.

1.1 Platform overview

The core IBM FileNet P8 Platform consists of two primary engines, Content Engine (CE) and Process Engine (PE). On top of these engines are built all user interfacing applications, such as the Application Engine (AE), Workplace XT, and Enterprise Manager. As an additional layer on top of the base, optional components are available for license, including IBM FileNet Records Manager. Figure 1-1 shows the relationship between the core engines and data storage.

Ready-to-use Applications (Workplace XT, Enterprise Manager) Content Engine Process Engine

Data Storage

Figure 1-1 Core engines and data storage

The CE handles content storage, indexing, and retrieval. In addition, the CE offers services for enabling active content through events and subscriptions, allowing code to be executed on an event-driven basis. The PE handles all workflow and process routing. The applications that are ready for use include Workplace XT and IBM FileNet Enterprise Manager. In addition to the core engines, a number of add-on products offer additional functionality and are built on the core engines. Some of these applications include IBM FileNet Records Manager and IBM FileNet Capture. See Figure 1-2 on page 3.

Developing Applications with IBM FileNet P8 APIs

Records Manager

Capture

Application Engine and Workplace XT

Content Engine

Process Engine

Data Storage
Figure 1-2 Core engines with additional applications

1.2 IBM FileNet API overview

The IBM FileNet suite of products contains a set of robust application programming interfaces (APIs). These APIs range from core platform APIs to supporting application APIs.

Core platform APIs

Within the core products, both primary engines, the CE and the PE, offer a series of APIs that are available for different languages and uses. Table 1-1 lists the primary APIs and what they are intended to be used for.
Table 1-1 Core P8 Platform APIs by use Language or use Java .NET Web Services Other P8 3.x Compatibility Content Engine Java API .NET API CE Web Services COM Compatibility Layer Java Compatibility Layer Process Engine Java API PE Web Services REST -

Chapter 1. Platform and API overview

Supporting application APIs

Along with the core platform APIs are other add-on applications, which often have APIs that are specific to that application. Several of these APIs are: IBM FileNet Records Manager Java API eForms JavaScript API eForms Java API Capture COM API Remote Capture Services .NET API Image Services Resource Adapter API

1.2.1 Content Engine APIs

To provide services for creating, querying, and working with content, the CE offers two primary and feature-compatible APIs: the Java and .NET APIs. In addition, the CE offers two compatibility APIs for applications that were written with the CE 3.x versions.

Content Engine Java API

IBM FileNet Content Manager provides a full-featured CE Java API. Any feature that is available in the CE server is available to Java programmers. These features include routine operations, such as retrieving and updating Document objects, and specialized operations, such as adding a custom class or property to an object stores metadata definitions. This API can be configured to communicate with the CE server through the EJB or Web Service transports. The Java API set complies with the Enterprise JavaBeans 2.0 specification for the EJB transports that are used internally. The Java API uses Java Authentication and Authorization Service (JAAS) for authentication purposes. Fine-grained authorization is implemented within the CE server and does not use JAAS.

Content Engine .NET API

IBM FileNet Content Manager provides a full-featured .NET API, which you can use to write programs in any .NET-compatible language. With a couple of exceptions, any feature that is available in the CE server is available to .NET programmers. The exceptions are mainly custom code that must be executed within the server, for example, EventActions. Because the CE server is a J2EE application, internally executed custom code is limited to Java-compatible technologies. One significant feature available only with the .NET API is the use of Kerberos to perform authentication with Microsoft Windows integrated login. This is only

Developing Applications with IBM FileNet P8 APIs

possible when both the client application and the CE server are running on Microsoft Windows. The .NET API communicates with the CE server over the Web Services transport.

Content Engine Web Services

Modern, loosely coupled frameworks, such as service-oriented architecture, favor Web services protocols for connecting components. IBM FileNet Content Manager provides Content Engine Web Services (CEWS) for accessing nearly all features available in the CE server. Typically, if you, as a programmer, want to use a Web services interface, you obtain the interface description in the form of a Web Services Description Language (WSDL) file. You run the WSDL file through a toolkit to generate programming language objects for interacting with the Web services interface. You then usually build up a library of utilities to provide abstraction layers, caching, security controls, and other conveniences. The Java and .NET APIs provided by IBM FileNet Content Manager are already equivalent to that, and both APIs can use Web services as a transport. Consequently, there is not as much motivation to use CEWS directly. There are still occasions where the direct use of CEWS might be useful: You have an application already using CEWS, and no plans exist for immediately porting it to the Java or .NET API. You are building an application component as part of a framework or technology stack in which the use of Web services is the model for communicating with external systems and the use of the CE Java or .NET API is not a reasonable possibility. For these occasions, the direct use of CEWS is a good choice and is fully supported. In theory, any current Web services toolkit can use the Content Engine WSDL to generate the interfaces that you use for your application. In practice, however, toolkits are still individualistic in their handling of various WSDL features; writing a WSDL for a complex service that is usable by a wide cross-section of Web services toolkits is difficult. Check the latest hardware and software support documentation, IBM FileNet P8 Platform 4.5.x hardware and software requirements, and only use a supported toolkit. Your toolkit generates programming language stubs and other artifacts so that you can include CEWS calls in your program logic. The details of those artifacts vary from toolkit to toolkit, but you will surely see representations of CE objects,

Chapter 1. Platform and API overview

properties, and update operations. The IBM FileNet Content Manager product documentation describes how these pieces interrelate, but we recommend you also read the developer documentation for the Java or .NET API to get an additional understanding of how the things mentioned in the WSDL were intended to be used. A common question for CEWS developers is what the XML actually looks like for various requests. The official specification, of course, is the WSDL, and any XML generated by your client must comply with that. The simplest way to obtain samples of actual XML is to use either the .NET or Java API with CEWS transport and use a network tracing tool to capture the data from the interactions. You will see both the requests and responses. Because of the rich variety of possible interactions in CEWS, IBM support does not have a set of ready-made XML samples. Note: As of release 4.5.0, CE supports three Web services endpoints. The difference is in their handling of content attachments. The SOAP endpoint (FNCEWS40SOAP) uses inline content and carries significant performance costs. It should therefore be avoided. Support for the Direct Internet Message Encapsulation (DIME) endpoint (FNCEWS40DIME) is documented as deprecated and will eventually be removed. Therefore, all new code should be written for the Message Transmission Optimization Mechanism (MTOM) endpoint (FNCEWS40MTOM), and existing DIME code should be migrated to MTOM as soon as possible.

Content Engine Java Compatibility Layer API

IBM FileNet Content Manager 3.x provided a set of Java APIs which is now called the Java Compatibility Layer in P8 4.x. This Java compatibility layer allows almost all custom applications that are written in P8 3.x to continue functioning in P8 4.x with a few configuration changes. All new development should be done with the Content Manager 4.x Java APIs because new features are incorporated only into the 4.x APIs. The Java Compatibility Layer API is only mentioned in this book in terms of how to configure the environment. The book does not discuss Java Compatibility Layer programming.

Content Engine COM Compatibility Layer API

IBM FileNet Content Manager 3.x also provided a COM API set. This COM API set is now supported as a compatibility layer in P8 4.x, allowing many COM-based custom applications that are written in P8 3.x to continue functioning in P8 4.x with a few configuration changes.

Developing Applications with IBM FileNet P8 APIs

Any new Microsoft environment development should be done by using the IBM FileNet Content Manager 4.x .NET API because all new features will be incorporated into the 4.x APIs. The COM Compatibility Layer API is only mentioned in this Redbooks publication in terms of how to configure the environment. There will be no discussion on COM Compatibility Layer programming.

1.2.2 Process Engine APIs

To provide services for creating and managing workflows, the PE offers three APIs: the Java API, Process Engine Web Services, and the REST API.

Process Engine Java API

IBM FileNet Business Process Manager contains a full featured Java API set that allows for various interactions with the PE server. This API set may be familiar to a lot of IBM FileNet experienced programmers from the eProcess versions through IBM FileNet P8 4.x. This API set provides access to define, administer, and control runtime workflow processing. It can also be used to build custom Step Processors and Work Performers. The PE Java API uses the CE Java API for authentication. Authentication-related calls can be done with EJB or WSI transports, though the use of the Web services transport is the normal configuration.

Process Engine Web Services

Process Engine Web Services (PEWS) provides a WSDL-based interface to communicate with PE. PEWS is a functional subset of the PE Java API. Because there is no native PE .NET API, PEWS is ideal for development with .NET. Similar to the CEWS, PEWS is also compliant with Web Services Interoperability Organization (WS-I) Basic Profile 1.0 and WS-Security.

Process Engine REST Services API

A new feature introduced in IBM FileNet P8 4.5 is the PE REST Service. The PE REST Service provides access to the PE by following the Representational State Transfer (REST) architecture style. PE objects are exposed as REST resources which are accessible by standard HTTP methods. This API is ideal for Ajax-based Web applications that need to talk to PE.

Chapter 1. Platform and API overview

1.2.3 Records Manager Java API

The IBM FileNet Records Manager Java API set allows for various interactions with IBM FileNet Records Manager including declaring, classifying, storing, and disposing of records. This API set is currently an extension of the IBM FileNet Content Manager Java 3.5 API set. This API set is not covered in this book.

1.2.4 The eForms APIs

To provide services for managing electronic forms, eForms offers two APIs: the eForms JavaScript API and the eForms Java API.

The eForms JavaScript API

The eForms JavaScript API enables programmers to customize the client side processing of forms in a users Web browser. Examples include: Retrieving and populating form data Attaching JavaScript event handlers Performing client side validation of user inputted data This API set is not covered in this book.

The eForms Java API

The eForms Java API provides a low level access to the eForms objects and methods. This API performs the server side data processing. In practice, for most implementations, the eForms JavaScript API is the most useful in customizing eForms solutions. This API set is not covered in this book.

1.2.5 Capture APIs

To provide services for acquiring images into IBM FileNet content repositories, IBM FileNet Capture offers two APIs: the Capture COM API and the Remote Capture Services .NET API.

Capture COM API

The Capture COM API is an existing IBM FileNet API set that allows for image acquisition into IBM FileNet content repositories including CE and Image Services. This API set provides Capture repository objects and ActiveX controls

Developing Applications with IBM FileNet P8 APIs

for image acquisition (for example, scanning and document upload), batching, advanced document recognition (ADR), and others. This API set is not covered in this book.

Remote Capture Services .NET API

Remote Capture Services (RCS) provides a .NET framework for developers to build custom .NET Web applications. The RCS toolkit includes Web interface controls, the RCS framework, as well as Capture repository objects. This API set is not covered in this book.

1.2.6 Image Services Resource Adapter

The Image Services Resource Adapter (ISRA) is an existing IBM FileNet API that was developed for interaction with IBM FileNet Image Services. It is a Java based API that is J2EE Connector Architecture (JCA) compliant. This API set is not frequently used as interactions with Image Services can be accomplished through Content Federation Services for Image Services (CFS-IS). However, for systems that do not use Content Federation Services, ISRA can be used to communicate with Image Services. This API set is not covered in this book.

1.3 Communication between applications

When approaching development for the IBM FileNet P8 Platform, an important aspect to remember is how the various components integrate and communicate with each other, because this helps with debugging communications issues in custom applications. Figure 1-3 on page 10 shows the communications paths between CE, PE, Application Engine, Workplace XT, and Enterprise Manager.

Chapter 1. Platform and API overview

Application Engine and Workplace XT Enterprise Manager .NET WSI Java API Java API

Content Engine

Java API

Process Engine

Figure 1-3 Application communication between core engines and available applications

For custom applications, the communications path used depends on both the API used and which transport is selected for use. Figure 1-4 shows the typical communication path for .NET-based applications.

Windows Thick-Client / ASP.NET

.NET

PE/WSI

Content Engine

Process Engine

Figure 1-4 .NET application communication

Figure 1-5 on page 11 shows the typical communication path for Java-based applications. Although a J2EE application can use Web services transport to communicate with the CE, doing so is unusual. After an application is inside an application server, the EJB protocol has advantages in performance and ease of integration.

10

Developing Applications with IBM FileNet P8 APIs

Java Application / Java Web Application

EJB or WSI Content Engine

Java

Process Engine

Figure 1-5 Typical Java application communication

In the remaining chapters of this book, we describe how to set up development environments and provide detailed discussions about the CE and PE APIs with a sample use case application.

Chapter 1. Platform and API overview

11

12

Developing Applications with IBM FileNet P8 APIs

Chapter 2.

Setting up development environments

In this chapter, we discuss setting up Java development environments for Content Engine (CE) and Process Engine (PE). We discuss the basic requirements for setting up programming with IBM FileNet APIs as well as providing instructions, where appropriate, for the sample application included with this book. This chapter discusses the following topics: Content Engine Java development setup Process Engine Java development setup .NET environment setup for CE and PE PE REST API sample code development setup ECM Widgets development setup

Copyright IBM Corp. 2009. All rights reserved.

13

2.1 Content Engine Java development setup

This section specifies setting up Java development environment to work with Content Engine Java API. We include the required libraries, transport protocols, thick client versus thin client requirements, and sample CE Java API application setup using Eclipse.

2.1.1 Required libraries

The two primary Java archive (JAR) files that are required for CE 4.0 development are: Jace.jar log4j-<version>.jar Where <version> is the supported version that is installed with CE. These two .jar files can be obtained through installing the CE Client installer. By default, the CE Client is installed to c:\Program Files\FileNet\CEClient and the two .jar files are in the lib subdirectory. See Figure 2-1.

Figure 2-1 Content Engine Client libraries installation directory

2.1.2 Transport protocols

Communication with CE can occur through the EJB transport or Content Engine Web Services (CEWS) transport. The underlying communication protocols are handled internally by CE APIs so the application code remains the same regardless of transport selected. Chapter 3, Introduction to Content Engine API

14

Developing Applications with IBM FileNet P8 APIs

programming on page 39 discusses more details of the transport protocols and the Uniform Resource Identifiers (URIs) to use for each.

2.1.3 Thick client versus thin client requirements

The term thick client as used in this book refers to stand-alone Java applications that are running outside of a J2EE application server such as a scheduled nightly batch job that uploads documents to CE. Thin client as used in this book refers to Java applications deployed and running in a J2EE application server such as a JSP-based Web application. The requirements for thick client versus thin client vary slightly, so we discuss each in detail.

Thick client EJB transport requirements

For the EJB transport, .jar files that are specific to the application server are required in the thick client environment. The required .jar files can change with each application server version. For the application servers supported in the current CE 4.5.0 release, Table 2-1 lists the required .jar files.
Table 2-1 Thick client EJB transport library requirements Application server WebSphere Required .jar files Install the WebSphere Application Client version and patch that matches the CE application server. Use the WebSphere JRE. During runtime, use the java.ext.dirs JVM parameter mentioned in the next section wlclient.jar or weblogic.jar jbossall-client.jar

WebLogic JBOSS

Figure 2-2 on page 16 shows the WebSphere 6.1 Application Client installed directories.

Chapter 2. Setting up development environments

15

Figure 2-2 WebSphere 6.1 Application Client installation directory

In addition to the application server .jar files, the Java virtual machine (JVM) parameters listed in this section are required for thick client applications: For EJB transport with WebSphere (Figure 2-3 on page 17): -Dcom.ibm.CORBA.ConfigURL=<WebSphereHome>\properties\sas.client.prop s -Djava.ext.dirs=<WebSphereHome>\java\jre\lib\ext;<WebSphereHome>\lib ;<WebSphereHome>\plugins -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.WebSphere Where, <WebSphereHome> refers to the WebSphere Application Client home directory, for example, c:\Program Files\IBM\WebSphere\AppClient.

16

Developing Applications with IBM FileNet P8 APIs

Figure 2-3 Eclipse runtime configuration for WebSphere EJB transport JVM parameters

Note: The CE server host name you use in your connection URI and the sas.client.props must be resolvable through DNS from your client machine. For WebSphere, the host name and port specified in your client application must match the host name and port number specified in the BOOTSTRAP_ADDRESS parameter in the WebSphere Administration Console. In sas.client.props, set the following properties (see Figure 2-4): com.ibm.CORBA.securityServerHost=ceServerName com.ibm.CORBA.securityServerPort=2809 com.ibm.CORBA.loginSource=none

Figure 2-4 Sample WebSphere sas.client.props entries

For EJB transport with Weblogic:

-Djava.naming.factory.initial=weblogic.jndi.WLInitialContextFactory -Djava.naming.provider.url=t3://ceServer:7001 -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.WebLogic

Chapter 2. Setting up development environments

17

For EJB transport with JBoss:

-Djava.naming.factory.initial=org.jnp.interfaces.NamingContextFactory -Djava.naming.factory.url.pkgs=org.jboss.naming:org.jnp.interfaces -Djava.naming.provider.url=jnp://ceServer:1099 -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.JBoss

Note that the port numbers assume default application server ports. Change them as needed for your specific environments.

Thick client CEWS transport requirements

The CEWS transport does not require the client application and CE server to be on the same application server libraries. The CEWS transport is independent of the CE application server so the following instructions can be followed on any CEWS transport client. Note: For the CEWS transport in a thick client environment, the requirements are slightly different depending on which version of CE you are using. From CE 4.0 to 4.5.0, CE Java API utilized Systinet WASPJ libraries for communication with the CE server. Starting with CE 4.5.1, Systinet libraries are no longer used. We describe both configurations. For CE 4.0 to 4.5.0, the Systinet libraries and configuration files are installed with the CE Client installation (default installation directory is c:\Program Files\FileNet\CEClient\wsi). Table 2-2 lists the required .jar files.
Table 2-2 Thick client CEWS transport library requirements for CE 4.0 to 4.5.0 Application server Applies to all application servers Required .jar files wasp.jar This file can be obtained from the c:\Program Files\FileNet\CEClient\wsi\lib directory. The other .jar files in that directory should not be placed on your classpath.

Figure 2-5 on page 19 shows the CEWS installation directory where wasp.jar can be obtained.

18

Developing Applications with IBM FileNet P8 APIs

Figure 2-5 CE 4.5.0 CEWS transport installation directory

Starting with CE 4.5.1, Systinet libraries are no longer used. Instead, three new IBM libraries are required for CEWS transport as listed in Table 2-3.
Table 2-3 Thick client CEWS transport library requirements for CE 4.5.1 and beyond Application server Applies to all application servers Required .jar files stax-api.jar xlxpScanner.jar xlxpScannerUtils.jar

As mentioned earlier, the CEWS transport requirements are the same for any application server. For the CEWS transport in CE 4.0 to 4.5.0, two JVM parameters are required (Figure 2-6 on page 20): -Dwasp.location=c:\Program Files\FileNet\CEClient\wsi -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.WSI

Chapter 2. Setting up development environments

19

Figure 2-6 Eclipse runtime configuration for CEWS transport JVM parameters

For the CEWS transport in CE 4.5.1 and beyond, the only JVM parameter required is the java.security.auth.login.config file, thereby eliminating the requirement to define the wasp.location parameter.

Thin client EJB transport requirements

Thin client J2EE Web applications using the EJB transport have slightly different requirements than thick client stand-alone applications because much of the application server libraries and JVM variables are already defined or available within the application server container. Thin client applications do not require explicit references to application server libraries such as the WebSphere application client, weblogic.jar, or jbossall-client.jar. The only JVM parameter required for thin client EJB transport applications is: -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.AppServer Replace AppServer with your application server such as WebSphere, WebLogic, or JBoss. The requirements to setup a thin client Web application in a J2EE application server using EJB transport would be similar to the instructions for deploying WorkplaceXT. WorkplaceXT is a Java application that communicates over the EJB transport. The main point is to set up the trust relationships between the client application server and the CE application server. Refer to the P8 installation guide (under the section for Deploying Application Engine) for the steps to deploy and configure WorkplaceXT. It describes setting up the trust relationships, the required JVM parameter, classloader settings, and others.

20

Developing Applications with IBM FileNet P8 APIs

One way to troubleshoot connectivity issues is to deploy WorkplaceXT on the client application server. If WorkplaceXT can connect to CE successfully, you can then focus on configuring your custom application in the same fashion.

Thin client CEWS transport requirements

Note: You should use EJB transport with thin clients when possible. Especially in releases prior to 4.5.1, there is a strong possibility of having classloader conflicts with application server classes. Resolving classloader issues in J2EE applications can be time-consuming and frustrating. Similar to thick clients, thin clients that are using the CEWS transport do not have a requirement for homogeneous application server versions and patches. So you can have a Weblogic thin client application communicate using the CE Java API over the CEWS transport to a CE server running on WebSphere. From CE 4.0 to 4.5.0, a thin client application that communicates over CEWS transport must copy all eight .jar files mentioned in Table 2-4 on page 22 into the custom application WAR files WEB-INF\lib directory. Example steps for a WebSphere thin client application are as follows: 1. Create and deploy a custom EAR file in which the WAR\WEB-INF\lib folder contains the eight .jar files in Table 2-4 on page 22. 2. In the WebSphere administration console, set the EAR and WAR classloader mode to parent last. (WebSphere 6.0 refers to this as parent last; WebSphere 6.1 refers to it as application first). 3. Install the CE client libraries with the Client installer as mentioned previously. 4. Add the following JVM arguments by using the WebSphere administration console and selecting server Process Definition Java Virtual Machine Generic JVM arguments: -Dwasp.location=c:\Program Files\FileNet\CEClient\wsi -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.WSI

Chapter 2. Setting up development environments

21

Table 2-4 Thin client CEWS transport library requirements for CE 4.0 to 4.5.0 Application server Applies to all application servers Required .jar files wasp.jar jaxrpc.jar jetty.jar runner.jar saaj.jar wsdl_api.jar activation.jar builtin_serialization.jar These eight .jar files can be obtained from the c:\Program Files\FileNet\CEClient\wsi\lib directory. See Figure 2-5 on page 19.

For CE 4.5.1 and later, there are no requirement differences between a thick client stand-alone application and a thin client application. The three .jar files that are still required for a thin client application using CEWS transport are listed in Table 2-5.
Table 2-5 Thin client CEWS transport library requirements for CE 4.5.1 and beyond Application server Applies to all application servers Required .jar files stax-api.jar xlxpScanner.jar xlxpScannerUtils.jar

With the 4.5.1+ implementations, the only required JVM parameter is: -Djava.security.auth.login.config=c:\Program Files\FileNet\CEClient\config\samples\jaas.conf.WSI

2.1.4 Sample Content Engine Java API application setup in Eclipse

This book provides a list of CE Java API code samples. In addition to this book, you can find a number of other programming samples on the IBM Web site: http://www.ibm.com/support/docview.wss?rs=3278&uid=swg27010422 We use one of the samples found there to demonstrate setting up a Java development environment on Eclipse. At this Web address, scroll to the FileNet P8 Platform Developer Samples section and select FileNet Content Engine Java API Demo Sample Code.

22

Developing Applications with IBM FileNet P8 APIs

Download the Demo_Java.zip package. Use the following steps to configure and run this demonstration application in Eclipse: 1. Extract the Demo_Java.zip to a folder on your development environment. We refer to this folder where Demo_Java.zip is extracted to as <DEMO_HOME>. 2. Create a Java project in Eclipse. For the demo application, we use the project name ContentEngineDemo. 3. Add all the required CE .jar files as listed in 2.1.1, Required libraries on page 14. This demonstration is an example of a thick client application. Figure 2-7 shows the required .jar files for CE.

Figure 2-7 Required .jar files for CE and CEWS transport

4. For the demonstration (Demo) application, we use the CEWS transport. Add the .jar file listed in 2.1.3, Thick client versus thin client requirements on page 15. Figure 2-7 shows the required .jar file for thick client CEWS transport. 5. Create a package called cesample under the src folder in Eclipse. See Figure 2-8.

Figure 2-8 ContentEngineDemo create cesample package

6. Import all the classes from the folder <DEMO_HOME>/src/cesample as shown in Figure 2-9 on page 24.

Chapter 2. Setting up development environments

23

Figure 2-9 Imported demo Java classes

7. Create a Java application runtime profile in Eclipse for the MainFrame.java class. See Figure 2-10.

Figure 2-10 Java Application runtime profile for MainFrame.java

8. Add all the required JVM parameters as listed in 2.1.3, Thick client versus thin client requirements on page 15. See Figure 2-11 on page 25.

24

Developing Applications with IBM FileNet P8 APIs

Figure 2-11 Runtime JVM parameters for MainFrame.java

9. Run the Java class MainFrame.java as a Java Application from Eclipse. See Figure 2-12.

Figure 2-12 Demo application main page

Chapter 2. Setting up development environments

25

10.Enter the following input values to test the CE connectivity. For this example, we use the CEWS transport to connect to CE: CE URI as: http://<server-name>:<port-number>/wsi/FNCEWS40MTOM JAAS Stanza as FileNetP8WSI Valid user name Password of user 11.After entering the input values, click Test Connection. The demo application connects to CE and retrieves the domain name along with a list of existing object stores. Figure 2-13 shows a successful connection and lists the COLL, and E2E object stores.

Figure 2-13 Demo application: Test Server Connection

12.You can now click the other tabs of the Demo application such as Create Doc, Get Doc, and Create Folder, to test each feature. A Readme.doc file accompanies the Demo application. Refer to it for further information about using the Demo application.

26

Developing Applications with IBM FileNet P8 APIs

2.2 Process Engine Java development setup

PE relies on CE for authentication and directory service access operations. This section assumes a development environment is properly configured for CE and the selected transport protocol.

2.2.1 Required libraries

The PE API consists of the following libraries: pe.jar pe3pt.jar peResources.jar These three files can be obtained by running the PE Client Installer. By default, the PE client is installed to c:\Program Files\FileNet\BPMClient. The .jar files are located in the files subdirectory, as shown in Figure 2-14.

Figure 2-14 Process Engine Client libraries installation directory

In Eclipse, a PE development environment looks similar to Figure 2-15 on page 28.

Chapter 2. Setting up development environments

27

Figure 2-15 Eclipse jar files for Process Engine development

Java virtual machine parameters

Ensure the Java virtual machine (JVM) parameters required to set up CE for Java development are already configured. Add the following parameter to set up PE for Java development (and see Figure 2-16): -Dfilenet.pe.bootstrap.ceuri=[Content Engine URI] Add the following parameter to use EJB transport with WebSphere: -Dfilenet.pe.bootstrap.ceuri=iiop://ceserver:2809/FileNet/Engine Add the following parameter to use CEWS transport with WebSphere: -Dfilenet.pe.bootstrap.ceuri=http://ceserver:9080/wsi/FNCEWS40MTOM/

Figure 2-16 Process Engine JVM parameters

When the PE Java development project has the WcmApiConfig.properties file in a directory as part of the classpath, there is no need to add the filenet.pe.bootstrap.ceuri parameter. This setting is taken from the WcmApiConfig.properties values (see Example 2-1 on page 29 for using with EJB transport with WebSphere and Example 2-2 on page 29 for using with CEWS transport with WebSphere).

28

Developing Applications with IBM FileNet P8 APIs

Example 2-1 WcmApiConfig.properties using EJB transport with WebSphere

RemoteServerUrl = cemp:iiop://ceserver:2809/FileNet/Engine RemoteServerUploadUrl = cemp:iiop://ceserver:2809/FileNet/Engine RemoteServerDownloadUrl = cemp:iiop://ceserver:2809/FileNet/Engine ...

Example 2-2 WcmApiConfig.properties using CEWS transport with WebSphere

RemoteServerUrl = cemp:http://ceserver:9080/wsi/FNCEWS40MTOM/ RemoteServerUploadUrl = cemp:http://ceserver:9080/wsi/FNCEWS40MTOM/ RemoteServerDownloadUrl = cemp:http://ceserver:9080/wsi/FNCEWS40MTOM/ ...

2.3 .NET environment setup for CE and PE

In addition to Java, IBM FileNet P8 offers connectivity and API sets for the Microsoft .NET 2.0 environment. Configuring a client workstation or server to use the P8 APIs typically involves installing the prerequisites and adding references to the P8 API library DLL or Web Services Description Language (WSDL) file. In this section, we provide examples of setting up a Visual Studio.NET 2005 development environment for: CE .NET API Content Engine Web Services API Process Engine Web Services API

2.3.1 Prerequisites
For a .NET environment, the two primary prerequisites that apply to both the CE and PE are: Microsoft .NET Runtime 2.0 Microsoft Web Services Enhancements 3.0

2.3.2 Running the sample application supplied by IBM

After the prerequisites have been installed, you can run the sample applications supplied by IBM. If they come compiled, you run them simply by double-clicking on the supplied executable; however, if they are source-only, then you must compile them using Microsoft Visual Studio.NET.

Chapter 2. Setting up development environments

29

2.3.3 Configuring VisualStudio.NET 2005

To build an application in VisualStudio using the P8 API, create the solution and then reference the appropriate .NET assemblies, as follows: 1. Install the CE .NET API (if not already installed). 2. Start Visual Studio 2005. 3. Create the solution. 4. Add a reference to the Microsoft.Web.Services3 assembly by selecting Project Add Reference .NET Select Microsoft.Web.Services3.

Content Engine .NET API

CE 4.x provides a native .NET API. To use the CE .NET API, install the CE .NET API library using one of the following ways: Run the IBM FileNet CE install CD and selecting install the .NET API. Copy the FileNet.Api.dll to the target system. Bundle the above FileNet.Api.dll in with a custom application being deployed. To use the CE .NET API in Visual Studio.NET 2005, add a reference to the FileNet.Api.dll by selecting Project Add Reference File Find FileNet.Api.dll, as shown in Figure 2-17 on page 31.

30

Developing Applications with IBM FileNet P8 APIs

Figure 2-17 Adding a reference to the CE API DLL

Content Engine Web Services

Content Engine Web Services (CEWS) can be consumed by various client environments including Java and .NET. In this example, we show how to add CEWS in Visual Studio.NET 2005. In Visual Studio .NET 2005, add a Web Reference to the Content Engine WSDL, as follows (see Figure 2-18 on page 32): 1. Select Project Add Web Reference. 2. Enter: http://<server-name>:<port-number>/wsi/FNCEWS40MTOM 3. Give the reference a name (the samples use CEWSI). 4. Click Add Reference.

Chapter 2. Setting up development environments

31

Figure 2-18 Adding a Content Engine Web Service .NET Web Reference

Process Engine Web Services API

Similar to CEWS, Process Engine Web Services (PEWS) can also be consumed by various client environments including Java and .NET. In this example, we show how to add PEWS in Visual Studio.NET 2005. In Visual Studio.NET 2005, add a Web Reference to the Process Engine WSDL as follows (see Figure 2-19 on page 33): 1. Select Project Add Web Reference. 2. Enter: http://<server-name>:<port-number>/wsi/ProcessEngineWS 3. Give the reference a name (the samples use peWS). 4. Click Add Reference.

32

Developing Applications with IBM FileNet P8 APIs

Figure 2-19 Adding a Process Engine Web Service .NET Web Reference

2.4 PE REST API sample code development setup

To demonstrate the PE REST Service API, we include a sample application with this book. To run the PE REST sample application, you need to create a Web application in Eclipse. To create and deploy the application follow these steps: 1. Download Dojo library version 1.0.2 from the following Dojo toolkit Web site and save it to your local hard disk: http://download.dojotoolkit.org/release-1.0.2/dojo-release-1.0.2.zip 2. In Eclipse, select File New Other to open the wizard dialog. Select Dynamic Web Project. See Figure 2-20 on page 34.

Chapter 2. Setting up development environments

33

Figure 2-20 Create New Web Application

3. In the next dialog, specify the project name PERESTSample. Then, click Finish to create the project. See Figure 2-21 on page 35.

34

Developing Applications with IBM FileNet P8 APIs

Figure 2-21 Specify the Project Name

4. Extract the Dojo library .zip file to a temporary folder and copy the Dojo library into project folder dojoroot. After this step, the project structure looks like Figure 2-22.

Figure 2-22 Code structure after you copy Dojo library

5. Create a test page to run the sample code for PE REST API under folder samples. See Figure 2-23 on page 36.

Chapter 2. Setting up development environments

35

Figure 2-23 Test page for PE REST API

6. In the testPEREST.html test page, copy and paste the code skeleton shown in Example 2-3.
Example 2-3 Code skeleton to test PE REST API

<html> <head> <title>Test PE REST API</title> <script type="text/javascript" src="../dojoroot/dojo/dojo.js" djConfig="isDebug:true"></script> <script type="text/javascript"> dojo.addOnLoad(function(){ // copy the sample code here }) </script> </head> <body> </body> </html> 7. Copy the sample code of PE REST API into the body of the anonymous function called by dojo.addOnLoad. Example 2-4 on page 37 shows a code example.

36

Developing Applications with IBM FileNet P8 APIs

Example 2-4 A code sample to run PE REST API <html> <head> <title>Test PE REST API</title> <script type="text/javascript" src="../dojoroot/dojo/dojo.js" djConfig="isDebug:true"></script> <script type="text/javascript"> dojo.addOnLoad(function(){ // Ensure to be authenticated with the Application Server container var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; var applicationSpaceName = "GenericApproval"; // Construct the URI for roles var url = baseURL + "appspaces/" + applicationSpaceName + "/myroles"; // Use GET method to retrieve the roles dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { // Callback to handle the data for(var roleName in data) { var role = data[roleName]; console.log("AuthoredName:" + roleName); console.log("Display Name:" + role.name); console.log("URI:" + role.URI); } }, error: function(data) { // The error callback console.dir(data); } } ); }) </script> </head> <body> </body> </html>

8. After you create the project and sample code, deploy this project to WebSphere to test it. In Eclipse, select File Export to open the export wizard. In the pop-up dialog, select WAR File, and save it to a folder. See Figure 2-24 on page 38.

Chapter 2. Setting up development environments

37

Figure 2-24 Export the project as WAR file

9. Deploy the WAR file to WebSphere. The context root for the deployed Web application should be PERESTSample. 10.Access the test page to run the sample code for PE REST API. The Web address of the test page is: http://<server-name>:<port-number>/PERESTSample/testPEREST.html Note: Before you test the sample code, log in to WorkplaceXT first to ensure you can be authenticated with the application server container.

2.5 ECM Widgets development setup

ECM Widgets are explained in detail in Chapter 6, Advanced Process Engine API programming on page 167. It also covers the development environment requirements.

38

Developing Applications with IBM FileNet P8 APIs

Chapter 3.

Introduction to Content Engine API programming

This chapter provides an introduction to developing with the Content Engine (CE) APIs. It covers the basic principles and concepts to create, find, and work with CE objects. Most of the code snippets we show in this chapter are from the sample applications that we created for this book. For more details about the sample applications, refer to Chapter 7, Sample applications for Fictional Auto Rental Company A on page 223. This chapter discusses the following topics: Content Engine API class overview Making the initial connection Exception handling Creating, retrieving, updating, and deleting objects getInstance() versus fetchInstance() Querying Viewing documents Batching and batch execution

Copyright IBM Corp. 2009. All rights reserved.

39

3.1 Content Engine API class overview

The CE API is divided into logical groups of related items that use packages in Java and namespaces in .NET. For ease of use, the class structure generally mirrors the classes in CE metadata (for example, documents are accessed by using the Document and DocumentSet classes, folders with Folder and FolderSet classes, and so on). The API also offers a set of Factory classes that exist to create new objects and fetch existing objects from the CE. In this book, we describe the two primary CE APIs: CE Java API CE .NET API The Java and .NET APIs are similar, varying only in stylistic naming conventions. Classes and interfaces in packages that are directly under com.filenet.api (for Java) and namespace FileNet.Api (for .NET) are exposed, meaning they are available for your use. Unless specifically documented otherwise, classes in any other package or namespace are not exposed; they are strictly internal and not supported for external use. For CE development, the solution platform and the business use cases dictate what API to use. For example, if you plan to implement a Windows-based thick client, using the .NET API might be appropriate. However, for event action development or in a pure Java shop, the Java API would be appropriate. In this book, we include code snippets that show how to use the CE APIs for many use cases that are related to CE development. Because both the Java and .NET APIs are feature-compatible, most code snippets in this chapter are shown in both C# and Java languages. Of course, for the .NET API, any .NET-compatible language binding can be used for development.

3.1.1 Content Engine API class model

Note: Most of the CE API types are interfaces, but a few are classes. To avoid continuously repeating that we are talking about classes or interfaces, we refer to them all simply as classes.

40

Developing Applications with IBM FileNet P8 APIs

The CE API has over a hundred classes. They are arranged into a number of packages or namespaces, organized by functional areas, which include: Core Core (com.filenet.api.core and FileNet.Api.Core) provides classes related to the core business objects and other classes that are used in most applications. Examples include: Java: EntireNetwork, Document, Folder, CustomObject, Factory, UpdatingBatch, Connection .NET: EntireNetwork, IDocument, IFolder, ICustomObject, Factory, UpdatingBatch, Connection Meta Meta (com.filenet.api.meta and FileNet.Api.Meta) provides classes for holding immutable metadata for CE classes and properties. Examples include: Java: ClassDescription, PropertyDescription, PropertyDescriptionDateTime .NET: ClassDescription, IPropertyDescription, IPropertyDescriptionDateTime Admin Admin (com.filenet.api.admin and FileNet.Api.Admin) provides classes used in the administration of a CE, including classes for updating metadata objects. Examples include: Java: ClassDefinition, PropertyDefinition, PropertyDefinitionDateTime, DirectoryConfiguration, PEConnectionPoint, ServerInstance, TableDefinition .NET: ClassDefinition, IPropertyDefinition, IPropertyDefinitionDateTime, IDirectoryConfiguration, IPEConnectionPoint, IServerInstance, ITableDefinition Security Security (com.filenet.api.security and FileNet.Api.Security) provides classes related to authentication, authorization, and user-specific and group-specific data. Examples include: Java: User, Group, AccessPermission, MarkingSet .NET: IUser, IGroup, IAccessPermission, IMarkingSet

Chapter 3. Introduction to Content Engine API programming

41

Query Query (com.filenet.api.query and FileNet.Api.Query) provides classes related to constructing and performing CE searches. Examples include: Java: SearchScope, RepositoryRow .NET: SearchScope, IRepositoryRow Collection Collection (com.filenet.api.collection and FileNet.Api.Collection) provides type-safe classes that are related to collections of objects. Examples include: Java: FolderSet, ContentElementList .NET: IFolderSet, IContentElementList Events Events (com.filenet.api.events, not available in .NET) provides classes representing events triggered on CE objects, as well as classes related to handling those events raised within the CE. Because all events are run inside the CE context, there is no equivalent .NET package. Examples include: Java: FileEvent, UnfileEvent, EventAction, InstanceSubscription Property Property (com.filenet.api.property and FileNet.Api.Property) provides classes related to CE properties and property values. Examples include: Java: Properties, PropertyDateTime, PropertyDateTimeList, PropertyFilter .NET: IProperties, IPropertyDateTime, IPropertyDateTimeList, IPropertyFilter Constants Constants (com.filenet.api.constants and FileNet.Api.Constants) provides classes defining collections of related, type-safe constant values. Examples include: Java and .NET: AccessRight, Cardinality, DatabaseType, PropertyNames, ReservationType Figure 3-1 on page 43 shows the relationship between key classes that are used for creating, storing, searching, and retrieving objects within the CE.

42

Developing Applications with IBM FileNet P8 APIs

Factory Classes

ObjectStore

Creates And Retrieves

Document

CustomObject Uses Folder Contains

Contains

Property Collection

Contains

SearchScope

IndependentObjectSet

Property

SearchSQL

RepositoryRow

Contains

Figure 3-1 Class relationships and general object flow

3.2 Making the initial connection

Before working with the CE, an initial connection must be made. Making this initial connection to the CE typically involves the following steps: 1. 2. 3. 4. Authenticate or prepare the UserContext. Get a connection object. Get a domain object. Get an object store object.

After the connection has been made, assuming the application user is properly authenticated, the core objects (Connection, Domain, and ObjectStore) will be available for use.

Connection
A connection object tells the API how to connect to the CE server, primarily through the Uniform Resource Identifier (URI). The API then deduces both the method for connecting and the location of the CE from the URI. A connection object does not contain any user identification information. A connection is a lightweight API object that does not maintain any server state or tie up any server resources except when a server call is actually in progress.

Chapter 3. Introduction to Content Engine API programming

43

Domain
A domain is an object holding IBM FileNet P8 resources at or above the object store level. These resources include: Object stores Fixed content devices Marking sets

ObjectStore
An object store is a collection of classes, objects, and metadata, and represents a logical division of data within a CE domain. Other objects can be retrieved directly from the object store or can be used with the Factory methods to create and retrieve objects such as: Documents Custom objects Folders Class and property definitions

3.2.1 User authentication

When making an initial connection to the CE, the user must be authenticated to access resources and to retrieve objects. Java and .NET have slightly different requirements for user authentication.

Authentication in Java
For a Java application, the three primary methods for authenticating a user are: Container managed authentication A custom JAAS LoginModule that authenticates directly with the authentication provider The UserContext object with its pushSubject method For container-managed and LoginModule-based authentication, work with your application server provider or internal administration and development teams for the appropriate code. Although the ideal situation for authentication is either container-managed or LoginModule-based authentication, the Java API does provide a convenience method for authentication through the UserContext object. When doing a UserContext-based login, you create a Java Authentication and Authorization Service (JAAS) subject and push it onto the UserContext stack. After all CE-related operations are complete, you should pop the Subject back off the

44

Developing Applications with IBM FileNet P8 APIs

stack. See Example 3-1 for how to create the subject, push it onto the UserContext, and then pop it off the stack.
Example 3-1 UserContext authentication in Java

Subject subject = UserContext.createSubject(connection, "username, "password", null); UserContext.get().pushSubject(subject); try { // do something } finally { UserContext.get().popSubject(); }

User authentication in .NET

For .NET-based applications, authentication is always handled through the UserContext object. The UserContext requires a valid SecurityToken be passed to either the SetProcessSecurityToken or the SetThreadSecurityToken static method. In a .NET application, this security token can be either a UsernameToken provided by Microsoft Web Services Enhancements or a KerberosToken for Kerberos-based single sign on. Example 3-2 shows basic UsernameToken authentication.
Example 3-2 UsernameToken authentication in .NET

UsernameToken token = new UsernameToken("username", "password", PasswordOption.SendPlainText); UserContext.SetProcessSecurityToken(token);

3.2.2 Java
Java offers two data transport layers: the native Enterprise JavaBeans (EJB) transport offered by the application server or the CE Web Services (CEWS) transport. The transport used is indicated by the CE server URI passed into the Connection object. Table 3-1 on page 46 shows examples of available CE connection URIs.

Chapter 3. Introduction to Content Engine API programming

45

Note: For the Web services transport, the three endpoints are: FNCEWS40MTOM, FNCEWS40DIME, and FNCEWS40SOAP. The primary endpoint is MTOM. The other two endpoints, DIME and SOAP, exist for compatibility purposes only. All examples in this book use MTOM as the endpoint. For historical reasons, CEWS transport is sometimes referred to as Web Services Interface (WSI) transport or Web services transport. These are just naming differences and all refer to exactly the same thing. The CEWS transport is exactly the same set of Web services as would be used by someone calling CEWS directly (assuming that the persons uses a FNCEWS40* endpoint).
Table 3-1 Content Engine connection URIs Protocol Web Services EJB/WebSphere EJB/Weblogic EJB/JBoss URI http://server:port/wsi/FNCEWS40MTOM/ iiop://server:port/FileNet/Engine t3://server:port/FileNet/Engine jnp://server:port/FileNet/Engine

Getting an object store in Java consists of the following steps (see Example 3-3 on page 47 and Example 3-4 on page 47): 1. Get the connection object from Factory.Connection. 2. Create the Java Authentication and Authorization Service (JAAS) subject either by using UserContext.createSubject and pushing that subject into the UserContext, or by any way that creates a valid JAAS subject. 3. Get the domain object by using Factory.Domain.getInstance(). 4. Fetch the object store by using Factory.ObjectStore.fetchInstance(). Note: The main differences between CEWS and EJB when connecting are the URI and the JAAS stanza, as shown in bold in Example 3-3 and Example 3-4 on page 47.

46

Developing Applications with IBM FileNet P8 APIs

Example 3-3 Getting the initial connection with Web services transport in Java

// Set the constants // Note: use /wsi/FNCEWS40MTOM/ for CEWS transport String uri = "http://ceserver:port/wsi/FNCEWS40MTOM/"; String username = "username"; String password = "password"; // Get the connection Connection conn = Factory.Connection.getConnection(uri); // Get the user context UserContext uc = UserContext.get(); // Build the subject using the FileNetP8WSI stanza // Use the FileNetP8WSI stanza for CEWS uc.pushSubject( UserContext.createSubject(conn,username,password,"FileNetP8WSI" ) ); try { // Get the default domain Domain domain = Factory.Domain.getInstance(conn, null); // Get an object store ObjectStore os = Factory.ObjectStore.fetchInstance(domain, "ObjectStoreName", null); } finally { uc.popSubject(); }
Example 3-4 Getting the initial connection with EJB transport in Java

// Set // Use String String String

the constants /FileNet/Engine for EJB uri = "http://ceserver:port/FileNet/Engine"; username = "username"; password = "password";

// Get the connection Connection conn = Factory.Connection.getConnection(uri);

Chapter 3. Introduction to Content Engine API programming

47

// Get the user context UserContext uc = UserContext.get(); // Build the subject using the FileNetP8 stanza // Use FileNetP8 for the EJB transport (also the default) uc.pushSubject( UserContext.createSubject(conn,username,password,"FileNetP8" ) ); try { // Get the default domain Domain domain = Factory.Domain.getInstance(conn, null); // Get an object store ObjectStore os = Factory.ObjectStore.fetchInstance(domain, "ObjectStoreName", null); } finally { // Pop the subject off the UserContext stack uc.popSubject(); } Note: There are other ways to handle authentication. All the API requires is a valid JAAS Subject. To use alternate authentication, work with your application server provider.

3.2.3 .NET
Unlike the Java API, the .NET API uses only the Web services transport for communicating with the CE. Making the initial connection to the CE and retrieving an object store involves the following steps: 1. 2. 3. 4. Create the UsernameToken for authentication. Pass the token to the UserContext. Get the domain using Factory.Domain.GetInstance(). Fetch the object store Factory.ObjectStore.FetchInstance().

Example 3-5 on page 49 is adapted from: ITSO.AutoRental.DataAccess.ContentEngineConnector.Connect() It shows the process of getting the initial connection, retrieving the domain, and retrieving an object store in C#.

48

Developing Applications with IBM FileNet P8 APIs

Example 3-5 Getting the initial connection in C#

public void Connect() { string uri = "http://ceserver:port/wsi/FNCEWS40MTOM/" string username = "username"; string password = "password"; // Create a security token using the username and password UsernameToken token = new UsernameToken(username, password, PasswordOption.SendPlainText); // Associate this UserContext with the whole process UserContext.SetProcessSecurityToken(token); // Now get the connection IConnection conn = Factory.Connection.GetConnection(uri); // Get the default domain _domain = Factory.Domain.GetInstance(conn, null); // Do the initial connection _os = Factory.ObjectStore.FetchInstance(_domain, objectStore, null); }

3.3 Exception handling

When the CE signals an error, it throws a runtime exception. Within the API, there is only one type of exception: EngineRuntimeException. Instead of multiple exception types for the various error conditions available, the EngineRuntimeException carries with it an exception code that indicates the actual failure. Calling EngineRuntimeException.getExceptionCode() will return the exception code for the specific exception condition. After an exception has been caught, it is up to the developer to determine if the application can recover from the exception or if the exception should be logged or displayed to the user to indicate that something failed. A complete list of the exception codes is available in the JavaDoc and .NET DNA online reference as properties of ExceptionCode.

Chapter 3. Introduction to Content Engine API programming

49

Example 3-6 and Example 3-7 on page 51 show an attempt to fetch an object and, in case of failure, attempt to create the object. This code has been adapted from the following method in the sample application Java code: RentalActivityHandler.getRentalNumber()
Example 3-6 Exception handling with Java

while (confirmationId == null) { try { Integer temp; if( createObject ) // Calls the create custom object code temp = createRentalNumber(dom,os); else // Fetches the custom object temp = getRentalNumber(os); confirmationId = temp; } catch( EngineRuntimeException ex) { ExceptionCode e = ex.getExceptionCode(); if( e == ExceptionCode.E_OBJECT_NOT_FOUND ) { // We can't find the requested object, so we should create it createObject = true; continue; } else if ( e == ExceptionCode.E_NOT_UNIQUE ) { // We tried creating an object and it's already there, use it createObject = false; continue; } // We hit an error we can't handle, so rethrow the exception throw ex; } }

50

Developing Applications with IBM FileNet P8 APIs

Example 3-7 Exception handling in C#

int confirmationId = -1; bool createObject = false; while( confirmationId == -1 ) { try { int temp; if( createObject ) temp = CreateRentalNumber(dom,os); else temp = GetRentalNumber(os); confirmationId = temp; } catch( EngineRuntimeException ex) { ExceptionCode e = ex.GetExceptionCode(); if( e == ExceptionCode.E_OBJECT_NOT_FOUND ) { // We can't find the requested object, so we should create it createObject = true; continue; } else if ( e == ExceptionCode.E_NOT_UNIQUE ) { // We tried creating an object and it's already there, use it createObject = false; continue; } // We've hit an error we can't handle, so rethrow the exception throw ex; } }

Chapter 3. Introduction to Content Engine API programming

51

3.4 Creating, retrieving, updating, and deleting objects

Within the CE, there are a number of common objects that an application will have to use. These include: Documents Folders Custom Objects For these objects, there are a number of common needs in accessing these objects: Creating objects Working with properties Retrieving objects Deleting objects Retrieving content Working with property filters Working with versions (described in 4.2, Versioning on page 90.)

3.4.1 Pending actions

Within the CE API, all changes (creations, updates, and deletions) are treated as local changes only until the .save() (Java) or .Save() (.NET) method has been called. Instead of immediately saving the changes, the API creates a pending action describing the change. Changes to objects can be committed individually or as a batch. See 3.8, Batching and batch execution on page 81 for details about batch execution.

3.4.2 Creating objects

To create any object with the CE API: 1. Use the Factory convenience methods to create the object locally. 2. Apply any metadata locally to the newly created object. See 3.4.3, Working with properties on page 53 for details. 3. For objects that can have versions, optionally handle check-in. 4. Commit the object to the CE, creating the object in the CE. Example 3-8 on page 53 and Example 3-9 on page 53 are adapted from the ITSONewVehicleActivity.java program and show how to create a CustomObject using both Java and C#.

52

Developing Applications with IBM FileNet P8 APIs

Note: When creating objects, specifying the object ID for the object being created is possible by setting the value of the ID property. In addition, the ID class offers a method to create an IBM FileNet P8 object ID. Unless you have a specific reason for doing so, you should not explicitly specify the object ID. An ID will be created by the API or by the CE server.
Example 3-8 Creating objects in Java

CustomObject activity = Factory.CustomObject.createInstance(os, "ITSOIdleActivity"); // Property values are set here // Save the new IdleActivity activity.save(RefreshMode.NO_REFRESH);
Example 3-9 Creating objects in C#

ICustomObject activity = Factory.CustomObject.CreateInstance(os, "ITSOIdleActivity"); // Property values are set here // Save the new IdleActivity activity.Save(RefreshMode.NO_REFRESH);

3.4.3 Working with properties

Each object in the CE has metadata. This metadata can range from the date that the object was created to the title of a document to a binary large object (BLOB) of data. Metadata is stored and accessed using the properties of the given object. The CE provides the following metadata types for use: Text 32-bit Integer 64-bit Floating Point Binary Boolean Date/Time GUID Object

Chapter 3. Introduction to Content Engine API programming

53

Properties can also can be either single-valued or multi-valued, meaning the property can store either a single value or one or more values.

Object properties
Object-valued properties (OVP) work a bit differently than the other property types. For more information, see 4.3.1, Object-valued properties on page 94.

Properties collections
All CE objects have properties, or metadata, that describe that object. When an object has been retrieved with the API, these properties are stored locally in a properties collection. This collection holds property values for existing objects or stores new or updated values to be committed to the CE. The primary property collection classes are Properties for Java and IProperties for .NET and are typically accessed with the .getProperties() method in Java or the .Properties property in .NET. Although the properties collection can be iterated or enumerated, a more common approach is to access a given property directly through the .get(name) method or .Item[name] property, which returns a Property object that represents that property and its value (or values). Dynamically determining property type requires a cascade of instanceof in Java or is in .NET. Note: For clarity, the following code examples omit the save method call, so, as written, the examples would not commit any property changes.

Setting single-valued property value

To set the value of a single-valued property in Java: 1. Create it locally or retrieve the object. 2. Get the properties collection. 3. Add the value by using the .putValue() method. Example 3-10 on page 55, adapted from the ITSONewVehicleActivity.java program, shows how to retrieve the properties of a document and to update the value of the ITSOStartDate property with the current date and time.

54

Developing Applications with IBM FileNet P8 APIs

Example 3-10 Setting a single-valued property in Java

// Get the properties collection Properties props = activity.getProperties(); // Get the start date Date start = new Date(System.currentTimeMillis()); // Set the properties props.putValue("ITSOStartDate", start); The .NET API works slightly different from the Java API. To set a single value property in .NET: 1. Create locally or retrieve the object. 2. Get the properties collection. 3. Set the value of the property by using the properties["propertyname"] convention. Note: For clarity, the following code examples omit the save method call, so, as written, the examples would not commit any property changes. Example 3-11 shows how to retrieve the properties of a document and to update the value of the ITSOStartDate property with the current date and time.
Example 3-11 Getting and setting properties in C#

// Get the properties collection for a Document IProperties properties = d.Properties; Property["ITSOStartDate"] = DateTime.Now;

Setting multi-valued property values

To set a multi-valued property in Java: 1. Retrieve or create the object. 2. Set the value of the property to be a list using the .putValue() method. Note: Setting the value of a multi-valued object-valued property is not possible. For details, see 4.3.1, Object-valued properties on page 94. Example 3-12 on page 56 shows how to retrieve the properties of a document, create a list of integer properties with two values, and to set the local value of the property "MutliValuedIntProperty" to be those values.

Chapter 3. Introduction to Content Engine API programming

55

Example 3-12 Setting a multi-valued property in Java

// Get the properties collection for a Document Properties properties = document.getProperties(); // Create a multi-value integer list Integer32List list = Factory.Integer32List.createList(); list.add(100); list.add(212); // Add that to the properties for the document properties.putValue("MutliValuedIntProperty",list); To set a multi-valued property in .NET: 1. Create or retrieve the object. 2. Get the properties collection. 3. Create the value list. 4. Add the values. 5. Set the value of the property to be the list using the properties["name"] convention. See Example 3-13.
Example 3-13 Setting multi-valued property in C#

IProperties properties = d.Properties; // Create a multi-value integer list and add 2 values IInteger32List list = Factory.Integer32List.CreateList(); list.Add(100); list.Add(212); property["MultiValuedIntProperty"] = list;

Retrieving property values

To retrieve property values in Java: 1. Retrieve the object. 2. Get the properties collection. 3. Get the value (or values) by using either of the following methods: .get<type>Value() .get<type>ValueList()

56

Developing Applications with IBM FileNet P8 APIs

Example 3-14 shows, using Java, how to retrieve the value of a single-valued string property, StringProperty, and a multi-valued ID property, RelatedIds.
Example 3-14 Retrieving property values from a CE object using Java

// Get the properties collection from a custom object Properties properties = o.getProperties(); // Get a string value and print to the console System.out.println( properties.getStringValue("StringProperty") ); // Get a multi-value ID property IdList ids = properties.getIdListValue("RelatedIds"); // Iterate through the list of values, printing them to the console for( Iterator i = ids.iterator(); i.hasNext(); ) { // Get the value and cast to an Id property type Id property = (Id)i.next(); // Print the value to the console System.out.println(property.toString()); } To retrieve property values in C#: 1. Retrieve the object. 2. Get the properties collection. 3. Get the value(s) using .Get<type>Value() or .Get<type>ValueList() (for example, GetStringValue or GetInt32ListValue()) Example 3-15 shows how to retrieve the value of a single-valued string property, StringProperty, and a multi-valued ID property, RelatedIds.
Example 3-15 Retrieving property values from a CE object using C#

// Get the properties for a custom object Properties properties = o.Properties; // Write out the string property value Console.WriteLine("Property value: " + properties.GetStringValue("StringProperty")); // Get a multi-valued ID property IIdList ids = properties.GetIdListValue("RelatedIds"); // Loop through the properties

Chapter 3. Introduction to Content Engine API programming

57

for( IEnumerator i = ids.GetEnumerator(); i.MoveNext(); ) { // Cast the current property to an Id property Id property = (Id)i.Current; // Print out the value to the console Console.WriteLine(property.ToString()); }

Properties in local cache

The CE APIs give you explicit control over whether or not any particular property is in the local cache. For example, when retrieving an object from the CE, having the result exclude some of the properties is possible. For details about how, see 3.4.7, Working with property filters on page 66. If a PropertyFilter has been applied during the fetchInstance, refresh, or fetchProperties calls, not all properties of an object will be available. Trying to get any of the missing properties will result in an EngineRuntimeException exception.

3.4.4 Retrieving objects

Objects from CE are typically retrieved by using the Factory methods getInstance() or fetchInstance(), depending on how you will use the instantiated object and whether the object must be used immediately. For the differences between the two, see 3.5, getInstance() versus fetchInstance() on page 70. There are methods for both retrieval techniques using either an object ID or a path as a document retrieval identifier. In addition, the API can be directed whether to retrieve all properties or a subset of the properties for the requested object. See 3.4.7, Working with property filters on page 66 for the details about using property filters. Example 3-16 on page 59 and Example 3-17 on page 59 show how to retrieve documents, folders, and custom objects both by ID and by path. Note: The following examples all do an immediate fetch with no property filter for demonstration purposes. This approach allows for immediate usage of the objects. However, it may might be the best for performance. Review 3.5, getInstance() versus fetchInstance() on page 70 to understand when to get and when to fetch.

58

Developing Applications with IBM FileNet P8 APIs

Example 3-16 Fetching objects in Java // Retrieve document by path Document d1 = Factory.Document.fetchInstance(os,"/path/to/doc",null); // Retrieve document by ID Document d2 = Factory.Document.fetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null); // Retrieve folder by path Folder f1 = Factory.Folder.fetchInstance(os,"/path/to/folder",null); // Retrieve folder by ID Folder f2 = Factory.Folder.fetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null); // Retrieve custom object by path CustomObject o1 = Factory.CustomObject.fetchInstance(os,"/path/to/object", null); // Retrieve custom object by ID CustomObject o2 = Factory.CustomObject.fetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null); Example 3-17 Fetching objects in C# // Retrieve by path IDocument d1 = Factory.Document.FetchInstance(os,"/path/to/doc",null); // Retrieve by ID IDocument d2 = Factory.Document.FetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null); // Retrieve by path IFolder f1 = Factory.Folder.FetchInstance(os, "/path/to/folder", null); // Retrieve by ID IFolder f2 = Factory.Folder.FetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null); // Retrieve by path ICustomObject o1 = Factory.CustomObject.FetchInstance(os, "/path/to/object", null); // Retrieve by ID ICustomObject o2 = Factory.CustomObject.FetchInstance(os, new Id("{C6E0130B-F679-4244-9269-CCA073AE31C9}"),null);

Chapter 3. Introduction to Content Engine API programming

59

3.4.5 Deleting objects

An object can be deleted from the CE by using the .delete() method (in Java) and the .Delete() method (in .NET). Deleting an object removes both the metadata and associated content (in the case of documents and annotations). Note: Versionable objects have special behaviors related to deletion. For more information, see 4.2, Versioning on page 90. To delete an object from the CE: 1. Be sure that the object is instantiated first (with either getInstance or fetchInstance). 2. Call the delete method to add the delete operation to the objects PendingActions. 3. Call the save method to perform the delete. Example 3-18 and Example 3-19 on page 61 show how to instantiate and delete documents, folders, and custom objects. Note: In the following examples, all objects are created locally by using a getInstance call instead of fetching. This approach cuts the number of round-trips in half, improving performance, and reducing network utilization.
Example 3-18 Deleting objects in Java

// Get the object Document doc = Factory.Document.getInstance(os,"Document","/Doc"); // Mark the document for deletion locally doc.delete(); // Perform the delete doc.save(Refresh.NO_REFRESH); // Get the folder for deletion Folder folder = Factory.Folder.getInstance(os,"Folder","/Folder"); // Prepare the folder for deletion f.delete(); // Commit the delete f.save(RefreshMode.NO_REFRESH); // Get the object CustomObject o = Factory.CustomObject.getInstance(os,"CustomObject","/Object");

60

Developing Applications with IBM FileNet P8 APIs

// Prepare the object for deletion o.delete(); // Commit the deletion o.save(RefreshMode.NO_REFRESH);
Example 3-19 Deleting objects in C#

// Get the object IDocument doc = Factory.Document.GetInstance(os,"Document","/Doc"); // Mark the document for deletion locally doc.Delete(); // Perform the delete doc.Save(RefreshMode.NO_REFRESH); // Get the folder for deletion IFolder folder = Factory.Folder.GetInstance(os,"Folder","/Folder"); // Prepare the folder for deletion f.Delete(); // Commit the delete f.Save(RefreshMode.NO_REFRESH); // Get the object ICustomObject o = Factory.CustomObject.GetInstance(os,"CustomObject","/Object"); // Prepare the object for deletion o.Delete(); // Commit the deletion o.Save(RefreshMode.NO_REFRESH);

3.4.6 Retrieving content

When working with objects, it will likely be necessary to access the content directly. Retrieving the content allows the application to use or save the binary (or text) content of the object for processing elsewhere. Note: See 3.7, Viewing documents on page 78 for more information about simply viewing the object if that behavior is necessary. The CE allows a single object to have multiple content elements. This approach allows for operations, such as storing a multi-page scanned document as a series of individual files, one for each page, instead of having to store a single file with multiple pages. From a performance and network usage standpoint, this can reduce the load on the network by allowing only the desired page to be transferred instead of having to transfer the complete multi-page file just to access a single page.

Chapter 3. Introduction to Content Engine API programming

61

Finally, the CE stores a few metadata items for each content element attached, stored as a ContentTransfer. This metadata includes: Content size: Numerical size in bytes for the content element Content type: Content type of the element in MIME format Retrieval name: Optional property that indicates the original file name of the element Element sequence number: Ordered index for the content element indicating which element it is See Using ContentElement objects on page 63 for details about using ContentElement objects.

Retrieving content using Document object

To retrieve content from an object: 1. Retrieve the object. 2. Get the InputStream (Java) or Stream (.NET) for the desired content element by using .accessContentStream() (Java) or .AccessContentStream() (.NET). 3. Use the InputStream.read() (Java) or Stream.Read() (.NET) methods to retrieve the binary content. Example 3-20 and Example 3-21 on page 63, adapted from ContentEngineConnector.cs from the .NET Kiosk application, show how to retrieve a document and save its content to a file. In this case, we know we are dealing with a short, line-oriented text file. In the more general case, you would want to treat the content as an arbitrary stream of bytes.
Example 3-20 Directly accessing a documents content in Java

InputStream stream = document.accessContentStream(0); BufferedReader reader = new BufferedReader(new InputStreamReader(stream)); String text = "", line; while ((line = reader.readLine()) != null) text += line; reader.close();

62

Developing Applications with IBM FileNet P8 APIs

Example 3-21 Directly accessing a documents content in C#

Stream stream = document.AccessContentStream(0); TextReader reader = new StreamReader(stream); string text = "", line; while ((line = reader.ReadLine()) != null) text += line; reader.Close();

Using ContentElement objects

To retrieve the content collection and associated ContentElements: 1. Retrieve the object. 2. Call the get_ContentElements() method (Java) or .ContentElements property (.NET). 3. Access the appropriate content element using the .get() method (Java) or .Items[] property (.NET). 4. Optionally, iterate through the content elements by using the .iterator() (Java) or .GetEnumeration() (.NET) methods. Example 3-22 and Example 3-23 on page 65 show how to retrieve a document, loop through the content elements, and save each content element to the file system as "element#".
Example 3-22 Working with content elements in Java

public static void main(String[] args) { // Authentication and connection goes here (left out for space) // Retrieve the document Document d = Factory.Document.fetchInstance(os,"/document",null); // Get the content elements ContentElementList elements = d.get_ContentElements(); // Grab the first content element ContentTransfer element = (ContentTransfer)elements.get(0); String filename = element.get_RetrievalName(); InputStream stream = element.accessContentStream();

Chapter 3. Introduction to Content Engine API programming

63

Double size = writeContent(stream,filename); Double expected = element.get_ContentSize() if( size != expected ) System.err.println("Invalid content size retrieved"); d = Factory.Document.fetchInstance("/document2",null); elements = d.get_ContentElements(); for( Iterator i = elements.iterator(); i.hasNext(); ) { element = (ContentTransfer)i.next(); writeContent( element.accessContentStream(), "element" + element.get_ElementSequenceNumber() ); } } public static Double writeContent(InputStream s, String filename) throws IOException { // Open a buffered output stream to write the content to // a file named the same as the retrieval name BufferedOutputStream writer = new BufferedOutputStream( new FileOutputStream(filename) ); Double size = new Double(0); int bufferSize; byte[] buffer = new byte[1024]; // Loop through the content and write it to the system while( ( bufferSize = s.read(buffer) ) != -1 ) { size += bufferSize; writer.write(buffer, 0, bufferSize); } writer.close(); s.close(); return size; }

64

Developing Applications with IBM FileNet P8 APIs

Example 3-23 Working with content elements in C#

static void Main(string[] args) { // Authenticate and connect (left out for space) // Retrieve the document IDocument d = Factory.Document.FetchInstance(os,"/document", null); // Get the content elements IContentElementList elements = d.ContentElements; // Grab the first content element IContentTransfer element = (IContentTransfer)elements[0]; // Grab the retrieval name string filename = element.RetrievalName; // Get the content stream Stream stream = element.AccessContentStream(); // Write the content and compare the size expected double size = writeContent(stream,filename); if( size != element.ContentSize ) Console.Error.WriteLine("Invalid content size retrieved"); // Fetch second document d = Factory.Document.FetchInstance(os,"/document2",null); // Grab the content elements elements = d.ContentElements; // Iterate through each element and write the content for( IEnumerator i = elements.GetEnumerator(); i.MoveNext(); ) { element = (IContentTransfer)i.Current; writeContent( element.AccessContentStream(), "element" + element.ElementSequenceNumber ); } } // Convenience method that writes the content to the given file

Chapter 3. Introduction to Content Engine API programming

65

public static double writeContent(Stream stream, String filename) { byte[] buffer = new byte[4096]; int bufferSize; double size = 0; // Open a binary writer for the stream BinaryWriter writer = new BinaryWriter( File.Open(filename, FileMode.Create)); // Write the content while ((bufferSize = stream.Read(buffer, 0, buffer.Length)) != 0) { size += bufferSize; writer.Write(buffer, 0, bufferSize); } // Close the streams writer.Close(); stream.Close(); return size; }

3.4.7 Working with property filters

Property filters are optional parameters to a number of methods that fetch
objects or properties from the CE. They allow highly granular control of the objects or properties being returned. Property filters can be used two ways to help boost performance: Limiting the number of properties to be retrieved from the CE, reducing the volume of data transmitted. Including more properties through recursion, reducing the total number of requests to the CE. Recursion implicitly follows chains of object-valued properties.

Limiting requested properties

When fetching an object from the CE, it is possible to have the API retrieve all properties for the request object. However, if the application only needs a few properties to complete a procedure, fetching all properties wastes time and network resources. Instead, limiting the data returned to only the applicable properties will make the call and application more efficient.

66

Developing Applications with IBM FileNet P8 APIs

To limit data returned using property filters: 1. Directly fetch the properties using the .fetchProperties() (Java) or .FetchProperties() (.NET) methods. 2. Create a PropertyFilter object. 3. Set the include and exclude properties. 4. Set the recursion to determine how far to follow object properties. 5. Pass the object to a .fetchInstance() (Java) or .FetchInstance() (.NET) call. Note: Trying to use a property that has not been retrieved will result in an exception. If you do not use property filters, the API generally manages the retrieval of additional properties. This process avoids the exceptions but can cost extra round-trips to the server. When you use a property filter, you get explicit control over the presence or absence of all properties. Example 3-24 and Example 3-25 on page 68 show how to use property filters to control which properties are retrieved when fetching a document and fetching a documents properties.
Example 3-24 Limiting properties returned using property filters in Java

// Get the document Document d = Factory.Document.getInstance(os, "Document", "/Doc1"); // Fetch properties DocumentTitle and DateCreated d.fetchProperties(new String[] { "DocumentTitle", "DateCreated" }); // Write out the Document Title System.out.println(d.getProperties().getStringValue("DocumentTitle")); // Write out the Creator (throws EngineRuntimeException) //System.out.println(d.getProperties().getStringValue("Creator")); // Same result using PropertyFilter PropertyFilter filter = new PropertyFilter(); filter.addIncludeProperty( new FilterElement(null, null, null, "DocumentTitle DateCreated")); // Calling fetch instance instead of getInstance d = Factory.Document.fetchInstance(os, "/Doc2", filter);

Chapter 3. Introduction to Content Engine API programming

67

Example 3-25 Limiting properties returned using property filters in C#

// Retrieve the document IDocument d = Factory.Document.GetInstance(os, "Document", "/Doc1"); // Fetch Document Title and Date Created d.FetchProperties(new string[] { "DocumentTitle", "DateCreated" }); // Write out the DocumentTitle (works) Console.WriteLine(d.Properties.GetStringValue("DocumentTitle")); // Write out the Creator (throws EngineRuntimeException) //Console.WriteLine(d.Properties.GetStringValue("Creator")); // Same result using PropertyFilter PropertyFilter filter = new PropertyFilter(); filter.AddIncludeProperty( new FilterElement(null, null, null, "DocumentTitle DateCreated")); // Calling fetch instance instead of get instance d = Factory.Document.FetchInstance(os, "/Doc2", filter);

Using property filters to eliminate round trips

Property filters can also be used to return more data than normal. This approach is especially helpful when retrieving an object that has other objects as the values of properties. Instead of retrieving the object itself in one call and then the object values in multiple additional calls, adding a property filter to do recursion can cause all the objects to be returned in the initial call. For example, an application might need to retrieve not only a document, but also information about all annotations on that document. If a document has two annotations, normally this call would take three round-trips to the CE: one to fetch the document and potentially one to fetch each annotation. With an appropriately configured property filter, the retrieval can happen in a single round-trip. Example 3-26 on page 69 and Example 3-27 on page 69 both demonstrate using property filters, and are adapted from GetRentalRecord in ContentEngineConnector.cs in the sample .NET application.

68

Developing Applications with IBM FileNet P8 APIs

Example 3-26 Property filter in Java

SearchSQL sql = new SearchSQL( "select " + "ITSOCustomer, ITSOStartDate, ITSOEndDate, ITSOChargesNet," + "ITSOChargesTotal, ITSORentalAgreement ITSOVehicle," + "ITSODailyRate, ITSOConfirmationId " "from " + "ITSORentalActivity " + "where " + "ITSOConfirmationId = " + identifier); // Search the instantiated object store SearchScope scope = new SearchScope(os); // Create a property filter object PropertyFilter filters = new PropertyFilter(); filters.addIncludeProperty(2, null, null, "ContentElements Containees ITSOVehicleId ITSOVehicleYear " + "ITSOVehicleMake ITSOVehicleColor ITSOCustomerName"); IndependentObjectSet set = scope.fetchObjects(sql, 1, filters, false);
Example 3-27 Property filters in .NET

SearchSQL sql = new SearchSQL( "select " + "ITSOCustomer, ITSOStartDate, ITSOEndDate, ITSOChargesNet," + "ITSOChargesTotal, ITSORentalAgreement ITSOVehicle," + "ITSODailyRate, ITSOConfirmationId " + "from " + "ITSORentalActivity " + "where " + "ITSOConfirmationId = " + identifier); SearchScope scope = new SearchScope(_os); PropertyFilter filters = new PropertyFilter(); filters.AddIncludeProperty(2, null, null, "ContentElements Containees ITSOVehicleId ITSOVehicleYear " + "ITSOVehicleMake ITSOVehicleColor ITSOCustomerName"); IIndependentObjectSet set = scope.FetchObjects(sql, 1, filters, false);

Chapter 3. Introduction to Content Engine API programming

69

Without property filters, the operation in the examples would require four round-trip calls to the CE, as follows: Execute the search and retrieve the ITSORentalActivity custom object. Retrieve the ITSOVehicle folder properties. Retrieve the ITSORentalAgreement properties. Retrieve the ITSOCustomer properties. Using the code examples, the total number of round-trips to the CE is reduced from four to one. Figure 3-2 shows the objects and properties that are returned from the code in the examples.
ITSORentalActivity ITSOCustomer ITSOStartDate ITSOEndDate ITSOChargesNet ITSOChargesTotal {4B608AFB-E6EA-4e76-A2CB-71543EAE5789} 2009-11-11 09:00:00 2009-12-12 09:00:00 100000 120000 ITSOCustomer Id ITSOUserId ITSOCustomerName ITSORentalAgreement Id ContentElements DocumentTitle ITSOVehicle Id ITSOVehicleID ITSOVehicleYear ITSOVehicleMake ITSOVehicleModel ITSOVehicleColor Containees FolderName {C4D27FCE-13A7-405c-8BDD-7EF09FB67251} 1F786HH76DCC1 1975 Dacia 1310 Blue {E407D1FC-EF11-48bc-A977-55DF8A9A9B60} <not retrieved> {35F89870-64A4-4e75-A71B-596E7215473F} Agreement.txt <not retrieved> {4B608AFB-E6EA-4e76-A2CB-71543EAE5789} <not retrieved> John Q. Renter

ITSORentalAgreement {35F89870-64A4-4e75-A71B-596E7215473F} ITSOVehicle ITSODailyRate ITSOConfirmationID {C4D27FCE-13A7-405c-8BDD-7EF09FB67251} 25000 1009792649

Figure 3-2 Objects and properties returned

3.5 getInstance() versus fetchInstance()

The API offers two methods for instantiating CE objects: getInstance() and fetchInstance().

getInstance() and GetInstance()

A getInstance() (Java) or GetInstance() (.NET) call takes an object store, class, and retrieval identifier and creates locally a representation of the requested object, whether or not it actually exists. After the call, the object is immediately available locally to use, however no properties or content is available until one of the following methods is called: .refresh() or .Refresh() .fetchProperties() or .FetchProperties()

70

Developing Applications with IBM FileNet P8 APIs

A getInstance() call is useful for batching, or for calls where the properties are not necessary, just the action upon the object (such as deleting an object). Because there is no initial round trip to the server to fetch the object, this is also called fetchless instantiation.

fetchInstance() and FetchInstance()

A fetchInstance() (Java) or FetchInstance() (.NET) call results in an immediate round-trip to the CE, retrieving any requested properties or all object properties, depending on the value of the property filter. The result is a local object that contains properties that can be immediately queried and modified as necessary. The downside is that the call to the CE is immediate, guaranteeing the network round-trip.

Deciding which to use

The advantages and disadvantages to get and fetch methods are: You get a performance improvement by using fetchless instantiation, but the corresponding cost is that your application's error handling might have to deal with missing objects at a later stage. In practice, that error-handling cost is not too severe. Another cost is the loss of first writer wins protection in the CE. Fetchless instantiation explicitly ignores any other updates to the same object. With a fetched object, the update sequence number is checked for consistency by the CE before any changes are committed for that object. If you actually need to get the values of any properties of an object, at least one round trip to the CE server will occur, in any case. This approach erases the performance advantage of fetchless instantiation.

3.6 Querying
When developing for the CE, one of the primary ways of finding documents is to do a search, based on data from a user. Workplace and WorkplaceXT both offer search templates and stored searches that a user can execute. When you program, doing the search directly in code by using the SearchScope and SearchSQL classes is necessary. Searching the CE requires the following steps: 1. 2. 3. 4. Create the SQL statement. Set the SearchScope. Execute the search. Work with the results.

Chapter 3. Introduction to Content Engine API programming

71

3.6.1 SearchSQL
The CE offers a language for executing queries within single or multiple object stores. The SearchSQL class offers helper methods to assist you in constructing a SQL statement, or the SQL statement can be constructed independently and passed to a SearchSQL instance as a string. SQL statements must follow the CE SQL syntax, which generally conforms to SQL-92, with extensions for FileNet-specific constructs. The complete SQL syntax guide is available in the IBM FileNet P8 Platform documentation by selecting ECM Help Developer Help Content Engine Development Reference SQL Syntax Descriptions.

Building a query with SearchSQL

The CE API has a SearchSQL object with a number of convenience methods for creating the search SQL. In addition, the SearchSQL object can be instantiated with an actual SQL statement. Example 3-28 and Example 3-29 show an example of building a query with SearchSQL.
Example 3-28 Building a query with SearchSQL in Java

SearchSQL sqlObject = new SearchSQL("SELECT DocumentTitle, Id FROM Document WHERE Creator = 'jsmith'");
Example 3-29 Building a query with SearchSQL in C#

SearchSQL sqlObject = new SearchSQL("SELECT DocumentTitle, Id FROM Document WHERE Creator = 'jsmith'"); Note: The helper methods on SearchSQL and a SearchSQL object that are instantiated with a SQL statement are mutually-exclusive. The SQL passed in (on creation) cannot be updated with the convenience methods. Experienced developers typically find that supplying the SQL as a string is the simpler approach.

Using the IBM FileNet Enterprise Manager to build SQL

When you initially build the SQL, use the Enterprise Manager's Query Builder tool to construct your query or to quickly validate that it works as intended. After the query has been constructed, it can be copied to the code and passed to the SearchSQL object. To create SQL by using the Enterprise Manager: 1. Select the target Object Store. 2. Right-click the Search Results node.

72

Developing Applications with IBM FileNet P8 APIs

3. Select new search. 4. Build the search in the Content Engine Query Builder. 5. Select View SQL View. Figure 3-3 shows how to use the Enterprise Managers Content Engine Query Builder to create the search SQL for use in a SearchSQL object.

Figure 3-3 Compiled SQL statement from Content Engine Query Builder

Executing the query and working with the results

After a search has been executed, the SearchScope returns an IndependentObjectSet or RepositoryRowSet of the returned items, depending on whether you fetch objects or rows, respectively. See Example 3-30 and Example 3-31 on page 74.
Example 3-30 Search example in Java

// Construct the sql statement SearchSQL sql = new SearchSQL( "select ISTOStartDate, ITSOEndDate, ITSOVehicle " + "from ITSOIdleActivity " + "where + "ITSOVehicle = OBJECT('{D5DC8C04-2625-496f-A280-D791AFE87A73}') " + "AND ITSOStartDate < 20090801T000000Z OR " + "ITSOEndDate > 20090701T000000Z" ); // Search the object store with the sql and get the returned items SearchScope scope = new SearchScope(os);

Chapter 3. Introduction to Content Engine API programming

73

IndependentObjectSet set = scope.fetchObjects(sql, null, null, false); // Loop through the returned results for( Iterator i = set.iterator(); i.hasNext(); ) { // Get the document CustomObject obj = (CustomObject)i.next(); // Code to work with object goes here }
Example 3-31 Search example in C#

// Construct the sql statement SearchSQL sql = new SearchSQL( "select ITSOStartDate, ITSOEndDate, ITSOVehicle " + "from ITSOIdleActivity " + "where " + "ITSOVehicle = OBJECT('{D5DC8C04-2625-496f-A280-D791AFE87A73}') " + "AND ITSOStartDate < 20090801T000000Z OR " + "ITSOEndDate > 20090701T000000Z" ); // Search the object store with the sql and get the returned items SearchScope scope = new SearchScope(os); IIndependentObjectSet set = scope.FetchObjects(sql, null, null, false); IEnumerator e = set.GetEnumerator(); while(e.MoveNext()) { // Get the current object ICustomObject obj = (ICustomObject)e.Current; // Code to work with object goes here }

3.6.2 Search scope

When you execute a search, the scope of the search must be defined. This scope can be either a single object store or the search scope can be expanded to include multiple object stores. Methods on the search scope define the type of objects to be returned.

74

Developing Applications with IBM FileNet P8 APIs

The SearchScope methods execute the SQL statement on one or more object stores to find objects (IndependentObject instances), database rows (RepositoryRow instances), or metadata (ClassDescription instances). If the query includes a JOIN operator, you must fetch rows instead of objects.

Cross object store searching

When doing a cross-object store search (also called a merged scope search), the search results are merged based on the merge mode. The merge mode can be either: INTERSECTION: The search results will contain the classes occurring in all repositories searched. UNION: The search results will contain the classes occurring in any repository searched. When the merge mode is UNION and a class or property is not found in any repository, the following occurs: For classes, an INNER JOIN returns no rows, and an OUTER JOIN returns nulls. JOIN types are specified in the JoinOperator class For properties, the property value is null in a selection list or WHERE clause, and is omitted from an ORDER BY clause. To search across multiple object stores: 1. Create an array of object stores to search. 2. Create the search scope with the array of object stores to search and the merge mode. 3. Create the SQL. 4. Execute the search. 5. Work with the results. See Example 3-32 and Example 3-33 on page 76.
Example 3-32 Setting the scope to be multiple object stores in Java

// Create the object store array ObjectStore[] osArray = new ObjectStore[]{os1,os2}; // Create the search scope SearchScope objStores = new SearchScope(osArray, MergeMode.INTERSECTION);

Chapter 3. Introduction to Content Engine API programming

75

Example 3-33 Setting the scope to be multiple object stores in C#

// Create the object store array IObjectStore[] osArray = new IObjectStore[] { os1, os2 }; // Create the search scope SearchScope objStores = new SearchScope(osArray, MergeMode.INTERSECTION);

3.6.3 Content searches

Full-text searches (also known as content-based retrieval (CBR)) can be used to search for words or phrases that are part of object content or that occur in string properties of these objects. For the content in an object or its string properties to be searched, you must enable CBR for the object class and any of its string properties to be included in a content search. A content search is indicated by either a CONTAINS or FREETEXT operator in the SQL statement contained in SearchSQL. The CONTAINS and FREETEXT operators have somewhat different operand formats and provide somewhat different search characteristics. Although the CONTAINS operator can search content in all properties, or in a single property, the FREETEXT operator can search content only in all properties. Note that attempting to specify a property name for FREETEXT will generate an exception. The CONTAINS operator is generally preferable to the FREETEXT operator. For more information about the CONTAINS and FREETEXT operators, go to ECM Help Developer Help Content Engine Development Reference SQL Syntax Descriptions Full-Text Queries. Note: Full-text queries can have unexpected performance impact. You should experiment with CBR queries on realistic data sets to ensure that you have taken performance into account.

3.6.4 Paging support

When working with a large result set, returning too many items at once can cause memory usage problems. The API allows for breaking the result set into groups of items called pages. In this way, an application can request and work with a smaller subset of items at a time. You indicate that you want paged results by making your query continuable. Example 3-34 on page 77 and Example 3-35 on page 77 show how to page through a result in Java and .NET.

76

Developing Applications with IBM FileNet P8 APIs

Example 3-34 Paging in Java // A query that will return a large result set SearchSQL sql = new SearchSQL("select [DocumentTitle] from document"); SearchScope scope = new SearchScope(os); // Set the paging to be 50 items per page and enable continuation IndependentObjectSet s = scope.fetchObjects(sql, 50, null, true); // Get the page iterator PageIterator p = s.pageIterator(); // Loop through each page while(p.nextPage()) { // Loop through each item in the page for(Object obj : p.getCurrentPage()) { // Get the document object and write Document Title Document doc = (Document)obj; System.out.println( doc.getProperties().getStringValue("DocumentTitle")); } } Example 3-35 Paging in C# SearchSQL sql = new SearchSQL("select [DocumentTitle] from document"); SearchScope scope = new SearchScope(os); IIndependentObjectSet s = scope.FetchObjects(sql, 50, null, true); IPageEnumerator p = s.GetPageEnumerator(); while(p.NextPage()) { Console.WriteLine(p.ElementCount); foreach(object obj in p.CurrentPage ) { IDocument doc = (IDocument)obj; Console.WriteLine( doc.Properties.GetStringValue("DocumentTitle")); } }

Chapter 3. Introduction to Content Engine API programming

77

3.7 Viewing documents

The CE can store a wide variety of document file types, including images. For image file types that users need to view, applications can choose to retrieve the image content and display it in their preferred viewer application, or they can use WorkplaceXTs Java viewer. The supported image file types for WorkplaceXTs Java Viewer can be found in ECM Help User Help Actions, preferences, and tools Work with Image Viewer. For applications that choose to use the WorkplaceXT Java viewer applet, the supported way to instantiate the viewer is to obtain a user token and call the getContent servlet from WorkplaceXT. Using the Java viewer applet directly with a separate application is not supported. Example 3-36 and Example 3-37 on page 79 show the code samples for calling the WorkplaceXT Java viewer applet.
Example 3-36 Document viewing in Java and JSP <%@ page import="java.util.*, java.io.*, java.net.*"%> <html> <body> <% String String String String String baseP8URL = "http://aeServer:9080/WorkplaceXT/"; user = "user"; password = "password"; objectStore = "ObjectStoreName"; docID = "{C9712786-4B17-4512-9E13-9F4154B35FC2}";

// Call WorkplaceXTs setCredentials servlet to obtain user token URL url = new URL(baseP8URL + "setCredentials?op=getUserToken&userId=" + user + "&password=" + password + "&verify=true"); HttpURLConnection p8con = (HttpURLConnection) url.openConnection(); p8con.setRequestMethod("POST"); p8con.connect(); InputStream in = p8con.getInputStream(); int c = -1; String tempUserToken = ""; while ( (c=in.read()) >= 0) tempUserToken += new Character((char)c).toString();

78

Developing Applications with IBM FileNet P8 APIs

String userToken = URLEncoder.encode(tempUserToken, "UTF-8"); // Build URL to getContent servlet String contentUrl = baseP8URL + "getContent?objectStoreName=" + objectStore + "&id=" + docID + "&objectType=document&ut=" + userToken + "&impersonate=true"; %> <script language="javascript"> window.top.location="<%=contentUrl%>" </script>

</body> </html> Example 3-37 Document viewing in C# and ASP.NET using System.Net; using System.IO; public partial class _Default : System.Web.UI.Page { public string WorkplaceRootUrl = ""; public string ContentUrl = ""; public string Username = ""; private void Page_Load(object sender, System.EventArgs e) { // set constant values // replace these values with values applicable to your site WorkplaceRootUrl = "http://aeServer:9080/WorkplaceXT/"; Username = "user"; string pwd = "password"; string objectStoreName = "ObjectStoreName"; string Id = "{C9712786-4B17-4512-9E13-9F4154B35FC2}";

// Call WorkplaceXTs setCredentials servlet to obtain user token string UserToken = getCEUserToken(WorkplaceRootUrl, Username, pwd); // create URLs for the JavaViewer ContentUrl = WorkplaceRootUrl + "getContent?objectStoreName=" + objectStoreName + "&id=" + Id + "&objectType=document&ut=" + UserToken + "&impersonate=true"; }

Chapter 3. Introduction to Content Engine API programming

79

private string getCEUserToken(string baseURL, string uid, string pwd) { string UserToken = ""; // make the request and get the response WebRequest request = WebRequest.Create(baseURL + "setCredentials?op=getUserToken&userId=" + uid + "&password=" + pwd + "&verify=true"); request.Method = "POST"; WebResponse response = request.GetResponse(); // read the response from the stream into a byte[] Stream stream = response.GetResponseStream(); byte[] token = new byte[response.ContentLength]; stream.Read(token, 0, (int)response.ContentLength); response.Close(); // and convert the bytes in the array into a string. foreach (byte chr in token) UserToken += System.Convert.ToChar(chr); // return the encoded string return Server.UrlEncode(UserToken); } }

// Below is the aspx page source <html> <body> <script language="javascript"> window.top.location="<%=ContentUrl%>" </script> </body> </html>

3.7.1 User tokens

The code examples in Example 3-36 on page 78 and Example 3-37 on page 79 for viewing documents use user tokens obtained from WorkplaceXT. The method used in those examples is an sample of how a custom application that does not have access to IBM FileNet API libraries can obtain a user token through an HTTP URL call by calling the WorkplaceXT setCredentials servlet.

80

Developing Applications with IBM FileNet P8 APIs

User tokens can also be obtained in other ways and are described in detail in ECM Help Developer Help Workplace Development Workplace Customization Guide User Tokens.

3.8 Batching and batch execution

To make more efficient use of network resources and to allow for grouping the related operations into a single work unit, the CE API offers batching capabilities. A batch accumulates and packages multiple operations (method calls) on objects. The batch is then executed in a single operation. Whether a batch is a transactional operation depends on its type: Batch updates type This type of batch operation creates, updates, or deletes persisted objects, and is executed transactionally. IndependentlyPersistableObject references are accumulated, and an instance of the UpdatingBatch class is executed as a single transaction (the updateBatch method). The batch execution does not return a value: all of the pending commits succeed or all fail. For the failure case the transaction is rolled back and an exception is thrown. Batch retrieval type This type of batch operation retrieves independent objects, and is not executed transactionally. IndependentObject references are accumulated, and each included object then is either refreshed (retrieved) or gets its own exception. As for single-operation object saves and retrievals, any changes to the retrieved objects are done in place, so the existing IndependentObject references continue to be valid and reflect the changes. Note: A batch can significantly improve performance. Use batching when application logic lends itself to executing a series of operations that can be completed (or be in progress) independently, without reliance on either the state of another object included in the batch, or the result of another operation in the batch.

Refresh versus no refresh

When committing operations and batches to the CE, the operations have an option to do a refresh or no refresh on completion. Setting this to RefreshMode.REFRESH causes all local properties to be reloaded from the CE, providing the most recent values for a given object. Refresh.NO_REFRESH indicates to the API that no additional information, other than success and failure, should be returned from the call.

Chapter 3. Introduction to Content Engine API programming

81

Selecting the right mode can affect performance of operations or the work unit as a whole. For example, when committing a group of documents to the CE, you might only want to know whether the commit failed or is successful. So doing a refresh in this situation is not necessary, and would cause the entire operation to be slower. However, if the code has to work with the committed object, doing a refresh on commit operation can save a round-trip to the CE to fetch any updated properties (such as the ID of a newly created object).

Batch updates
You can perform bulk updating of objects in a batch by using UpdatingBatch. To update objects in batch: 1. Create the UpdatingBatch. 2. Retrieve the objects to update. 3. Apply the changes to the local objects (but do not .save() the changes). 4. Add the objects to the batch. 5. Update the batch. Example 3-38 and Example 3-39 on page 83, adapted from RentalActivityHandler.insertVehicleActivity, show how to update an existing object and create a second object in a single batch operation.
Example 3-38 Creating a batch of update actions in Java

// Get the IdleActivity record (from a previous search) CustomObject previous = (CustomObject)set.iterator().next(); // Create the next custom object CustomObject next = Factory.CustomObject.createInstance(os,"ITSOIdleActivity"); // Set the properties Properties props = next.getProperties(); props.putValue("ITSOVehicle", vehicle); props.putValue("ITSOStartDate", end); // Update the previous object previous.getProperties().putValue("ITSOEndDate", start); // Create the batch UpdatingBatch batch = UpdatingBatch.createUpdatingBatchInstance(domain,

82

Developing Applications with IBM FileNet P8 APIs

RefreshMode.REFRESH); // Add the objects batch.add(previous, null); batch.add(next, null); // Execute the batch batch.updateBatch();
Example 3-39 Creating a batch of update actions in C#

// Get the IdleActivity record (from a previous search) ICustomObject previous = (ICustomObject)getPrevious(); // Create the next custom object ICustomObject next = Factory.CustomObject.CreateInstance(_os, "ITSOIdleActivity"); // Create the batch UpdatingBatch batch = UpdatingBatch.CreateUpdatingBatchInstance(domain, RefreshMode.REFRESH); // Add the objects batch.Add(previous, null); batch.Add(next, null); // Execute the batch batch.UpdateBatch();

Batch retrieval
In addition to bulk creating and updating objects, batches can also be used to retrieve objects in bulk. Using RetrievingBatch can help limit the number of round-trips to the CE, increasing performance, and reducing network traffic. To retrieve items in a batch: 1. Create the RetrievingBatch. 2. Instantiate the objects to be retrieved. 3. Add them to the batch. 4. Retrieve the batch. 5. Work with the returned item(s).

Chapter 3. Introduction to Content Engine API programming

83

Example 3-40 and Example 3-41 show how to retrieve two documents in a batch using RetrievingBatch.
Example 3-40 Batch retrieval in Java

RetrievingBatch rb = RetrievingBatch.createRetrievingBatchInstance(domain); Folder vehicle = Factory.Folder.getInstance(os, "ITSOVehicle", vehiclePath); PropertyFilter filter = new PropertyFilter(); filter.addIncludeProperty(0, null, null, PropertyNames.ID + " " + PropertyNames.NAME + " " + "ITSODailyRate ITSOFranchiseCode" ); rb.add(vehicle, filter); Document doc = Factory.Document.getInstance(os, "ITSODocument", path); filter = new PropertyFilter(); filter.addIncludeProperty(0, null, null, PropertyNames.ID + " ITSOFranchiseCode"); rb.add(doc, filter); rb.retrieveBatch();
Example 3-41 Batch retrieval in C#

RetrievingBatch rb = RetrievingBatch.CreateRetrievingBatchInstance(domain); IFolder vehicle = Factory.Folder.GetInstance(os, "ITSOVehicle", vehiclePath); PropertyFilter filter = new PropertyFilter(); filter.AddIncludeProperty(0, null, null, PropertyNames.ID + " " + PropertyNames.NAME + " " + "ITSODailyRate ITSOFranchiseCode"); rb.Add(vehicle, filter);

84

Developing Applications with IBM FileNet P8 APIs

IDocument doc = Factory.Document.GetInstance(_os, "ITSODocument", path); filter = new PropertyFilter(); filter.AddIncludeProperty(0, null, null, PropertyNames.ID + " ITSOFranchiseCode"); rb.Add(doc, filter); rb.RetrieveBatch();

Chapter 3. Introduction to Content Engine API programming

85

86

Developing Applications with IBM FileNet P8 APIs

Chapter 4.

Advanced Content Engine API programming

This chapter continues the discussion of Content Engine (CE) development topics. Although we refer to these topics as advanced, they are actually a continuation of the concepts we presented in Chapter 3, Introduction to Content Engine API programming on page 39. Most of the code snippets we show in this chapter are from the sample applications that we created for this book. For more details about the sample applications, refer to Chapter 7, Sample applications for Fictional Auto Rental Company A on page 223. This chapter discusses the following topics: Permissions and authorization Versioning Relationships Annotations Subscriptions and event actions Workflow subscriptions and workflow event actions Metadata discovery Dynamic security inheritance

Copyright IBM Corp. 2009. All rights reserved.

87

4.1 Permissions and authorization

The CE contains features for fine-grained control of access to objects. This control includes both discretionary and mandatory access control. The most fundamental of these features is the access control list (ACL). It is represented in the APIs as the class AccessPermissionList. An ACL is a list of access control entries (ACEs), each of which describes access for a particular user or group. An ACE is represented in the APIs as the class AccessPermission, which is the value type for the Permissions property accessor methods. Users and groups must be defined in the configured directory used by the CE. The CE actually stores the unique identifier for the user or group and translates it dynamically to the GranteeName property. If that translation fails (because the user or group has been deleted from the directory), the unique identifier is presented as the GranteeName, and the GranteeType is set to SecurityPrincipalType.UNKNOWN. The access being granted or denied is given by the AccessMask property. It is an integer-valued property representing a simple bit mask of AccessRight values. Bit-wise arithmetic operations can be used to manipulate the values. The APIs also contain a series of constants of type AccessLevel. They just represent combinations of AccessRight values that are useful for some scenarios. They are not additional types of access controls. Example 4-1 and Example 4-2 on page 89 show an update to the ACL for an object to remove AccessRight.DELETE permission for each non-inherited ACE. The net result is that those grantees will not be allowed to delete the object. (This does not affect the owner, and it does not affect a grantee with delete rights through inheritance.)
Example 4-1 AccessPermissionList manipulation in Java

/** * Removes the AccessRight.DELETE right from all direct ACEs for an * object. Caller must have fetched the Permissions property. * Return true if we actually changed anything. Caller must call save. */ public boolean removeDeletePerm(IndependentlyPersistableObject ipo) { boolean madeChanges = false; Properties props = ipo.getProperties(); if (!props.isPropertyPresent(PropertyNames.PERMISSIONS)) { return madeChanges; } AccessPermissionList acl =

88

Developing Applications with IBM FileNet P8 APIs

(AccessPermissionList)props.getDependentObjectListValue( PropertyNames.PERMISSIONS); for (Iterator it = acl.iterator(); it.hasNext();) { AccessPermission ace = (AccessPermission)it.next(); PermissionSource acePS = ace.get_PermissionSource(); // Skip any inherited ACEs if (acePS == PermissionSource.SOURCE_PARENT) continue; int rights = ace.get_AccessMask().intValue(); if ((rights & AccessRight.DELETE_AS_INT) != 0) { madeChanges = true; // Remove the DELETE bit. rights &= ~AccessRight.DELETE_AS_INT; ace.set_AccessMask(rights); } } return madeChanges; }
Example 4-2 AccessPermissionList manipulation in C#

/** * Removes the AccessRight.DELETE right from all direct ACEs for an * object. Caller must have fetched the Permissions property. * Return true if we actually changed anything. Caller must call save. */ public Boolean RemoveDeletePerm(IIndependentlyPersistableObject ipo) { Boolean madeChanges = false; IProperties props = ipo.Properties; if (!props.IsPropertyPresent(PropertyNames.PERMISSIONS)) { return madeChanges; } IAccessPermissionList acl = (IAccessPermissionList)props[PropertyNames.PERMISSIONS]; foreach (IAccessPermission ace in acl) { PermissionSource acePS = ace.PermissionSource; // Skip any inherited ACEs if (acePS == PermissionSource.SOURCE_PARENT) continue; int rights = (int)ace.AccessMask; if ((rights & (int)AccessRight.DELETE) != 0) {

Chapter 4. Advanced Content Engine API programming

89

madeChanges = true; // Remove the DELETE bit. rights &= ~((int)AccessRight.DELETE); ace.AccessMask = rights; } } return madeChanges; }

4.2 Versioning
One of the fundamentals of traditional document management, the forerunner of modern enterprise content management, is that documents change over time. Rather than simply replacing the content as a whole, IBM FileNet Content Manager allows you to make a completely new version of the document, thereby preserving the older versions of the content. A Document object in CE represents a single version of what you might think of as a document (for example, a spreadsheet, a text file, or a photograph). The collection of all versions of a given document is represented in CE by a VersionSeries object. Each version in a VersionSeries object has its own security, life cycle, and other characteristics as an independently persistable object. The latest version of a document in a VersionSeries is called the current version and is pointed to by the CurrentVersion property. There is a two-level version numbering scheme (major and minor versions). The latest major version in a VersionSeries is called the released version and is pointed to by the ReleasedVersion property. A checked-in document can also be promoted to a major version or demoted to a minor version. Figure 4-1 on page 91 illustrates a VersionSeries object with multiple independent objects. Version 1.0 and 1.1 are superseded. Version 2.0 is the current official release. Version 2.1 is a checked-out version of 2.0 with a reservation state.

90

Developing Applications with IBM FileNet P8 APIs

Independent Object Version 1.0 Superseded

Independent Object Version 1.1 Superseded

Independent Object Version 2.0 Released

Independent Object Version 2.1 Reservation

Version Series

Figure 4-1 VersionSeries object with multiple IndependentObjects

To create a new version of a document, perform a check-out and check-in cycle: 1. Check out the current version of the document. This creates a new reservation document version. A reservation is not a different type of object. It is merely a new document version marked as being in the reservation state. The immediately previous version is in the reserved state. Properties may be automatically copied from the current version to the reserved version, depending on the value of the PropertyDescription boolean CopyToReservation metaproperty. 2. Make content changes and possibly property changes. 3. Check in the reserved document version. If you change your mind and want to cancel the check-out step, you can call the cancelCheckout helper method or simply delete the reserved document version (these are exactly the same operation). Content can only be updated on a document object that is in the reserved state. You can update content as many times as you want, with new content completely replacing the old content for that reserved version. After you perform the check-in step, the content for that version becomes immutable. (Properties for document versions can be changed at any time. Locking out property changes for a particular document version is possible by calling the freeze method.)

Chapter 4. Advanced Content Engine API programming

91

Within a given version series, individual independent versions can be deleted and removed from the version series by instantiating a specific document object and calling its delete method. Deleting the entire version series requires instantiating the VersionSeries object and calling its delete method. The instantiation of the VersionSeries can be done by using a Factory.VersionSeries method or by getting the value of the VersionSeries property from any of the documents in the version series. Example 4-3 and Example 4-4 on page 93 show how to retrieve a document, check it out, add content, and check the document back in as major version.
Example 4-3 Check-out and check-in cycle in Java // Get the document (saving a round-trip that a fetch would require) Document doc = Factory.Document.getInstance(os,"Document","/Doc1"); // Checkout the document and save doc.checkout(ReservationType.EXCLUSIVE, null, null, null); doc.save(RefreshMode.REFRESH); // Get the reservation object Document res = (Document)doc.get_Reservation(); // Update the properties res.getProperties().putValue("DocumentTitle","NextVersion"); // Prepare the content for attaching // Create the element list ContentElementList list = Factory.ContentElement.createList(); // Create a content transfer element by attaching a simple text file ContentTransfer element = Factory.ContentTransfer.createInstance(); // Set the MIME type element.set_ContentType("text/plain"); // Set the retrieval name element.set_RetrievalName("file.txt"); // Set the content source element.setCaptureSource(new FileInputStream(new File("file.txt"))); // Add the item to the list list.add(element); // Add the content transfer list to the document res.set_ContentElements(list); // Set the PendingAction to be check-in as a major version without

92

Developing Applications with IBM FileNet P8 APIs

// automatic content classification res.checkin(AutoClassify.DO_NOT_AUTO_CLASSIFY, CheckinType.MAJOR_VERSION); // Save the document to the repository res.save(RefreshMode.NO_REFRESH); Example 4-4 Check-out and check-in cycle in C# // Get the document (saving a round-trip that a fetch would require) IDocument doc = Factory.Document.GetInstance(os,"Document","/Doc1"); // Checkout the document and save doc.Checkout(ReservationType.EXCLUSIVE, null, null, null); doc.Save(RefreshMode.REFRESH); // Get the reservation object IDocument res = (IDocument)doc.Reservation; // Update the properties IProperties properties = res.Properties; // Set the value properties["DocumentTitle"] = "NewVersion"; // Create the content element list IContentTransferList l = Factory.ContentTransfer.CreateList(); // Create the content element IContentTransfer element = Factory.ContentTransfer.CreateInstance(); element.RetrievalName = "file.txt"; element.ContentType = "text/plain"; element.SetCaptureSource( System.IO.File.Open("file.txt", FileMode.Open)); l.Add(element); // Add the elements to the object res.ContentElements = l; // Set the check-in type res.Checkin(AutoClassify.DO_NOT_AUTO_CLASSIFY, CheckinType.MAJOR_VERSION); // Commit the changes to the CE res.Save(RefreshMode.NO_REFRESH);

Chapter 4. Advanced Content Engine API programming

93

4.3 Relationships
The CE APIs and the CE server itself have many features which emulate features of modern object-oriented programming languages. One of those features is the ability to create explicit relationships between objects. In most cases, the server perform referential integrity checks based on those relationships.

4.3.1 Object-valued properties

The fundamental mechanism for relationships between objects is the object-valued property (OVP). In a programming language, the analogy would be to a reference or pointer to another object. Just as in a programming language, an OVP has a required type. The server enforces the restriction that only objects of the correct type (including possibly a subclass of the required type) are assigned to an OVP. Although defining an OVP (or any other property) programmatically is unusual, actually populating and using OVPs is very common. OVPs can be either single-valued or multi-valued (MVOVP). Except for a few system OVPs, specifically lists of dependent objects, MVOVPs are always read-only. This concept can be confusing, but it becomes very clear when you understand just a little about the implementation. MVOVPs are not persisted in object instances in the repository. Instead, the metadata notes that some single-valued OVPs point to the object holding the MVOVP. When the value of the MVOVP is needed, the server dynamically generates a query to find all objects that are pointing to this object and returns the results of that query. For example, the system class Annotation has an OVP named AnnotatedObject. Classes that can have annotations (folders, documents, and custom objects) have an MVOVP named Annotations. Because the value of Annotations reflects the values of another property on a different set of objects, it is called a reflective property. Combinations of the two such properties (for example, AnnotatedObject and Annotations) are sometimes called association properties. When you create a new annotation and populate its AnnotatedObject property, the corresponding Annotations property appears to be instantaneously updated. Example 4-5 on page 95 and Example 4-6 on page 95 show the assignment of OVPs. It is mechanically similar to any other property value assignment.

94

Developing Applications with IBM FileNet P8 APIs

Example 4-5 OVP manipulation in Java, adapted from ITSONewVehicleActivity.java

/* Folder vehicle = ... */ // Create the ITSOIdleActivity object CustomObject activity = Factory.CustomObject.createInstance(os, "ITSOIdleActivity"); // Get the properties collection Properties props = activity.getProperties(); // Set the properties props.putValue("ITSOFranchiseCode", vehicle.getProperties().getStringValue("ITSOFranchiseCode")); props.putObjectValue("ITSOVehicle", vehicle); // Save the new IdleActivity activity.save(RefreshMode.NO_REFRESH);
Example 4-6 OVP manipulation in C#

/* IFolder vehicle = ... */ // Create the ITSOIdleActivity object ICustomObject activity = Factory.CustomObject.CreateInstance(os, "ITSOIdleActivity"); // Get the properties collection IProperties props = activity.Properties; // Set the properties props["ITSOFranchiseCode"] = vehicle.getProperties().getStringValue("ITSOFranchiseCode"); props["ITSOVehicle"] = vehicle; // Save the new IdleActivity activity.Save(RefreshMode.NO_REFRESH);

4.3.2 Filing in a folder

Perhaps the most well-known feature involving OVPs is the containment relationship whereby objects are filed into a folder. Besides the folder and the object being filed, there is an intermediate relationship object.

Chapter 4. Advanced Content Engine API programming

95

It is either of the following objects: ReferentialContainmentRelationship (RCR) object DynamicReferentialContainmentRelationship (DRCR) object Both RCR and DRCR are extensible. The relationship object, being an intermediary, enables a folder to contain many objects and an object to be contained in many folders. In fact, a given object may be contained more than once in the same folder. The RCR (and DRCR) Head property is reflected by containees Containers property. The RCR Tail property is reflected by the folder Containees property. To understand the nomenclature of head and tail, imagine arrows pointing from the folder to the various containees. The RCR properties give you the tails and heads of those arrows. Imagine the RCR object as a single marker of sorts in the middle of each arrow. The APIs provide convenience methods for filing and unfiling objects. They are convenience methods because they do not represent fundamental operations and are instead built upon creation or deletion of RCR or DRCR objects. Newcomers to the APIs generally find the convenience methods easier to understand, but developers experienced in the APIs generally find direct creation and manipulation of the RCR and DRCR objects much clearer. The difference in terms of lines of code is very small. Example 4-7 and Example 4-8 on page 97 show a sample of filing, adapted from RentalActivityHandler.java in the sample application. Because it happens to be filing a CustomObject, the sample uses RCR. For filing a document object, you want to use DRCR unless you have a very specific use case for using RCR.
Example 4-7 Filing a document to a folder in Java

Document d = Factory.Document.getInstance(os, "Document", "/Document"); Folder f = Factory.Folder.getInstance(os, "Folder", "/Folder"); ReferentialContainmentRelationship rcr = Factory.ReferentialContainmentRelationship.createInstance(os, ClassNames.DYNAMIC_REFERENTIAL_CONTAINMENT_RELATIONSHIP); rcr.set_Head(d); rcr.set_Tail(f); rcr.save(RefreshMode.NO_REFRESH);

96

Developing Applications with IBM FileNet P8 APIs

Example 4-8 Filing a document in a folder in C#

IDocument d = Factory.Document.GetInstance(os, "Document", "/Document"); IFolder f = Factory.Folder.GetInstance(os, "Folder", "/Folder"); IReferentialContainmentRelationship rcr = Factory.ReferentialContainmentRelationship.CreateInstance(os, ClassNames.DYNAMIC_REFERENTIAL_CONTAINMENT_RELATIONSHIP); rcr.Head = d; rcr.Tail = f; rcr.Save(RefreshMode.NO_REFRESH);

4.3.3 Compound documents

Certain documents can be thought of as being comprised of component parts. For example, a Web page consists of a text part containing the HTML, perhaps additional text parts for the cascading style sheets and JavaScript libraries. There are likely additional parts representing graphical images. Taken together, these can be thought of as a type of compound document. The component document parts typically form some sort of hierarchy for a given complete document. Component parts may be reused in several related or unrelated documents, depending on the application use case. All of that taken together prompts the need for creating many-to-many relationships among an arbitrary collection of document components. There are many ways to model that in CE, but the feature provided specifically for it is the ComponentRelatonship class. In many ways, the ComponentRelationship class resembles the other relationship classes described in 4.3.2, Filing in a folder on page 95. In compound document terminology, the relationships are between component parents and children. The ComponentRelationship class has several properties describing or allowing you to manipulate various aspects of the parent and child relationships. For example, unlike folder containment, which is unordered, compound document relationships can be explicitly ordered. Example 4-9 on page 98 and Example 4-10 on page 98 show simple manipulation of compound documents.

Chapter 4. Advanced Content Engine API programming

97

Example 4-9 Compound documents in Java

/** * Count the total number of children and parent documents */ public int childrenAndParents(Document doc) { System.out.println("document: " + doc); int childCount = 0; // Treat this doc as a parent and count its children if (doc.get_CompoundDocumentState() == CompoundDocumentState.COMPOUND_DOCUMENT) { // Could get_ChildRelationships instead. DocumentSet children = doc.get_ChildDocuments(); for (Iterator it = children.iterator(); it.hasNext();) { Document child = (Document)it.next(); System.out.println("child: " + child); ++childCount; } } int parentCount = 0; // Any document can be a child, so always count parents. // Could get_ParentRelationships instead. DocumentSet parents = doc.get_ParentDocuments(); for (Iterator it = parents.iterator(); it.hasNext();) { Document parent = (Document)it.next(); System.out.println("parent: " + parent); ++parentCount; } return childCount + parentCount; }
Example 4-10 Compound documents in C#

/// <summary> /// Count the total number of children and parent documents /// </summary> public int ChildrenAndParents(Document doc) { Console.WriteLine("document: " + doc); int childCount = 0; // Treat this doc as a parent and count its children

98

Developing Applications with IBM FileNet P8 APIs

if (doc.CompoundDocumentState == CompoundDocumentState.COMPOUND_DOCUMENT) { // Could use ChildRelationships instead. IDocumentSet children = doc.ChildDocuments; foreach (IDocument child in children) { Console.WriteLine("child: " + child); ++childCount; } } int parentCount = 0; // Any document can be a child, so always count parents. // Could use ParentRelationships instead. IDocumentSet parents = doc.ParentDocuments; foreach (IDocument parent in parents) { Console.WriteLine("parent: " + parent); ++parentCount; } return childCount + parentCount; }

4.4 Annotations
An annotation, as its name implies, is some kind of additional information associated with an object. For example, you can think of using a pen to physically mark up a photograph or to add review comments to a document. Although one typically thinks of an annotation as being graphically overlaid on the annotated object, there is no requirement for that to be so. The non-versionable Annotation class may contain content elements. It can also be made into a subclass, so any desired custom properties may be added to it. The specific subclass type can be used to distinguish among different annotation types on an object. The CE annotation feature provides relationships and convenient default security and other behaviors for annotation objects. Documents, folders, and custom objects may have associated annotations. For Documents, the Annotation property AnnotatedContentElement provides the opportunity to affiliate an annotation with a specific content element. When you do that, using the invariant content element ElementSequenceNumber to make that association is better than using the volatile position number of the content element within a Documents content elements collection.

Chapter 4. Advanced Content Engine API programming

99

An annotation can be for any purpose that an application developer feels is appropriate. The Application Engines Java viewer applet has its own specific implementation of annotations, but the formats of those annotations are subject to change at any time without notice. Annotations may also be carried over from Image Services with Content Federation Services for Image Services (CFS-IS). Example 4-11 and Example 4-12 on page 101 illustrate the creation of a custom annotation and associating it with a specific Document content element. The custom properties are adapted from the sample application.
Example 4-11 Custom annotation in Java

public Annotation createThumbnail(Document gallery, int eltNum) { ContentElementList galleryElements = gallery.get_ContentElements(); ContentElement contentElement = (ContentElement)galleryElements.get(eltNum); if (!(contentElement instanceof ContentTransfer)) { throw new RuntimeException("Cannot make thumbnail for content reference"); } ContentTransfer gCt = (ContentTransfer)contentElement; String type = gCt.get_ContentType(); String rName = gCt.get_RetrievalName(); Integer esn = gCt.get_ElementSequenceNumber(); Properties gProps = gallery.getProperties(); String franchiseCode = gProps.getStringValue("ITSOFranchiseCode"); // Instantiate the new thumbnail Annotation ObjectStore os = gallery.getObjectStore(); Annotation thumbnail = Factory.Annotation.createInstance(os, "ITSOThumbnail"); Properties tProps = thumbnail.getProperties(); tProps.putValue("ITSOFranchiseCode", franchiseCode); thumbnail.set_AnnotatedObject(gallery); // Better to use immutable ESN than element position thumbnail.set_DescriptiveText("thumbnail image [" + esn + "]"); thumbnail.set_AnnotatedContentElement(esn); ContentTransfer tCt = Factory.ContentTransfer.createInstance(); tCt.set_ContentType(type);

100

Developing Applications with IBM FileNet P8 APIs

if (rName != null && rName.length() > 0) { tCt.set_RetrievalName("t_" + rName); } // The computation of the thumbnail is done elsewhere InputStream photoStream = gCt.accessContentStream(); tCt.setCaptureSource(computeThumbnailGraphics(type, photoStream)); ContentElementList thumbElements = Factory.ContentElement.createList(); thumbElements.add(tCt); thumbnail.set_ContentElements(thumbElements); // Return the new Annotation. return thumbnail; }
Example 4-12 Custom annotation in C#

Caller must save.

public IAnnotation CreateThumbnail(IDocument gallery, int eltNum) { IContentElementList galleryElements = gallery.ContentElements; IContentElement contentElement = (IContentElement)galleryElements[eltNum]; if (!(contentElement is IContentTransfer)) { throw new Exception("Cannot make thumbnail for content reference"); } IContentTransfer gCt = (IContentTransfer)contentElement; string type = gCt.ContentType; string rName = gCt.RetrievalName; int? esn = gCt.ElementSequenceNumber; IProperties gProps = gallery.Properties; string franchiseCode = gProps.GetStringValue("ITSOFranchiseCode"); // Instantiate the new thumbnail Annotation IObjectStore os = gallery.GetObjectStore(); IAnnotation thumbnail = Factory.Annotation.CreateInstance(os, "ITSOThumbnail");

Chapter 4. Advanced Content Engine API programming

101

IProperties tProps = thumbnail.Properties; tProps["ITSOFranchiseCode"] = franchiseCode; thumbnail.AnnotatedObject = gallery; // Better to use immutable ESN than element position thumbnail.DescriptiveText = "thumbnail image [" + esn + "]"; thumbnail.AnnotatedContentElement = esn; IContentTransfer tCt = Factory.ContentTransfer.CreateInstance(); tCt.ContentType = type; if (rName != null && rName.Length > 0) { tCt.RetrievalName = "t_" + rName; } // The computation of the thumbnail is done elsewhere Stream photoStream = gCt.AccessContentStream(); tCt.SetCaptureSource(ComputeThumbnailGraphics(type, photoStream)); IContentElementList thumbElements = Factory.ContentElement.CreateList(); thumbElements.Add(tCt); thumbnail.ContentElements = thumbElements; // Return the new Annotation. return thumbnail; } Caller must save.

4.5 Subscriptions and event actions

A key enabler of active content is the CE event subscription subsystem. Event subscriptions may be created for most types of update request arriving at the CE. These subscriptions are often set up manually through Enterprise Manager, but they can also be established programmatically. The subscriptions establish the triggers leading to the execution of user-provided code running inside the CE server. Subscriptions may be established on specific instances of objects, through the class InstanceSubscription, or for an entire class (and, optionally, any subclasses) of objects with the class ClassSubscription.

102

Developing Applications with IBM FileNet P8 APIs

Note: The event subscription subsystem is strongly related to the audit subsystem. That approach is understandable when you consider that the same activities are interesting both for auditing and for event notifications. The audit subsystem can also be enabled for content retrieval but the event subsystem cannot. We do not cover the audit subsystem in this book. Refer to the product documentation for further information. Event subscriptions are divided into synchronous and asynchronous types. The subscription object sets the type to either synchronous or asynchronous. Informally, the associated EventAction is commonly referred to as a synchronous or asynchronous event handler. Note the following information about synchronous and asynchronous: A synchronous event handler is executed as part of the path to persisting a change. It is within the same transaction scope. In other words, the change has not actually happened yet. If the event handler throws an exception, the entire transaction is aborted. In this way, the synchronous event action can veto a change. An important restriction on synchronous event handlers is that they are not allowed to update the source object, either directly or by acting on a copy. An asynchronous event handler is executed after a change has been committed. The delay between the triggering change and the running of the asynchronous event handler is typically small, but guaranteed maximum delay. An asynchronous event handler has its own transaction scope. Although it cannot veto the change that has already happened, the asynchronous event handler can update the source object. It can also do just about anything else that the developer wants it to do if it fits within the transaction scope and timeout and can be done in the context of a J2EE application server container. Both types of event handlers allow the insertion of user code in a common location in the CE server. The synchronous case can be used to approve or veto updates to keep them consistent with an application policy. The asynchronous case can be used to make changes to keep state consistent, again according to an application policy. Either type of event handler can be used to perform notifications or to initiate actions in external systems. Unlike other forms of application code, an event handler has the nature of a callback and is called by the CE server code. It receives a copy of the source object along with sufficient information to understand the nature of the triggering update activity. Because event handlers execute in the context of the server, normal access control checks do not apply. Another consequence of executing in

Chapter 4. Advanced Content Engine API programming

103

server context, however, is that event handlers must be written in Java. Interactions with the CE server are through the Java API. Example 4-13 shows an example of an event handler. It is adapted from ITSONewVehicleActivity.java in the sample application.
Example 4-13 Event handler in Java

public class ITSONewVehicleActivity implements EventActionHandler { public void onEvent(ObjectChangeEvent event, Id subId) { try { // Get the object store and the new vehicle folder ObjectStore os = event.getObjectStore(); Folder vehicle = (Folder) event.get_SourceObject(); // Create the ITSOIdleActivity object CustomObject activity = Factory.CustomObject.createInstance(os, "ITSOIdleActivity"); // Get the properties collection Properties props = activity.getProperties(); // Get the start date Date start = new Date(System.currentTimeMillis()); // Calculate an end date of 20 years from now GregorianCalendar end = new GregorianCalendar(); end.setTime(start); end.add(Calendar.YEAR, 20); // Set the properties props.putValue("ITSOFranchiseCode", vehicle.getProperties().getStringValue("ITSOFranchiseCode")); props.putObjectValue("ITSOVehicle", vehicle); props.putValue("ITSOStartDate", start); props.putValue("ITSOEndDate", end.getTime()); // Save the new IdleActivity activity.save(RefreshMode.NO_REFRESH); } catch (RuntimeException r) {

104

Developing Applications with IBM FileNet P8 APIs

throw r; } catch (Throwable t) { // Catch any other exception and wrap it throw new RuntimeException("Unexpected exception", t); } } }

4.6 Workflow subscriptions and workflow event actions

A special case of the event subscription subsystem is represented by classes that are related to the launching of workflows. InstanceWorkflowSubscription and ClassWorkflowSubscription are subclasses of the more general subscription classes. They have properties that associate the necessary artifacts (the workflow definition, property mapping, and so on) needed for a successful workflow launch. An interface to the PE is integrated with the CE, and setting up the automated launching of workflows that are triggered by events on objects is very simple to do. A workflow subscription takes a WorkflowEventAction to handle events instead of the more general EventAction class. A requirement is to use a WorkflowEventAction with a workflow subscription. Because the programming model for WorkflowEventAction is exactly the same as for EventAction, refer to Example 4-13 on page 104 for a code sample.

4.7 Metadata discovery

Nearly everything about CE content classes and properties is discoverable at run time. You can discover data types, localizable display names, default values, and many more attributes. In some sense, this feature is analogous to the reflection features that are available in Java and C#. Every class in an ObjectStore is defined in a ClassDefinition. Every property of a class is defined by a PropertyDefinition. A ClassDefinition has a list of PropertyDefinitions. Although the definition objects are fully read-write capable, updating these objects programmatically is unusual. Alternatively, consulting the metadata attributes to control application logic is quite common. To facilitate that,

Chapter 4. Advanced Content Engine API programming

105

there is a read-only and slightly more compact representation of the metadata embodied in ClassDescription and PropertyDescription objects. Note: Newcomers to CE often find the similar terminology of definition and description confusing. Think of it this way: The definition objects are the place where you can actively define various attributes for a class or property. The description objects merely describe what has been defined (and so are read-only). The high-level steps for finding a piece of metadata for a particular property is: 1. Find the applicable ClassDescription object. 2. Iterate over the PropertyDescriptions collection looking for the PropertyDescription object of interest. 3. Read attributes from that PropertyDescription object. Consulting the metadata is common, and changes to it are so rare that the APIs implement a client-side metadata cache (CMC). The CMC transparently intercepts some metadata fetches and satisfies them with locally cached copies. Because the class and property description objects are read-only, there is no harm in sharing them. CMC automatically keeps track of the differences in requester locale and organizes the data by ObjectStore. Although most of the CMC behavior is automatic and transparent, there is also an explicit way of fetching things from CMC. Many developers find this clearer than the implicit behavior. To use the explicit behavior, use a Factory method to instantiate a MetadataCache object. From there, the use of MetadataCache methods is strightforward. Example 4-14 and Example 4-15 on page 107 show metadata discovery. If you want to find all of the custom classes and properties in the sample application, you can call this method with a value of itso for both prefix parameters.
Example 4-14 Metadata discovery in Java public void cmcDemo(ObjectStore os, String classPrefix, String propPrefix) { classPrefix = classPrefix.toLowerCase(); propPrefix = propPrefix.toLowerCase(); MetadataCache cmc = Factory.MetadataCache.getDefaultInstance(); // Use a query to get a list of class names from the ObjectStore. // Can't query on ClassDescription, so query on ClassDefinition // (the class names are identical by design). SearchScope ss = new SearchScope(os); String sqlString = "SELECT " + PropertyNames.SYMBOLIC_NAME

106

Developing Applications with IBM FileNet P8 APIs

+ " FROM " + ClassNames.CLASS_DEFINITION; SearchSQL sql = new SearchSQL(sqlString); RepositoryRowSet rrs = ss.fetchRows(sql, null, null, true); // Iterate over the results. For any class with a name that // starts with the classPrefix, pull a ClassDescription out of // CMC and look for PropertyDescriptions whose names start with // the propPrefix. for (Iterator it = rrs.iterator(); it.hasNext();) { RepositoryRow row = (RepositoryRow)it.next(); Properties props = row.getProperties(); String className = props.getStringValue(PropertyNames.SYMBOLIC_NAME); if (className.toLowerCase().startsWith(classPrefix)) { ClassDescription cd = cmc.getClassDescription(os, className); PropertyDescriptionList pds = cd.get_PropertyDescriptions(); for (Iterator itp = pds.iterator(); itp.hasNext();) { PropertyDescription pd = (PropertyDescription)itp.next(); String propName = pd.get_SymbolicName(); if (propName.toLowerCase().startsWith(propPrefix)) { TypeID type = pd.get_DataType(); // Print the results. System.out.println(className + "." + propName + " (" + type + ")"); } } } } } Example 4-15 Metadata discovery in C# public void cmcDemo(IObjectStore os, string classPrefix, string propPrefix) { classPrefix = classPrefix.ToLower(); propPrefix = propPrefix.ToLower(); IMetadataCache cmc = Factory.MetadataCache.GetDefaultInstance(); // Use a query to get a list of class names from the ObjectStore. // Can't query on ClassDescription, so query on ClassDefinition // (the class names are identical by design). SearchScope ss = new SearchScope(os); String sqlString = "SELECT " + PropertyNames.SYMBOLIC_NAME

Chapter 4. Advanced Content Engine API programming

107

+ " FROM " + ClassNames.CLASS_DEFINITION; SearchSQL sql = new SearchSQL(sqlString); IRepositoryRowSet rrs = ss.FetchRows(sql, null, null, true); // Iterate over the results. For any class with a name that // starts with the classPrefix, pull a ClassDescription out of // CMC and look for PropertyDescriptions whose names start with // the propPrefix. foreach (IRepositoryRow row in rrs) { IProperties props = row.Properties; string className = props.GetStringValue(PropertyNames.SYMBOLIC_NAME); if (className.ToLower().StartsWith(classPrefix)) { IClassDescription cd = cmc.GetClassDescription(os, className); IPropertyDescriptionList pds = cd.PropertyDescriptions; foreach (IPropertyDescription pd in pds) { string propName = pd.SymbolicName; if (propName.ToLower().StartsWith(propPrefix)) { TypeID type = pd.DataType; // Print the results. Console.WriteLine(className + "." + propName + " (" + type + ")"); } } } } }

4.8 Dynamic security inheritance

Dynamic security inheritance, introduced in IBM FileNet Content Manager 4.0.1, is not specifically an API feature. It is an important server feature that can greatly simplify security-related programming. The idea of security inheritance is that some or all of the access control entries (ACEs) from one object are automatically inherited by another object or objects. Dynamic security inheritance means that the inheritance of the ACEs is done as needed, when access checks for the inheriting object are being calculated. This is not in any way a copying of ACEs from one object to another. It is true inheritance. The CE server goes to a great deal of trouble to make the process extremely efficient.

108

Developing Applications with IBM FileNet P8 APIs

Some traditional inheritance features in the server are implemented in terms of dynamic security inheritance. Perhaps the most well-known of these is the inheritance of folder security by the folders containees. It is also possible for developers to define custom security inheritance relationships. The implementation mechanism is to define a custom object-valued property (OVP) and designate that property as pointing to a security proxy object. Specifically, you set the value of the corresponding PropertyDefinitionObject SecurityProxyType property to SecurityProxyType.FULL. On the security proxy object itself, you mark the InheritableDepth property of each ACE according to how you would like the inheritance to proceed (none, immediate children, or infinite). Dynamic security inheritance can act transitively. That is, object A can inherit security from object B, which in turn can inherit security from object C. There is no built-in limitation for how long such chains can be. Directly applied ACEs have precedence over inherited ACEs in the access check calculations. In practical terms, this means that a user or group that is denied access through inheritance can still be granted access by a directly applied ACE. An interesting use of dynamic security inheritance is to augment the enterprise directory group structure. This can be especially handy in situations where the natural organizational groups typically found in a directory do not represent security access groups very well. It can also be helpful in situations where security access groups or memberships in those groups changes frequently, and the site does not want to make such frequent changes to the directory. In some ways, security proxy objects can be used to implement an authorization scheme that resembles a limited form of role-based access control (RBAC) that is often sufficient to meet site needs. Rather than providing a code sample for this feature (which is typically an administrative set-up and the application of appropriate default values for the security proxy property on new object instances), we instead provide a description of how it can work with the sample application described elsewhere in this book. Understanding the sample application in detail is not necessary in order to follow this description. A franchise location consists of maintenance employees, front-office employees, and supervisors responsible for managing both maintenance and front-office operations. Maintenance records (ITSOMaintenanceActivity objects in the sample application data model) may be accessed by maintenance employees and supervisory employees, but only within that franchise location. Those objects (and all objects in the sample application data model) have an OVP called ITSOSecurityProxy. There is also a class of objects called ITSORole that is intended to act as security proxy objects, among other things; see the sample application description for more details.

Chapter 4. Advanced Content Engine API programming

109

Although directly adding all maintenance employees and supervisors of a franchise location to every object where they need access is possible, it then becomes necessary to update the ACL on many objects any time a maintenance employee or supervisor is hired, resigns, changes status, and so on. This is both an operational and performance burden, and the chances for data errors are high. We can solve it more efficiently and elegantly with a chain of proxy objects: Create an ITSORole object to represent the concept of employees who should have access to maintenance records at this franchise location. It should have no directly applied ACEs. We refer to this object as MS. Create an ITSORole object to represent supervisors at this franchise location. It has direct ACEs that are granting access to all the supervisors at that location. We refer to this object as S. Create an ITSORole object to represent maintenance employees at this franchise location. It has direct ACEs granting access to all of the maintenance employees at that location. We refer to this object as M. Maintenance records at this franchise location are created with inheritance from object MS. This is done by populating the ITSOSecurityProxy property with a pointer to MS. Object MS inherits security from both object M and object S. This is done by populating the ITSOSecurityProxy property with a pointer to object M and the ITSOSecurityProxy2 with a pointer to object S. With this arrangement, access for supervisors (object S) and maintenance employees (object M) flows to object MS. From there, the access for the combined groups flows to the maintenance record object. There are many ways to organize this sort of custom inheritance. If access control from groups is being combined, as in this example, you need a class with more than one security proxy property. That is why we used the intermediate object MS in this example; we only needed the second security proxy property on the ITSORole object instead of all the other objects in the data model. Although modeling any kind of aggregation with intermediate objects with just two security proxy properties should be possible, defining more might be more convenient for you. A good practice is to define as few security proxy properties as you think you need and then add more later if necessary.

110

Developing Applications with IBM FileNet P8 APIs

Chapter 5.

Introduction to Process Engine API programming

In this chapter, we introduce the concepts and elements of the IBM FileNet Business Process Manager API. We also introduce the usage of workflow-related operations to develop process applications using the available Process Engine (PE) APIs. Most of the code snippets we show in this chapter are from the sample applications that we created for this book. For more details about the sample applications, refer to Chapter 7, Sample applications for Fictional Auto Rental Company A on page 223. This chapter discusses the following topics: Process Engine API overview Establishing a Process Engine session Handling API exceptions Launching a workflow Search work items Process work items Work with process status

Copyright IBM Corp. 2009. All rights reserved.

111

5.1 Process Engine API overview

PE provides a number of API packages available for process application development. Among this set of APIs are packages that are most commonly used for PE programming, which at the same time are the base for other sets of PE APIs. In this book, we present the following PE APIs: Process Engine Java API This interface includes all the functionality exposed by the PE. It is the API base for both the PEWS and REST API interfaces used for complex PE development. Process Engine Web Services Use this API for .NET development, which uses wrapper classes generated from the Web Service Description Language (WSDL) of the Process Engine Web Services (PEWS). This API provides functionality for most of the PE use cases presented in this chapter. Process Engine REST API The Representational State Transfer (REST) service is included in a servlet deployed in the WorkplaceXT application. This service is a style of accessing PE resources over HTTP using query string format for both request and response data. The REST API resources use the JavaScript Object Notation (JSON) MIME type as a lightweight data-interchange format. For further details about the JSON response schemas for the REST resources, refer to the following P8 documentation path: ECM Help Developer Help Process Engine Development Developers Guide Process Engine REST Service Process Engine REST Service Reference. PE also provides a Process Orchestration API, which enables the coordination of events within a workflow definition based on Web services. This API is implemented through the Web services adaptor as part of Component Integrator. For further details about the Component Integrator Web services adaptor and how to implement it, refer to Chapter 8: Implementing Component integrator and Web services in Introducing IBM FileNet Business Process Manager, SG24-7509. For PE development, the solution platform and the process use cases dictate which API to use. For example, if you plan to implement a complex process application by using the J2EE platform, select the PE Java API for this scenario. In this book, we include code snippets showing how to implement the PE APIs for each of the use cases related to PE development. Because the PE Java API

112

Developing Applications with IBM FileNet P8 APIs

exposes all PE functionality, this API is used for all use cases that are presented; however, for certain cases, we also show how those cases can be implemented using the PEWS API for .NET clients or the PE REST API, depending on the use case and whether the API includes the required functionality.

5.1.1 Functional groups

There are over 100 classes in the PE Java API. Although all these classes are arranged in the single filenet.vw.api package, there are three functional groups within this API. The following list describes the three functional groups that comprise the PE Java API package: Administration and configuration API Use this API to access and modify system-wide administration information or tasks, such as initializing and emptying regions or removing databases, accessing configuration information for an isolated region, and accessing information for a user and groups. Workflow definition API This API exposes the workflow definition classes to create, delete, and access a workflow and its sub-objects such as step and field objects. Runtime API This API handles database and work objects that are used to log in to the PE and create a process session, retrieve work objects, query queues, retrieve step processor information, and perform other workflow-related operations. Figure 5-1 presents the stack of primary objects (classes) in the PE Runtime API, which is used to perform the most common actions for process application development. This chapter focuses on use cases related to the PE Runtime API.

Figure 5-1 Primary Runtime API objects

Chapter 5. Introduction to Process Engine API programming

113

5.1.2 Available API functionality

To better understand the functionality available across the PE APIs presented in this book, Table 5-1 lists available operations for each API.
Table 5-1 Process Engine APIs and available functionality Process Engine API Java API PEWS API Available functionality This API exposes full functionality included in all PE API functional groups. This API exposes limited functionality available in the PE Runtime API. The operations that are available by using this API are: Launch a workflow Get queue elements Get queue names Get roster elements Get roster names Get step elements from a roster Get step elements from a queue Get work class names Retrieve step element Reassign step element Unlock step element Update step element Get milestones This API exposes limited functionality available in the PE Runtime API. The operations that are available by using this API are: Launch a workflow Get queue elements Get queue elements count Get application space roles Get role definition Retrieve step element Update step element Get queue workbasket definition Get queue workbasket attributes Get workbasket column definitions Get workbasket filter definitions Get workbasket filter attributes Get work classes collection Get work class definition

REST API

114

Developing Applications with IBM FileNet P8 APIs

5.1.3 Naming conventions

The PE Java API includes naming conventions for objects and methods to assist process application developers in keeping the usage of PE programming more understandable by making the APIs easier to read. Table 5-2 describes the naming conventions used by PE API objects.
Table 5-2 Process Engine API objects naming convention Object naming convention VW + ObjectName VWObjectName + Element VWObjectName + Definition VWObjectName + Type Description Convention used by API base objects Convention used by API objects to access a base object element or instance Convention used by API objects to retrieve and administer base objects Convention used by API objects to retrieve a base objects type Examples VWRoster, VWQueue VWRosterElement, VWQueueElement VWRosterDefinition, VWQueueDefinition VWFetchType, VWAttachmentType

PE Java API object methods also use naming conventions to perform actions over certain PE objects. Table 5-3 on page 116 describes the naming convention used by PE API object methods.

Chapter 5. Introduction to Process Engine API programming

115

Table 5-3 Process Engine API object methods naming convention Object method fetchMethod() Description Retrieves an object from a PE isolated region executing a database operation. Retrieves a parameter from the PE object. These methods usually do not require additional database operations because the parameters are retrieved from cache. Instantiates a VW object, retrieving it from the PE isolated region. Sets a parameter value for a PE object. Tests presence of an object state. Processes work on the PE isolated region, usually with a void return type. Examples VWRoster.fetchCount() method gets the number of work item in the roster. VWRoster.getName() method returns the translated name.

getMethod()

createMethod()

VWQueue.createQuery() method performs a filtered fetch of queue items and returns a VWQueueQuery object. VWDataField.setValue() method sets the value of the data contained in an editable field. VWDataField.isArray() method determines whether the data field is an array. VWWorkObject.doDispatch() method saves changes made in the work object, unlocks it, and moves the current step to the next workflow step. VWSession.convertIdToUserName () method converts the user ID assigned by PE to a user name.

setMethod()

isMethod()

doMethod()

convertMethod()

Retrieves an object parameter corresponding to another one.

5.1.4 Core classes

The following API classes are the most commonly used according to their typical purpose: PE session class PE error handling class Create and process work classes Search work classes Retrieve work status classes

116

Developing Applications with IBM FileNet P8 APIs

The PE session class consists of the VWSession object used to establish a session and log on to a PE server. Several PE operations can be performed, such as querying rosters and queues, retrieving a list of rosters or queues, administering the system, and establishing an audit trail. The PE error handling class consists of the VWException object, which handles all workflow-related exceptions. The create and process work classes are used to create a workflow processes, get and assign parameter values to work items, and process work items across the workflow life cycle. This group includes the following classes: VWWorkObject VWStepElement VWParameter VWFieldType VWDataField The search work classes are used to find work items either on a PE roster, work queue, or event log. This group includes the following classes: VWRoster VWRosterQuery VWRosterElement VWQueue VWQueueQuery VWQueueElement VWLog VWLogQuery VWLogElement The retrieve work status classes are used to fetch step history data for workflow process associated with a workflow map and information related to milestone events. This group includes the following classes: VWProcess VWWorkflowDefinition VWWorkflowHistory VWStepHistory VWStepOcurrenceHistory VWStepWorkObjectHistory VWWorkflowMilestones VWMilestoneElement

Chapter 5. Introduction to Process Engine API programming

117

5.1.5 Functional relationship

Figure 5-2 illustrates important functional relationships and calling sequences among the core PE Runtime API classes. The figure includes the core classes and methods that are the focus of this section.

Figure 5-2 Runtime API core classes calling sequences

In the next section, we put in practice these API classes for process use cases.

118

Developing Applications with IBM FileNet P8 APIs

5.2 Establishing a Process Engine session

To establish or log on to a PE session, the following elements are required: A connection point must be specified. The Content Engine (CE) host must be accessible. The CE host Uniform Resource Identifier (URI) must be set. Note: The CE URI is not needed for Process Engine Web Service (PEWS) connections. The PEWS is deployed as part of the CE enterprise application.

Connection point
A connection point name identifies a specific isolated region in the workflow database connecting the PE API to an associated isolated region. A defined connection point consists of a PE server, communications port, and isolated region number. For Process Engine Web Service (PEWS) clients, the PE connection point must be specified in the SOAP header request using the router SOAP header element. If no router value is specified, the default value is used. The default connection point for the Process Engine Web Service is PEWSConnectionPoint. PE API usage tips: Performance improvements can be achieved by minimizing the number of Remote Procedure Calls (RPCs) to the PE. Methods that fetch data or cause certain action to occur typically cause RPCs to the PE server, whereas methods that get and set data are local to the object. Similarly, minimize logging on and logging off the PE.

5.2.1 Java API scenario

The VWSession object establishes a session and logs onto the PE. This object allows the caller to query rosters and queues, retrieve lists of roster and queue names, and administer the system. PE relies on CE for authentication and directory service access operations. For further details about the available transport protocols, see the CE transport protocols from Chapter 3, Introduction to Content Engine API programming on page 39. The PE Java API offers two data transport layers for connecting to CE, the native Enterprise JavaBeans (EJB) transport that is offered by the application server or the Content Engine Web Services (CEWS) transport.

Chapter 5. Introduction to Process Engine API programming

119

Setting the Content Engine URI

The CE URI can be specified in any of the following ways: Call VWSession.setBootstrapCEURI to specify the URI as a String. Call VWSession.setBootstrapConfiguration to specify the URI as an InputStream. Specify the value of the system property filenet.pe.bootstrap.ceuri. Specify the value of the RemoteServerUrl property in a file named WcmApiConfig.properties located in your application's class path. Table 5-4 shows the possible CE URIs to be used depending on the selected connection transport and application server where CE is deployed.
Table 5-4 Content Engine connection URIs Protocol Web services EJB/WebSphere EJB/Weblogic EJB/Jboss URI http://<server>:<port>/wsi/FNCEWS40MTOM/ iiop://<server>:<port>/FileNet/Engine t3://<server>:<port>/FileNet/Engine jnp://<server>:<port>/FileNet/Engine

Note: You might sometimes see CE connection URIs with a prefix of cemp:. This prefix is for historical purposes only and is ignored by the PE APIs. When you are specifying URIs directly, there is no reason to use that prefix.

Authentication
This section assumes the authentication mechanism between CE and the directory service is properly configured. Example 5-1 on page 121 shows how to create the VWSession object, set the CE URI, and log on to the PE passing the required parameters.

120

Developing Applications with IBM FileNet P8 APIs

Example 5-1 Getting the initial connection using Java API

// User Information String userName = "Administrator"; String password = "filenet"; // Connection Point String connectionPoint = "CEPoint"; // Create a Process Engine Session Object VWSession myPESession = new VWSession(); // Set Bootstrap Content Engine URI myPESession.setBootstrapCEURI("iiop://ceserver:2809/FileNet/Engine"); // Log onto the Process Engine Server myPESession.logon(userName, password, connectionPoint);

Session logoff
The sessions logoff method ends the session with the PE and frees all resources. Although the system calls this method internally, an explicit call makes finalization more immediate and certain, as shown in Example 5-2. Otherwise, finalization is uncertain even when the Java virtual machine (JVM) shuts down.
Example 5-2 Log off from Process Engine using Java API

// Log off from the Process Engine myPESession.logoff();

5.2.2 PEWS API scenario

Unlike the PE Java API, the Process Engine Web Service (PEWS) API implements only the Web services transport for communicating with the CE. In fact, that connectivity is implicit because the PEWS server is collocated in the same J2EE application as the CEWS server. The PEWS URL must be specified by the client in order to lookup for the PEWS Web Service Description Language (WSDL) available through the CE. The URL for the Process Engine Web Service is as follows: http://content_engine_host:port/wsi/ProcessEngineWS

Chapter 5. Introduction to Process Engine API programming

121

The SOAP header request must include the correct elements and values, such as account credentials. These credentials are verified when functional operations are attempted. Example 5-3 shows how to initialize the PEWS object and set the correct parameter values including the request SOAP context.
Example 5-3 Creating a PEWS RequestSoapContext header using a C# client

// User Information string userName = "Administrator"; string password = "filenet"; // Connection Point Value string connectionPoint = "CEPoint"; // URL for the PEWS string wsUrl = "http://ceserver:9080/wsi/ProcessEngineWS"; // PE Web Service Policy private static Policy pePolicy = new Policy(new PolicyAssertion[] {new PEAssertion(),new PESecurityAssertion()}); // Create the WS Service Port object peWS.ProcessEngineServiceWse peWSServicePort = new peWS.ProcessEngineServiceWse(); // Set PE Web Service Policy peWSServicePort.SetPolicy(pePolicy); // Set connection point name and wsURL parameters peWSServicePort.RequestSoapContext["router"]=connectionPoint; peWSServicePort.Url = wsUrl; // Create User Context for the RequestSoapContext SecurityToken tok = new UsernameToken(userName, password, PasswordOption.SendPlainText); UserContext.SetProcessSecurityToken(tok); In the example, the C# client application uses authentication helper classes to initiate a Process Engine Web Service Policy object and a UserContext object, which are required to interject the WS-Security and other SOAP headers for authentication purposes. You can refer to these helper classes that are included with the pick-up and drop-off kiosk application as part of the provided sample application in this book.

122

Developing Applications with IBM FileNet P8 APIs

See Chapter 7, Sample applications for Fictional Auto Rental Company A on page 223.

5.2.3 REST API scenario

For the REST API, there is no explicit PE logon action, however, the REST API resource requests require the calling client to be authenticated with the application server container. The HTTP Basic authentication is supported as the default authentication method. The HTTP Basic authentication passes credentials in the clear, so you should secure the connection with Transport Layer Security and Secure Sockets Layer (TLS/SSL) or other means. Note: The LDAP registry referenced must be the same registry used by the CE. The client's authorization for accessing REST operations on a specified resource is determined by the permissions that are assigned to the resource.

5.3 Handling API exceptions

Exception handling is the way that applications detect and recover from exceptional conditions. Exceptional conditions are any unexpected occurrences that are not accounted for in a system's normal operation. Exception handling techniques can be separated into two main categories: Expected exception handling Unexpected exception handling In certain cases, expected exception handling is capable of doing forward error recovery, but both expected and unexpected exception handling methods can perform backward error recovery. Forward error recovery can mask any exceptional occurrences and continue normal operation. Backward error recovery must halt normal system execution and attempt to return to a previous normal state to continue execution and retry the operation. PE has two models to handle these exception categories for process applications, reflecting the C++ implementation of a server module and the Java implementation of the server and remaining modules. In this section, we describe how to handle PE Java exceptions.

Chapter 5. Introduction to Process Engine API programming

123

5.3.1 VWException object

VWException is the base class for all workflow exceptions and is the only exception thrown by PE Java API method calls. Table 5-5 shows the VWException main methods to get the String properties related to a PE Java exception.
Table 5-5 Principal String properties of the VWException object Method VWException.getKey() Description Gets the exception key, which is used in looking up a resource in the VWExceptions file. This method shows what interface produced the exception Contains the associated string for the key from the exceptions resource file. Customized messages can be obtained from the exceptions resource file Returns the class name for the cause of the PE exception. Returns the description for the cause of the PE exception.

VWException.getMessage()

VWException.getCauseClassName() VWException.getCauseDescription()

5.3.2 Steps to handle an exception

The following steps show how to handle a PE exception by using the Java API: 1. Include a try-catch block. 2. Call a method that can throw a VWException object. 3. Catch the exception, and identify and handle the error. Example 5-4 shows how to handle an expected exception during PE logon. In this particular example, an empty password value is specified; therefore, an exception is thrown.
Example 5-4 Handle Process Engine exceptions using the Java API

// User Information String userName = "Administrator"; String password = ""; // Connection Point String connectionPoint = "CEPoint";

124

Developing Applications with IBM FileNet P8 APIs

// Create a Process Engine Session Object VWSession myPESession = new VWSession(); // Set Bootstrap Content Engine URI myPESession.setBootstrapCEURI("iiop://ceserver:2809/FileNet/Engine"); // Start try-catch block to handle exception when log onto PE try { myPESession.logon(userName, password, connectionPoint); } catch (VWException vwe) { System.out.println("\nVWException Key: " + vwe.getKey() + "\n"); System.out.println("VWException Cause Class Name: " + vwe.getCauseClassName() + "\n"); System.out.println("VWException CauseDescription: " + vwe.getCauseDescription()+ "\n"); System.out.println("VWException Message: " + vwe.getMessage()); // Perform error recovery ... }

5.4 Launching a workflow

IBM FileNet Business Process Manager combines processes with active content management. This combination facilitates a variety of process management scenarios. Certain scenarios might require launching workflow processes automatically within applications. The following steps retrieve and launch a workflow process using the PE Java and Web services APIs: 1. Create a VWSession object and log on to PE. 2. Retrieve the workflow definition (work class) names from the VWSession object. 3. Create the workflow process for the selected work class. 4. Get and set workflow parameters for the launch step (getting and setting parameters is described in 5.6, Process work items on page 140). 5. Dispatch the workflow launch step. Steps to retrieve and launch a workflow using the PE REST API are described in 5.4.1, REST API scenario on page 127.

Chapter 5. Introduction to Process Engine API programming

125

Example 5-5 and Example 5-6 show how to retrieve and create a workflow using both the Java API and a PEWS client.
Example 5-5 Create a workflow process using the Java API

// Create session object and log onto Process Engine ... // Workflow name to launch String workflowName = "myWorkflow"; // Retrieve transfered work classes String[] workClassNames = myPESession.fetchWorkClassNames(true); for (int i=0;i<workClassNames.length;i++) System.out.println(workClassNames[i] ); // Launch Workflow VWStepElement stepElement = myPESession.createWorkflow(workflowName); // Get and Set Workflow parameters for the Launch Step ... // Dispatch Worflow Launch Step stepElement.doDispatch();
Example 5-6 Create a workflow process using PEWS

// Create the PEWS RequestSoapContext header ... // Workflow name to launch string workflowName = "myWorkflow"; // Get the available workclasses ArrayList arrayWorkClasses = new ArrayList(); string[] workClassNames = peWSServicePort.getWorkClassNames(); foreach (string workClassName in workClassNames) { ... } private peWS.StepElement stepElement; stepElement = peWSServicePort.createWorkflow(workflowName); // Get and Set Workflow parameters for the Launch Step ...

126

Developing Applications with IBM FileNet P8 APIs

// Dispatch Worflow Launch Step peWS.UpdateStepRequest updStepRequest = new peWS.UpdateStepRequest(); peWS.UpdateFlagEnum updFlagEnum = peWS.UpdateFlagEnum.UPDATE_DISPATCH; updStepRequest.stepElement = stepElement; updStepRequest.updateFlag = updFlagEnum; peWSServicePort.updateStep(updStepRequest);

5.4.1 REST API scenario

The following steps launch a workflow process by using the PE REST API: 1. Send HTTP GET using the workclasses resource to retrieve work class information. 2. Get and update work class launch step fields. 3. Send HTTP POST using the workclasses resource to create the workflow. Note: For this operation, the Post Once Exactly (POE) parameter value must be set to 1 (POE:1) in both request headers. Example 5-7 shows how to create a workflow process with a Dojo toolkit client.
Example 5-7 Create a workflow process with a Dojo toolkit client // Ensure to be authenticated with the Application Server container // Set request URI var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; // Set REST API resource and wokflow name var workflowName = "myWorkflow"; var url = baseURL + "workclasses/" + workflowName; var newWorkItemURL = ""; var defaultWorkObjectValue; // Use HTTP GET Method to retrieve work class information dojo.xhrGet({ url: url, // Handle the return result as JSON object handleAs: "json-comment-optional", // Set the POE (POST Once Exactly) parameter in header to true headers: {"POE": "1"}, // The response callback load: function(data, ioArgs) { // Get the resource URL for the new work object to create newWorkItemURL = ioArgs.xhr.getResponseHeader('POE-Links');

Chapter 5. Introduction to Process Engine API programming

127

// Get the work object's default values defaultWorkObjectValue = data; }, error: function(data) { // The error callback console.dir(data); }, sync: true } ); // Set URL for POST request url = baseURL + newWorkItemURL; var workObject = defaultWorkObjectValue; // Update work class lauch step fields workObject.dataFields["CompanyName"].value = "IBM"; // Set the attachment value workObject.attachments["Documents"].value = [{"vsId":"{997E0AAC-8068-4990-AC08-D748E3063162}", "title":"MyDocument1", "type":3, "libraryName":"objectstore", "libraryType":3, "version":"{D48EE0A9-61CA-4DAD-8B61-961CE362434B}", "desc":"web.xml"}]; // Use HTTP POST Method to create the new work object dojo.rawXhrPost({ // the response callback url: url, handleAs: "json-comment-optional", headers: {"Content-Type": "application/json", "POE": "1"}, _raw: true, // Post back the modified work object postData: dojo.toJson(workObject), load: function(data, ioArgs) { console.log(dojo.toJson(data)); }, error: function(data) { // the error callback console.log(data); }, sync: true } );

128

Developing Applications with IBM FileNet P8 APIs

5.5 Search work items

All work searching operations are performed on PE database objects. The primary database objects are: VWRoster A workflow roster stores the current location of and other information about all workflows. Workflow rosters provide the PE with an efficient way to locate specific workflows and work items. VWQueue A queue contains each individual work item in a table that is waiting to be processed. Each entry in the roster has a corresponding entry in a queue. The API can retrieve and modify work items in work queues and user queues. VWLog Event logs contain a record of specific system- or workflow-related events for each isolated region. This type of logging is useful for tracking workflow activity. Each event that is logged is part of an event category and has an associated event number. These PE objects are stored as tables in the database. In addition, a set of views are created to enable direct database access from a custom application. In this section, we present how to search work by using the APIs (however, direct read-only access to the database views is supported). When the PE database is accessed directly, it must only be for viewing information in a read-only manner. Any modification of table data without using the provided APIs or applications is not supported. Work search operations that use the PE REST API are based on Workbasket properties. For further details about how to search work items by using the REST API, see 6.2.3, Retrieve workbasket on page 192. Also see 6.2.4, Query work items from workBasket on page 194.

Chapter 5. Introduction to Process Engine API programming

129

To optimize PE queries: Set the appropriate query flags to return only user fields from roster, queue, and log queries. For example, to avoid retrieving system fields and helper data from the work items returned, set the query flags for the queryFlags parameter of createQuery(): QUERY_GET_NO_SYSTEM_FIELDS QUERY_GET_NO_TRANSLATED_SYSTEM_FIELDS Note that helper methods on elements are not functional without translated system fields. Create indexes and limit the number of indexes to achieve more efficient queries with less data transmitted. Indexes must be unique. Use filters to get ranges of elements for more efficient queries with less data transmitted. Use filters to limit elements retrieved to a specified range or to set a maximum number of objects to be retrieved for each fetch from the PE server (default is 50 items).

Create query parameters

To successfully return results from any of the PE database objects, the fetch criteria must be correctly specified by the input query parameters. Table 5-6 describes the parameters included in the createQuery method to query a PE database object by using the APIs.
Table 5-6 Parameters to build a query by using the APIs Parameter indexName minValues maxValues queryFlags filter Description Specify the name of the database search index for the queue. Specify an array of objects containing the minimum values with which to compare the index fields. Specify an array of objects containing the maximum values with which to compare the index fields. Specify a value for the search options Specify a SQL WHERE clause to be used as a filter; has placeholder variables that are replaced by values in substitutionVars parameter. Specify an array of objects containing the values to substitute for the placeholder variables specified in filter. Specify a value associated with the element type to be retrieved using the object returned by the query object.

substitutionVars fetchType

130

Developing Applications with IBM FileNet P8 APIs

Note: Use VWWorkObjectNumber class as a substitution value type for queries when using F_WobNum or F_WorkFlowNumber. The type is substituted into the filter string specified. This class takes care of formatting the work object or workflow number so the developer does not have to.

Query flags
These flags represent an integer value that specifies the search options. Setting this parameter refines the search for the work items that are specified by the search index. Table 5-7 shows the available query flags for work search.
Table 5-7 Process Engine query flag options for work search Query flag QUERY_GET_NO_SYSTEM_ FIELDS QUERY_GET_NO_TRANSLAT ED_SYSTEM_FIELDS QUERY_MAX_VALUES_ INCLUSIVE QUERY_MIN_VALUES_ INCLUSIVE QUERY_NO_OPTIONS QUERY_READ_UNWRITABLE QUERY_RESOLVE_NAMES Integer value 1024 2048 64 32 0 4 8192 Description Returns objects without system fields. Returns objects without translated fields. Specifies the inclusion of maximum search values. Specifies the inclusion of minimum search values. Specifies no search option. Specifies the return of read-only work items. Supports conversion of user names to IDs for Process Engine Web services to perform queries using user ID fields within an index or filter.

5.5.1 Query a roster

The following steps retrieve work items from a workflow roster by using the PE Java and Web services APIs: 1. 2. 3. 4. 5. Create a VWSession object and log on to PE. Retrieve the roster to search and the roster count. Set query parameters. Perform the query. Process the results.

Chapter 5. Introduction to Process Engine API programming

131

To optimize roster queries: Expose only infrequently updated fields. Exposed data fields on rosters are more costly for updates. Set an optimal buffer size on the roster-related queries. For example, consider increasing the buffer size (default is 50, maximum is 200) to minimize round-trips to the PE server. Query roster elements instead of work objects. Query work objects through the roster causes the application to access database records twice instead of once. Example 5-8 and Example 5-9 on page 133 show how to retrieve work items from a workflow roster by using both the Java API and a PEWS client.
Example 5-8 Query workflow roster using the Java API

// Create session object and log onto Process Engine ... // Set Roster Name String rosterName= "DefaultRoster"; // Retrieve Roster Object and Roster count VWRoster roster = myPESession.getRoster(rosterName); System.out.println("Workflow Count: " + roster.fetchCount()); // Set Query Parameters int queryFlags=VWRoster.QUERY_NO_OPTIONS; String queryFilter="F_WobNum=:A"; String wobNum = "5BDD567B70453C48A5388AC56ABC69E4"; // VWWorkObjectNumber class takes care of the value format // used in place of F_WobNum and F_WorkFlowNumber Object[] substitutionVars = {new VWWorkObjectNumber(wobNum)}; int fetchType = VWFetchType.FETCH_TYPE_ROSTER_ELEMENT; // Perform Query VWRosterQuery query = roster.createQuery (null,null,null,queryFlags,queryFilter,substitutionVars,queryType); // Process Results while(query.hasNext()) { VWRosterElement rosterItem = (VWRosterElement) query.next(); System.out.println("WF Number: "+rosterItem.getWorkflowNumber());

132

Developing Applications with IBM FileNet P8 APIs

System.out.println("WOB Number: "+rosterItem.getWorkObjectNumber()); System.out.println("F_StartTime: " + rosterItem.getFieldValue("F_StartTime")); System.out.println("F_Subject: " + rosterItem.getFieldValue("F_Subject")); }

Example 5-9 Query workflow roster using PEWS // Create the PEWS RequestSoapContext header ... // Set Roster Name string rosterName= "DefaultRoster"; // Set Query Parameters peWS.GetRosterElementsRequest query = new peWS.GetRosterElementsRequest(); query.queryFlags = new peWS.QueryFlagEnum?[] {peWS.QueryFlagEnum.QUERY_READ_LOCKED, peWS.QueryFlagEnum.QUERY_LITE_ELEMENTS}; query.nToReturn = 100; query.rosterName = rosterName; // Perform Query peWS.RosterElement[] elements = peWSServicePort.getRosterElements(query); // Process Results foreach (peWS.RosterElement element in elements) { peWS.Field[] fields = element.Fields; foreach(peWS.Field field in fields) { // Only process the field if there is a value if (field.Values.Length > 0) { // Process field values //field.Name; //field.Values[0].Item.GetType(); //field.Values[0].Item; } } }

Chapter 5. Introduction to Process Engine API programming

133

5.5.2 Workflow queues

A workflow queue contains each individual work item in a table that is waiting to be processed. Each entry in the roster has a corresponding entry in a queue. The three workflow queue categories are: User queues Each user has an inbox that holds work items assigned to that user. A user can also have a queue of tracker items. Work queues A work queue holds work items that can be completed by one of a number of users, rather than by a specific participant, or work items that can be completed by an automated process. System queues System queues (Conductor, Delay, and InstructionSheetInterpreter) hold work items that are undergoing system processes or are waiting. Note: The following cases describe the differences between the Inbox and the Inbox(0) user queues: Use Inbox queue to retrieve the inbox work items of the user that is currently logged on. Use Inbox(0) queue to retrieve inbox work items of all users. Use Inbox(0) queue and F_BoundUser system field in the query filter to retrieve inbox work items of a specific user (or users).

Query queues
The following steps retrieve work items from a workflow queue using the PE Java and Web services APIs: 1. 2. 3. 4. 5. 6. Create a VWSession object and log on to PE. Retrieve the queue to search. Get the queue depth. Set the query parameters. Perform the query. Process the results returned by the query.

The same steps are included to retrieve work items from a workflow queue for all queue categories described above. Refer to Table 5-6 on page 130 for a detailed description of the required parameters that are included in the createQuery method to query a queue by using the PE APIs.

134

Developing Applications with IBM FileNet P8 APIs

To optimize queue queries: Use the QUERY_LOCK_OBJECTS flag. This avoids the necessity of an additional RPC to check the lock status of an object (or to lock an object). Using the flag also reduces lock contention. Set an optimal buffer size on the query. For example, consider increasing the buffer size (default is 50, maximum is 200) to minimize round-trips to the PE server. Partition, if possible, to use more than one queue. Example 5-10 and Example 5-11 on page 136 show how to retrieve work items from a workflow queue using both the Java API and a PEWS client.
Example 5-10 Query workflow queue using the Java API

// Create session object and log onto Process Engine ... // Queue Name String queueName = "Inbox"; // Retrieve the Queue to be searched and Queue depth VWQueue queue = myPESession.getQueue(queueName); System.out.println("Queue Depth: " + queue.fetchCount()); // Set Query Parameters String wobNum = "5BDD567B70453C48A5388AC56ABC69E4"; VWWorkObjectNumber wob = new VWWorkObjectNumber(wobNum); VWWorkObjectNumber[] queryMin = new VWWorkObjectNumber[1]; VWWorkObjectNumber[] queryMax = new VWWorkObjectNumber[1]; String queryIndex = "F_WobNum"; queryMin[0] = wob; queryMax[0] = wob; int queryFlags = VWQueue.QUERY_MIN_VALUES_INCLUSIVE + VWQueue.QUERY_MAX_VALUES_INCLUSIVE; int fetchType = VWFetchType.FETCH_TYPE_QUEUE_ELEMENT; // Perform Query VWQueueQuery queueQuery = queue.createQuery(queryIndex,queryMin, queryMax,queryFlags,null,null,fetchType); // Process Results while(queueElement.hasNext()) {

Chapter 5. Introduction to Process Engine API programming

135

VWQueueElement queueElement = (VWQueueElement) queueQuery.next(); System.out.println("Username: "+ queueElement.getFieldValue("F_BoundUser").toString()); System.out.println("WOB Number: "+ queueElement.getWorkObjectNumber()); System.out.println("F_Subject: " + queueElement.getFieldValue("F_Subject")); }
Example 5-11 Query workflow queue using PEWS // Create the PEWS RequestSoapContext header ... // Set Queue Name string queueName= "Inbox"; // Set Query Parameters peWS.GetQueueElementsRequest query = new peWS.GetQueueElementsRequest(); query.queryFlags = new peWS.QueryFlagEnum?[1] { peWS.QueryFlagEnum.QUERY_READ_LOCKED }; query.nToReturn = 100; query.queueName = queueName; // Perform Query peWS.QueueElement[] queueelements = peWSServicePort.getQueueElements(query); // Process Results foreach (peWS.QueueElement queueelement in queueelements) { peWS.Field[] fields = queueelement.Fields; foreach(peWS.Field field in fields) { // Only process the field if there is a value if (field.Values.Length > 0) { // Process field values //field.Name; //field.Values[0].Item.GetType(); //field.Values[0].Item; } } }

136

Developing Applications with IBM FileNet P8 APIs

5.5.3 Query event log

The Process Configuration Console automatically creates an event log (that is called DefaultEventLog) for each isolated region including a set of event-logging options. The logging options determine whether PE logs a message when certain events occur within the region. Each event-logging option represents an event category; if enabled, the occurrence of any event within that category generates a log message. Note: A custom (user-defined) message can also be logged if a Log system function is executed within a workflow.

Event-logging categories
Each system event that is logged is part of an event category, and has an associated event number. Table 5-8 shows the primary event-logging categories used for workflow event tracking. For further details about all event-logging categories available, refer to the following P8 documentation path: ECM Help User Help Integrating workflow Workflow overview Process Engine Reference Events and statistics Event log categories.
Table 5-8 Primary event logging categories Event log category Creation Event logged VW_WOChildCreatio nMsg VW_WOParentCreati onMsg Termination VW_WOChildTermin ationMsg VW_WOParentTermi nationMsg VW_WFTermination Administration message VW_WOForcedToSki pInstructionMsg VW_WOForcedToTer minateMsg VW_WOForcedToDe leteMsg Event number 130 140 150 160 165 180 190 200 Description Records the creation of a child work item. Records the creation of a parent work item. Records the termination of a child work item. Records the termination of a parent work item. Records the completion of all work items in a workflow. Records when a work item is forced to skip an instruction. Records when a work item is forced to terminate. Records when a work item is forced to delete.

Chapter 5. Introduction to Process Engine API programming

137

Event log category Begin operation

Event logged VW_WPBeginServic eMsg VW_WPWorkObject QueuedMsg

Event number 350 352 360

Description Records when a step processor or user locks a work item. Records when a work item is queued. Records when a work item is updated and dispatched to the next queue. Records when a work item is saved while retaining the existing lock. Records when work item processing ends abnormally. Records when a work item is updated and unlocked in the same queue. Records when a work item is delegated to another user. Records when a work item is reassigned to another user. Records when a work item is returned to a user following delegation. Records when a user cancels, stops, or ends an operation. Records when a step is executed without specifying a queue. Records when a compound step is completed.

End operation

VW_WPEndService NormalMsg VW_WPWOBSaveW ithLockMsg VW_WPEndService AbnormalMsg VW_WPEndService ReleaseMsg VW_WPEndService ReleaseDelegateMs g VW_WPEndService ReleaseReassignMs g VW_WPEndService ReleaseReturnMsg VW_WPEndService AbortMsg

365

370 380

382

384

386

390 500

Empty step / System step

VVW_WOEmptyStep Msg VW_WOCompleteSy sStepMsg

510

The following steps retrieve work items from an event log using the PE Java API: 1. Create a VWSession object and log on to PE. 2. Retrieve the event log to search. 3. Set the query parameters.

138

Developing Applications with IBM FileNet P8 APIs

4. Perform the query and count the elements returned. 5. Process the results. Refer to Table 5-6 on page 130 for a detailed description of the required parameters included in the createQuery method to query a queue using the PE APIs. To optimize event log queries: Partition to use different logs for different workflows. Disable any logging options that you do not need. Note that although this improves performance, it also affects the Process Tracker and Process Analyzer. Manage log records. Use the vwlog tool to maintain logs. Use operating system scheduling tools to periodically remove unused log records. Save log information in text files. Example 5-12 shows how to retrieve work items from an event log by using the Java API.
Example 5-12 Query event log by using the Java API // Create session object and log onto Process Engine ... // Set Event Log Name String eventLogName = "DefaultEventLog"; // Retrieve the Event Log to be searched VWLog log = myPESession.fetchEventLog(eventLogName); // Set Query Parameters String queryFilter = "F_EventType = 240"; // user logged in event type int queryFlags = VWLog.QUERY_NO_OPTIONS; // Perform Query VWLogQuery query = log.startQuery(null,null,null,queryFlags,queryFilter,null); // Process Results VWLogElement logElement = (VWLogElement) query.next(); int eventType = logElement.getEventType(); String stringEventType = VWLoggingOptionType.getLocalizedString(eventType); String[] fieldNames = logElement.getFieldNames(); String FieldValue = logElement.getFieldValue(fieldNames[0]).toString();

Chapter 5. Introduction to Process Engine API programming

139

5.6 Process work items

Process applications enable users to manage information and access resources that are associated with a workflow. Developing a process application involves, among other things, accessing process data and resources, and interfacing to the PE to perform tasks that are associated with a workflow step. These tasks are usually performed by either a step processor or a work performer application. Step processor A step processor is an application that provides the information and resources for a participant to complete a step in a workflow. When a participant opens a work item at run time, the step processor displays the necessary instructions, attachments, field values, response options, and other resources that the participant needs to complete the work. Work performer A work performer is an application that performs an operation or set of operations that are associated with a workflow. Typically, work performers are designed without a user interface and are used to perform automatic workflow operations, such as those that are associated with a specific step in a workflow definition. These process applications must perform fundamental operations, which include: Retrieve step element. Get step element parameters. Set step element parameter values. Complete the work item.

5.6.1 Retrieve step element

To process a work item in a process application, the appropriate step element must be retrieved from the queue. The following steps are required to retrieve a step element from a workflow queue by using the PE Java and Web services APIs: 1. 2. 3. 4. 5. Create a VWSession object and log on to PE. Get the queue containing the work item. Set the query parameters to get step elements. Perform the queue query. Get the desired step element from the queue.

For further details about steps to retrieve a step element using the PE REST API, see Retrieve step element: REST API scenario on page 142.

140

Developing Applications with IBM FileNet P8 APIs

Example 5-13 and Example 5-14 show how to retrieve a step element from a user queue by using both the Java API and a PEWS client.
Example 5-13 Retrieve a step element from the Inbox by using the Java API

// Create session object and log onto Process Engine ... // Queue Name String queueName = "Inbox"; // Retrieve the Queue VWQueue queue = myPESession.getQueue(queueName); // Set Query Parameters String wobNum = "5BDD567B70453C48A5388AC56ABC69E4"; VWWorkObjectNumber wob = new VWWorkObjectNumber(wobNum); VWWorkObjectNumber[] queryMin = new VWWorkObjectNumber[1]; VWWorkObjectNumber[] queryMax = new VWWorkObjectNumber[1]; String queryIndex = "F_WobNum"; queryMin[0] = wob; queryMax[0] = wob; // Query Flags and Type to retrieve Step Elements int queryFlags = VWQueue.QUERY_READ_LOCKED; int queryType = VWFetchType.FETCH_TYPE_STEP_ELEMENT; VWQueueQuery queueQuery = queue.createQuery (queryIndex,queryMin,queryMax,queryFlags,null,null,queryType); // Get an individual Step Element VWStepElement stepElement = (VWStepElement) queueQuery.next();
Example 5-14 Retrieve a step element from the Inbox using PEWS

// Create the PEWS RequestSoapContext header ... // Set Queue Name string queueName= "Inbox"; // Set Query Parameters string wobNum= "5BDD567B70453C48A5388AC56ABC69E4"; peWS.RetrieveStepRequest request = new peWS.RetrieveStepRequest(); request.bLock = true; request.bOverrideLock = true; request.queueName = queueName; request.wobNum = wobNum;

Chapter 5. Introduction to Process Engine API programming

141

peWS.GetStepElementsFromQueueRequest qry = new peWS.GetStepElementsFromQueueRequest(); request.queryFlags = new peWS.QueryFlagEnum? []{peWS.QueryFlagEnum.QUERY_NO_OPTIONS}; // Perform Query and Retrieve Step Element peWS.StepElement = peWSServicePort.retrieveStep(request);

Retrieve step element: REST API scenario

The following steps retrieve a step element by using the PE REST API: 1. Send HTTP GET by using the stepelements resource to retrieve step element information, specifying queue name and work object number. 2. Process the response results. Example 5-15 shows how to retrieve a step element from the Inbox with Dojo toolkit client.
Example 5-15 Retrieve a step element from the Inbox with Dojo toolkit client

// Ensure to be authenticated with the Application Server container // Set request URI var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; // Set REST API resource and work item number var queueName = "Inbox"; var wobNumber = "5BDD567B70453C48A5388AC56ABC69E4"; var url = baseURL + "queues/" + queueName + "/stepelements/" + wobNumber; var stepElement; // Use HTTP GET method to retrieve the step element by specifying // the queue name and work object number dojo.xhrGet({ url: url, // Handle the result as JSON object handleAs: "json-comment-optional", // the callback to handle the response load: function(data) { stepElement = data; // Process the attachments for (var attachmentName in stepElement.attachments) { // .... console.log(attachmentName); }

142

Developing Applications with IBM FileNet P8 APIs

// Process the data fields for (var dataFieldName in stepElement.dataFields) { // .... console.log(dataFieldName); } // Process the system properties for (var systemPropName in stepElement.systemProperties) { // .... console.log(systemPropName); } // Process step processor //var stepProcessor = stepElement.stepProcessor; }, error: function(data) { // the error callback console.dir(data); } } );

5.6.2 Get step element parameters

To process a work item from a process application, current information included in the step element must be retrieved so that it can be displayed to the participant and to perform business rules that are based on the field properties. Table 5-9 on page 144 presents the required primary field flags necessary to properly handle step element and work item fields.

Chapter 5. Introduction to Process Engine API programming

143

Table 5-9 Primary process field flags Field flag Field type Values Integer (FIELD_TYPE_INT) String (FIELD_TYPE_STRING) Boolean (FIELD_TYPE_BOOLEAN) Float (FIELD_TYPE_FLOAT) Time (FIELD_TYPE_TIME) Attachment (FIELD_TYPE_ATTACHMENT) Participant (FIELD_TYPE_PARTICIPANT) All types (ALL_FIELD_TYPES) User (FIELD_USER_DEFINED) System (FIELD_SYSTEM_DEFINED) User and system (FIELD_USER_AND_SYSTEM_DEFINED) Read only (MODE_TYPE_IN) Write only (MODE_TYPE_OUT) Read and write (MODE_TYPE_IN_OUT) Description An integer value that specifies the parameter type or types

Field creator

An integer that specifies one of the field creator types An integer value that indicates the mode of the step parameter

Field mode

The following steps get step element parameters by using the PE Java and Web services APIs: 1. Retrieve the user- and system-defined VWParameters. 2. Get the name, data type, read and write mode, and value for each parameter. 3. Retrieve the step element responses. For further details about included steps to retrieve step element parameters using the PE REST API, see Get step element parameters: REST API scenario on page 146. Example 5-16 and Example 5-17 on page 145 show how to retrieve step element parameters and responses using both the Java API and a PEWS client.
Example 5-16 Get step element parameters and responses using the Java API

// Create session object and log onto Process Engine ... // Retrieve Step Element ... // Specify Field Type and Field Creator flags to retrieve // user- and system-defined parameters VWParameter[] parameters = stepElement. getParameters(VWFieldType.ALL_FIELD_TYPES, VWStepElement.FIELD_USER_AND_SYSTEM_DEFINED);

144

Developing Applications with IBM FileNet P8 APIs

// Get the name, type, mode, and value for each parameter for (int i = 0; i < parameters.length; i++ ) { // Get Parameter Name String parameterName = parameters[i].getName(); // Get Parameter Type (Field Type Flag) String parameterType = VWFieldType.getLocalizedString (parameters[i].getFieldType()); // Get Parameter Mode (Field Mode Flag) String parameterMode = VWModeType.getLocalizedString (parameters[i].getMode()); // Get Parameter Value String parameterValue =parameters[i].getStringValue(); } // Retrieve Step Element Responses String[] stepResponses = stepElement.getStepResponses(); if(stepResponses != null) { int len = stepResponses.length; for(int j = 0; j < len; j++ ) { // Process Response System.out.println(stepResponses[j]); } }
Example 5-17 Get step element parameters and responses using PEWS

// Create the PEWS RequestSoapContext header ... // Retrieve Step Element ... // Retrieve Step Element Parameters peWS.Parameter[] parameters = stepElement.Parameters; // Get the name, type, mode, and value for each parameter foreach(peWS.Parameter param in parameters) { // Get Parameter Name string paramName = param.Name; // Get Parameter Type (peWS.FieldTypeEnum) string paramType = param.Type; // Get Parameter Mode (peWS.ModeTypeEnum) string paramMode = param.Mode;

Chapter 5. Introduction to Process Engine API programming

145

// Get Parameter Value if( param.Values.Length > 0) { foreach( peWS.Value stepvalue in param.Values) { string paramValue = stepvalue.Item; } } } // Retrieve Step Element Responses peWS.ArrayOfResponse stepResponses = stepElement.Responses; if (stepResponses!=null) { foreach (string response in stepResponses.Response) { // Process Response System.Windows.Forms. MessageBox.Show("Step Element Response: "+response); } }

Get step element parameters: REST API scenario

The following steps retrieve step element parameters by using the PE REST API: 1. Send HTTP GET using the stepelements resource to retrieve the step element specifying queue name and work object number. 2. Process response results. Example 5-18 shows how to get step element parameters and responses with a Dojo toolkit client.
Example 5-18 Get step element parameters and responses with a Dojo toolkit client

// Ensure to be authenticated with the Application Server container // Set request URI var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; // Set REST API resource and work item number var queueName = "Inbox"; var wobNumber = "5BDD567B70453C48A5388AC56ABC69E4"; var url = baseURL + "queues/" + queueName + "/stepelements/" + wobNumber; // Helper funtion to print parameter properties

146

Developing Applications with IBM FileNet P8 APIs

var printParameter = function(parameter) { // get the attachment name var name = parameter.name; // get the mode var mode = parameter.mode; // get the type var type = parameter.type; // get the value var value = parameter.value; console.log("Attachment " + name + ": "); console.log(" - Mode: " + mode); console.log(" - Type: " + type); console.log(" - Value: " + value); }; // Use HTTP GET method to retrieve the step element by specifying // the queue name and work object number dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { // the callback to handle the response stepElement = data; var attachments = stepElement.attachments; var dataFields = stepElement.dataFields; var stepProcessor = stepElement.stepProcessor; var systemProperties = stepElement.systemProperties; // print the attachment parameters for (var attachmentName in attachments) { var attachment = attachments[attachmentName]; printParameter(attachment); } // print the dataField parameters for (var dataFieldName in dataFields) { var dataField = dataFields[dataFieldName]; printParameter(dataField); } // print the systemField parameters for (var systemPropName in systemProperties) { var systemPropValue = systemProperties[systemPropName]; console.log("systemProperty (" + systemPropName + "):"

Chapter 5. Introduction to Process Engine API programming

147

+ systemPropValue); } // print the step responses var responses = systemProperties.responses; for (var i = 0; i < responses.length; i++) { console.log("response (" + i + "):" + responses[i]); } }, error: function(data) { // the error callback console.dir(data); } } );

5.6.3 Set step element parameter values

To process work from a process application, step elements have to be locked in order to set parameter values, and either save or complete the work. Besides the common field types that comprise a workflow process, two workflow parameters require additional handling in order to set their values programmatically: workflow attachments and participants. In this section, we show how to get and set values for the common workflow field types, and for workflow attachments and participants: Workflow attachments This parameter specifies the workflow attachments that a participant can use to complete a step in the workflow. These attachments can be objects such as documents, document arrays, folders, stored searches, URLs, or files that are located on a shared file system. Workflow participants This parameter specifies the participants who can process work in the workflow. Participants can be a user, users, a group, or groups. When participants are assigned to a particular step in the workflow, and there are active work items in the step, the work items appear in the inbox of the participants. A workflow attachment can be a document, folder, or other type of objects. Table 5-10 on page 149 lists the workflow attachment types.

148

Developing Applications with IBM FileNet P8 APIs

Table 5-10 Workflow attachment types Attachment type ATTACHMENT_TYPE_CUSTOM_OBJECT ATTACHMENT_TYPE_DOCUMENT ATTACHMENT_TYPE_FOLDER ATTACHMENT_TYPE_LIBRARY Integer value 6 3 2 1 Description Indicates the attachment type is a custom object Indicates the attachment is the document type Indicates the attachment is the folder type Indicates the attachment type is library or Object Store Indicates the attachment is the stored search type Indicates the attachment type is not known Indicates the attachment type is a UNC or URL

ATTACHMENT_TYPE_STORED_SEARCH ATTACHMENT_TYPE_UNDEFINED ATTACHMENT_TYPE_URL

4 0 5

Note: The VWParticipant object represents the name of a user or user group in any of the following ways: Short (logon) name Distinguished name (DN) Display name (configurable) For a detailed description of the primary field flags required to properly handle step element and work item fields, see the list in Table 5-9 on page 144. The following steps are required to update step element parameter values using the PE Java and Web services APIs: 1. 2. 3. 4. 5. 6. Lock the step element. Check whether the parameter is read only. Switch through each data type for editable parameters. For each data type, check whether the parameter is single or an array. Set the value for the parameter. Set the value for the system-defined response parameter.

Chapter 5. Introduction to Process Engine API programming

149

For further details about included steps to update step element parameter values by using the PE REST API, see the Set step element parameter values: REST API scenario on page 154. Example 5-19 and Example 5-20 on page 152 show how to perform the steps required to set step elements parameter values, using both the Java API and a PEWS client.
Example 5-19 Set step element parameters and responses by using the Java API

// Create session object and log onto Process Engine ... // Retrieve Step Element ... // Retrieve Step Element Parameters ... // Lock the Step Element stepElement.doLock(true); // Process Step Element Parameters for (int i = 0; i < parameters.length; i++ ) { // Check parameter mode boolean readOnly = (parameters[i].getMode() == VWModeType.MODE_TYPE_IN); // If the parameter is editable, switch through each data type if (!readOnly) { // For each data type, // check whether the parameter is single or an array // and set the parameter value(s) switch (parameters[i].getFieldType()) { case VWFieldType.FIELD_TYPE_STRING: if (parameters[i].isArray()) { String[] arrParamValues = new String[] {"value_1", "value_2", "value_3"}; stepElement.setParameterValue (parameters[i].getName(),arrParamValues,true); } else { String paramValue = "value_1"; stepElement.setParameterValue (parameters[i].getName(),paramValue,true); } break;

150

Developing Applications with IBM FileNet P8 APIs

case VWFieldType.FIELD_TYPE_ATTACHMENT: if (!parameters[i].isArray()) { // Get the value for the VWAttachment VWAttachment attachment = (VWAttachment) parameters[i].getValue(); // Set the attachment name attachment.setAttachmentName("Document Title"); // Set the attachment description attachment.setAttachmentDescription ("A document added programmatically"); // Set the type of object (Document) attachment.setType (VWAttachmentType.ATTACHMENT_TYPE_DOCUMENT); // Set the library type and name (CE Object Store) attachment.setLibraryType (VWLibraryType.LIBRARY_TYPE_CONTENT_ENGINE); attachment.setLibraryName("ObjectStoreName"); // Set the document ID and version attachment.setId ("{BBE5AD7F-2449-4DC3-AA38-012A65EC4286}"); attachment.setVersion ("{BBE5AD7F-2449-4DC3-AA38-012A65EC4286}"); // Set the parameter value stepElement.setParameterValue (parameters[i].getName(),attachment,true); } break; case VWFieldType.FIELD_TYPE_PARTICIPANT: // Instantiate a new VWParticipant array VWParticipant[] participant = new VWParticipant[1]; // Set the participant name using username value String participantUserName = "Administrator"; participant[0].setParticipantName(participantUserName); // Set the parameter value stepElement.setParameterValue (parameters[i].getName(),participant,true); break; default: // Do not take action for other data types break; } }

Chapter 5. Introduction to Process Engine API programming

151

} // Set the value for the system-defined Response parameter if (stepElement.getStepResponses() != null) { String responseValue = "Ok"; stepElement.setSelectedResponse(responseValue); }
Example 5-20 Set step element parameters and responses by using PEWS

// Create the PEWS RequestSoapContext header ... // Retrieve and Lock Step Element // request.bLock = true; // request.bOverrideLock = true; ... // Retrieve Step Element Parameters // Process Step Element Parameters foreach(peWS.Parameter param in parameters) { // If the parameter is editable, switch through each data type bool readOnly = param.Mode == peWS.ModeTypeEnum.MODE_TYPE_IN; if (!readOnly) { switch(param.Type) { case peWS.FieldTypeEnum.FIELD_TYPE_STRING: // Set the parameter value string paramValue = "value_1"; peWS.Value stepvalue = new peWS.Value(); peWS.Value[] values = {paramValue}; param.Values = values; param.Modified = true; break; case peWS.FieldTypeEnum.FIELD_TYPE_ATTACHMENT: // Get the value for the attachment object peWS.Attachment peAttachment = (peWS.Attachment)stepvalue.Item; // Set the attachment name peAttachment.Name = "Document Title"; // Set the attachment description peAttachment.Description =

152

Developing Applications with IBM FileNet P8 APIs

"A document added programmatically"; // Set the library type and name (CE Object Store) peAttachment.LibraryType = peWS.LibraryTypeEnum.LIBRARY_TYPE_CONTENT_ENGINE; peAttachment.Library = "ObjectStoreName"; // Set the type of object (Document) peAttachment.Type = peWS.AttachmentTypeEnum.ATTACHMENT_TYPE_DOCUMENT; // Set the document ID and version peAttachment.Id = "{BBE5AD7F-2449-4DC3-AA38-012A65EC4286}"; peAttachment.Version = "{BBE5AD7F-2449-4DC3-AA38-012A65EC4286}"; // Set the parameter value peWS.Value stepvalue = new peWS.Value(); peWS.Value[] values = {peAttachment}; param.Values = values; param.Modified = true; break; case peWS.FieldTypeEnum.FIELD_TYPE_PARTICIPANT: // Set the participant name using username value string participantUserName = "Administrator"; // Set the parameter value peWS.Value stepvalue = new peWS.Value(); peWS.Value[] values = {participantUserName}; param.Values = values; param.Modified = true; break; default: // Do not take action for other data types break; } } } // Set the value for the system-defined Response parameter peWS.ArrayOfResponse stepResponses = stepElement.Responses; if (stepResponses!=null) { string responseValue = "Ok"; stepResponses.Selected = responseValue; }

Chapter 5. Introduction to Process Engine API programming

153

Set step element parameter values: REST API scenario

The following steps are required to update step element parameter values using the PE REST API: 1. Send HTTP GET using the stepelements resource to retrieve the step element and eTag parameter. 2. Send HTTP PUT using the stepelements resource to lock the step element. 3. Set step element parameter values and response. Example 5-21 shows how to set step element parameters and responses with a Dojo toolkit client.
Example 5-21 Set step element parameters and responses with a Dojo toolkit client

// Ensure to be authenticated with the Application Server container // Set request URI var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; // Set REST API resource and work item number var queueName = "Inbox"; var wobNumber = "5BDD567B70453C48A5388AC56ABC69E4"; var url = baseURL + "queues/" + queueName + "/stepelements/" + wobNumber; var eTag; var stepElement; // Use the HTTP GET method to retrieve the step element and eTag dojo.xhrGet({ url: url, handleAs: "json-comment-optional", // The callback to handle the response load: function(data, ioArgs) { // Get the step element stepElement = data; // Get the eTag eTag = ioArgs.xhr.getResponseHeader("ETag"); }, // The error callback error: function(data) { console.dir(data); }, sync: true // Use the HTTP PUT method to lock the step element

154

Developing Applications with IBM FileNet P8 APIs

var urlForLock = url + "?action=lock&responseContent=1"; dojo.rawXhrPut({ url: urlForLock, handleAs: "text", // Set the value of "If-Match" to the value of eTag // that you got in the 1st step headers: {"Content-Type":"application/json","If-Match":eTag}, _raw: true, // The callback to handle the response load: function(data, ioArgs) { console.log(dojo.toJson(data)); console.log("Lock successfully"); }, // The error callback error: function(data) { console.dir(data); }, sync: true } ); // Set the step element parameter values for (var dataFieldName in stepElement.dataFields) { var dataField = stepElement.dataFields[dataFieldName]; if (dataField.type === 2) { dataField.value = "Value_1"; dataField.modified = true; console.log("Set the value for data field: " + dataFieldName); } } for (var attachmentName in stepElement.attachments) { console.log("Set the value for attachment: " + attachmentName); var attachment = stepElement.attachments[attachmentName]; if (attachment.isArray) { attachment.value = [{"vsId":"{997E0AAC-8068-4990-AC08-D748E3063162}", "title":"Document Title", "type":3, "libraryName":"objectstore", "libraryType":3, "version":"{D48EE0A9-61CA-4DAD-8B61-961CE362434B}", "desc":"A document added programmatically"}]; } else { attachment.value = {"vsId":"{997E0AAC-8068-4990-AC08-D748E3063162}",

Chapter 5. Introduction to Process Engine API programming

155

"title":"Document Title", "type":3, "libraryName":"objectstore", "libraryType":3, "version":"{D48EE0A9-61CA-4DAD-8B61-961CE362434B}", "desc":"A document added programmatically"}; } attachment.modified = true; } // set workflow group participant for (var workflowGroupName in stepElement.workflowGroups) { console.log("Set the value for workflow group: " + workflowGroupName); var workflowGroup = stepElement.workflowGroups[workflowGroupName]; if (workflowGroup.isArray) { workflowGroup.value = [{"Administrator"}]; } else { workflowGroup.value = {"Administrator"}; } workflowGroup.modified = true; } // Set the selected Response stepElement.systemProperties.selectedResponse = "Complete";

5.6.4 Complete work items

To complete work, a step processor or work performer application must be able to save or cancel changes that are made to the work item, and be able to advance the work item to the next step in the workflow. The following steps are included in the examples in order to complete work items using the PE Java and Web services APIs: Save changes to the work item without advancing it in the workflow. Cancel changes to the work item without advancing it in the workflow. Save changes to the work item and advance it in the workflow. For further details about included steps to complete work items by using the PE REST API, see the Complete work items: REST API scenario on page 160. Example 5-22 on page 157 and Example 5-23 on page 158 show the possible actions to complete work items by using both the Java API and a PEWS client.

156

Developing Applications with IBM FileNet P8 APIs

Example 5-22 Complete a work item using the Java API

// Create session object and log onto Process Engine ... // Retrieve Step Element and Step Parameters ... // Set Step Element Parameters ... // Action Types final int ACTION_TYPE_REASIGN = 1; final int ACTION_TYPE_RETURN = 2; final int ACTION_TYPE_ABORT = 3; final int ACTION_TYPE_SAVE= 4; final int ACTION_TYPE_DISPATCH = 5; // Complete Step Element // Action to perform on the Step Element int actionToPerform = 5; // Dispatch switch (actionToPerform) { case ACTION_TYPE_REASIGN: // Determine whether a step element // can be reassigned and reassign it if (stepElement.getCanReassign()) { String participantName = "Administrator"; stepElement.doReassign(participantName,true,null); } break; case ACTION_TYPE_RETURN: // Determine whether a step element can be returned to the // queue from which the user delegated or reassigned it and // return it if (stepElement.getCanReturnToSource()) stepElement.doReturnToSource(); break; case ACTION_TYPE_ABORT: // Cancel the changes to the work item // without advancing it in the workflow stepElement.doAbort(); break;

Chapter 5. Introduction to Process Engine API programming

157

case ACTION_TYPE_SAVE: // Save the changes to the work item // and unlock it without advancing it in the workflow stepElement.doSave(true); break; case ACTION_TYPE_DISPATCH: // Save the changes to the work item // and advance it in the workflow stepElement.doDispatch(); break; } Note: Abort actions on step elements are not available for the Process Engine Web Services API. Instead, the unlockStep operation is available for step elements (to cancel changes on a work item and unlock it).
Example 5-23 Complete a work item using PEWS // Create the PEWS RequestSoapContext header ... // Retrieve Step Element and Step Parameters ... // Set Step Element Parameters ... // Action Types sealed int ACTION_TYPE_REASIGN = 1; sealed int ACTION_TYPE_RETURN = 2; sealed int ACTION_TYPE_SAVE= 3; sealed int ACTION_TYPE_DISPATCH = 4; // Complete Step Element // Action to perform on the Step Element int actionToPerform = 4; // Dispatch switch (actionToPerform) { case ACTION_TYPE_REASIGN: if (stepElement.CanReassign) { string participantName = "Administrator"; peWS.ReassignStepRequest reassignRqt = new peWS.ReassignStepRequest (); reassignRqt.delegate = true; reassignRqt.newUserName = participantName; reassignRqt.stepElement = stepElement;

158

Developing Applications with IBM FileNet P8 APIs

peWSServicePort.reassignStepRequest(reassignRqt); } break; case ACTION_TYPE_RETURN: // Determine whether a step element can be returned to the // queue from which the user delegated or reassigned it and // return it if (stepElement.CanReturnToSource) { peWS.UpdateStepRequest updStepRequest = new peWS.UpdateStepRequest(); peWS.UpdateFlagEnum updFlagEnum = peWS.UpdateFlagEnum.UPDATE_SAVE_RETURN; updStepRequest.stepElement = stepElement; updStepRequest.updateFlag = updFlagEnum; peWSServicePort.updateStep(updStepRequest); } break; case ACTION_TYPE_SAVE: // Save the changes to the work item // and unlock it without advancing it in the workflow peWS.UpdateStepRequest updStepRequest = new peWS.UpdateStepRequest(); peWS.UpdateFlagEnum updFlagEnum = peWS.UpdateFlagEnum.UPDATE_SAVE_UNLOCK; updStepRequest.stepElement = stepElement; updStepRequest.updateFlag = updFlagEnum; peWSServicePort.updateStep(updStepRequest); break; case ACTION_TYPE_DISPATCH: // Save the changes to the work item // and advance it in the workflow peWS.UpdateStepRequest updStepRequest = new peWS.UpdateStepRequest(); peWS.UpdateFlagEnum updFlagEnum = peWS.UpdateFlagEnum.UPDATE_DISPATCH; updStepRequest.stepElement = stepElement; updStepRequest.updateFlag = updFlagEnum; peWSServicePort.updateStep(updStepRequest); break; }

Chapter 5. Introduction to Process Engine API programming

159

Complete work items: REST API scenario

To complete work items by using the PE REST API, follow these steps: 1. Send HTTP GET using the stepelements resource to retrieve the step element and eTag parameter. 2. Send HTTP PUT using the stepelements resource to lock the step element. 3. Set step element parameter values and response. 4. Send HTTP PUT using the stepelements resource and the action parameter. Table 5-11 presents the possible values that the action parameter can contain to perform an action on a step element. This parameter is required for all HTTP PUT requests by using the stepelements resource.
Table 5-11 Action parameter values Action value lock overrideLock save Description Locks the work item. Locks the work item. If the work item is already locked, the current lock will be overridden. Updates the work item and leaves it locked. In the request content, sends the updated version of the representation that is returned in the GET response. Updates the work item and unlocks it. In the request content, sends the updated version of the representation that returned in the GET response. Unlocks the work item without making any updates. Updates the work item, unlocks it, and advances the work item in the workflow. In the request content, sends the updated version of the representation returned in the GET response.

saveAndUnlock

abort dispatch

Example 5-24 shows how to complete a work item with a Dojo toolkit client.
Example 5-24 Complete a work item with a Dojo toolkit client

// Ensure to be authenticated with the Application Server container // Set request URI var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; // Set REST API resource and work item number var queueName = "Inbox"; var wobNumber = "5BDD567B70453C48A5388AC56ABC69E4";

160

Developing Applications with IBM FileNet P8 APIs

var url = baseURL + "queues/" + queueName + "/stepelements/" + wobNumber; // Use the HTTP GET method to retrieve the step element and eTag ... // Use the HTTP PUT method to lock the step element ... // Set step element parameter values and response ... // Complete the step element by HTTP PUT method var urlForDispatch = url + "?action=dispatch&responseContent=0"; dojo.rawXhrPut({ url: urlForDispatch, handleAs: "text", // Set the value of "If-Match" to the value of eTag //retrieved from the initial request headers: {"Content-Type":"application/json","If-Match":eTag}, _raw: true, // Set the data you want to put back putData: dojo.toJson(stepElement), // The callback to handle the response load: function(data, ioArgs) { console.log("Complete the step element successfully"); }, // The error callback error: function(data) { console.dir(data); }, sync: true } );

Chapter 5. Introduction to Process Engine API programming

161

5.7 Work with process status

Use process status elements to keep track of a running workflow. Status elements can be retrieved and monitored. Elements might be milestones reached within a running workflow, workflow steps completed, which participants are completing these steps, and other valuable information about the workflow. The main components of process status indicators are workflow history and milestones elements: Workflow milestones To track the progress of a workflow, key points (milestones) are included within a workflow definition. On each workflow map, a milestone can be placed either before or after a step. Each of these milestones is related to a message that is specified by the workflow designer. The message is written to a log file when the running workflow reaches the milestone. Workflow history To track workflow activity, workflow-related historical information is extracted from the event logs, which contain specific records related to a workflow event for each PE-isolated region. This historical information is useful to track workflow-related activities, such as how long workflows take to complete and which workflows are currently running.

5.7.1 Retrieve process history

To retrieve the information about the workflow history by using the PE Java API: 1. 2. 3. 4. 5. 6. 7. 8. 9. Retrieve the work item object. Get VWProcess object from the work object. Get the workflow definitions from the VWProcess. Get the maps for each workflow definition. Get the workflow history information for each map. Get the step history objects for each workflow history. Get the step occurrence history objects for each step history object. Get the step work object information for each step occurrence. Get the participant information for each work object.

Table 5-12 on page 163 shows the information that can be retrieved from the PE objects to track the workflow-related historical information.

162

Developing Applications with IBM FileNet P8 APIs

Table 5-12 Process history information Process Engine object VWProcess VWProcessHistory VWStepHistory VWStepOcurrenceHistory VWStepWorkObjectHistory VWParticipantHistory Information to retrieve Status, history, child, and split work object information Launch step comments, launch date, workflow originator, launch step response Step occurrence completion date, date participant received work, step status Step name Whether this process was completed Comments, responses, dates completed and received, exposed log fields, activity status, name of participant who completed, delegated, reassigned, returned, or terminated step

Example 5-25 shows how to retrieve information about the workflow history by using the Java API
Example 5-25 Retrieve workflow history information using the Java API // Create session object and log onto Process Engine ... // Get the specific work item ... // Get VWProcess object from work object VWProcess process = stepElement.fetchProcess(); // Get workflow definitions from the VWProcess VWWorkflowDefinition workflowDefinition = process.fetchWorkflowDefinition(false); // Get maps for each workflow definition VWMapDefinition[] workflowMaps = workflowDefinition.getMaps(); // Iterate through each map in the workflow Definition for (int i = 0; i < workflowMaps.length; i++) { // Get map ID and map name for each map definition int mapID = workflowMaps[i].getMapId(); String mapName = workflowMaps[i].getName(); // Get workflow history information for each map VWWorkflowHistory workflowHistory = process.fetchWorkflowHistory(mapID); String workflowOriginator = workflowHistory.getOriginator();

Chapter 5. Introduction to Process Engine API programming

163

// Iterate through each item in the Workflow History while (workflowHistory.hasNext()) { // Get step history objects for each workflow history VWStepHistory stepHistory = workflowHistory.next(); // Iterate through each item in the Step History while (stepHistory.hasNext()) { // Get step occurrence history // objects for each step history object VWStepOccurrenceHistory stepOccurenceHistory = stepHistory.next(); Date stepOcurrenceDateReceived = stepOccurenceHistory.getDateReceived(); Date stepOcurrenceDateCompleted = stepOccurenceHistory.getCompletionDate(); while (stepOccurenceHistory.hasNext()) { // Get step work object information // for each step occurrence VWStepWorkObjectHistory stepWorkObjectHistory = stepOccurenceHistory.next(); stepWorkObjectHistory.resetFetch(); // Get participant information for each work object while (stepWorkObjectHistory.hasNext()) { VWParticipantHistory participantHistory = stepWorkObjectHistory.next(); Date participantDateReceived = participantHistory.getDateReceived(); String participantComments = participantHistory.getComments(); String participantUser = participantHistory.getUserName(); String participantName = participantHistory.getParticipantName(); } // while stepWorkObjectHistory } // while stepOccurenceHistory } // while stepHistory } // while workflowHistory } // for workflow maps

164

Developing Applications with IBM FileNet P8 APIs

5.7.2 Retrieve process milestones

The following steps are included to retrieve the workflow milestones history information using the PE Java and Web services APIs: 1. 2. 3. 4. 5. 6. Retrieve the work item object. Get VWProcess object from the work object. Get all milestones definition from the VWProcess object. Get reached milestones from the VWProcess object. Get milestone element information for each milestone. Handle information for each milestone element.

Each milestone that is defined within a workflow is associated with a level (1 - 99). The workflow-reached milestones are based on the milestone level that is specified. For example, if the milestone level is set to 5, reached-milestones of levels 1 - 5 are retrieved. Example 5-26 and Example 5-27 on page 166 show how to perform the steps required to retrieve the workflow milestones history information, using both the Java API and a PEWS client.
Example 5-26 Get milestones of a workflow using the Java API

// Create session object and log onto Process Engine ... // Get the specific work item ... // Get VWProcess object from work object ... // Get all milestones definition from the workflow process VWMilestoneDefinition[] milestonesDefinition = process.getMilestoneDefinitions(); for (int i = 0; i < milestonesDefinition.length; i++) { // Milestone element information VWMilestoneDefinition milestoneDefinition = milestonesDefinition[i]; String milestoneName = milestoneDefinition.getName(); String milestoneMessage = milestoneDefinition.getMessage(); int milestoneLevel = milestoneDefinition.getLevel(); } // Get reached milestones from the VWProcess object int level = 99; VWWorkflowMilestones milestones = process.fetchReachedWorkflowMilestones(level); while (milestones.hasNext()) { // Milestone element information

Chapter 5. Introduction to Process Engine API programming

165

VWMilestoneElement milestoneElement = milestones.next(); String milestoneName = milestoneElement.getName(); String milestoneMessage = milestoneElement.getMessage(); Date milestoneTimeLogged = milestoneElement.getTimestamp(); }
Example 5-27 Get milestones of a workflow using PEWS

// Create the PEWS RequestSoapContext header ... // Get milestones work object from the roster string rosterName = "DefaultRoster"; string wobnum = "5BDD567B70453C48A5388AC56ABC69E4"; peWS.GetMilestoneForRosterElementRequest query = new peWS.GetMilestoneForRosterElementRequest (); query.milestoneLevel = 99; query.rosterName = rosterName; query.queryValue = wobnum; query.queryEnum = peWS.MilestoneQueryEnum.MILESTONE_QUERY_WOBNUMBER; peWS.Milestone [] milestones = peWSServicePort.getMilestoneForRosterElement(query); if (milestones!=null) { foreach (peWS.Milestone m in milestones) { // Milestone element information string map = m.Map; string milestoneMessage = m.Message; string milestoneName = m.Name; int milestoneId = m.Id; int milestoneLevel = m.Level; int stepId = m.StepId; DateTime milestoneDateReached = m.Reached; } }

166

Developing Applications with IBM FileNet P8 APIs

Chapter 6.

Advanced Process Engine API programming

In this chapter, we introduce advanced topics for the IBM FileNet Business Process Manager API programming. Although we refer to these topics as advanced, they are actually a continuation of the concepts we presented in Chapter 5, Introduction to Process Engine API programming on page 111. Most of the code snippets we show in this chapter are from the sample applications that we created for this book. For more details about the sample applications, refer to Chapter 7, Sample applications for Fictional Auto Rental Company A on page 223. This chapter discusses the following topics: Component Integrator Application space, role, and workbasket Resource navigation in Process Engine REST API ECM Widgets overview Building a custom Get Next In-basket widget

Copyright IBM Corp. 2009. All rights reserved.

167

6.1 Component Integrator

Component Integrator (CI) is part of the IBM FileNet P8 platform, and that connects the PE with an external entity called a component. A component can be either a Java object or Java Messaging System (JMS) queue, making them available in a workflow. CI provides a technology platform that enables organizations to integrate and coordinate their business processes. This is typically called Enterprise Information System (EIS) integration. The use of CI provides enterprise solution capabilities, such as: Extend business functionality easily without full application development. Automate work processing. Perform external functions from within a workflow. Use existing Java business objects and components. Integrate with a Java Message Service (JMS). For further details about CI general topics, including how to implement the Web Services adaptor, refer to Chapter 8: Implementing Component Integrator and Web services from Introducing IBM FileNet Business Process Manager, SG24-7509. In this section, we present the available actions provided by the CE_Operations component, and how to implement a custom Java adaptor.

6.1.1 CE_Operations component

CE_Operations is a built-in component that are used to integrate the Content Engine (CE) and Process Engine (PE), facilitating interaction with documents and other objects in an object store from within a workflow. This component is provided as part of a typical Application Engine (AE) or WorkplaceXT installation (although it does not run within the J2EE application as part of AE or WorkplaceXT). The following required steps implement CE_Operations in a workflow definition: 1. Determine the symbolic name and data type of the object class property using IBM FileNet Enterprise Manager. 2. Define the appropriate attachments and data fields in the workflow definition. These data fields hold the values retrieved from the object property in the component step. 3. Add the component step on the workflow map, select the CE_Operations component queue, and define the appropriate parameters for each operation.

168

Developing Applications with IBM FileNet P8 APIs

Available actions
Table 6-1 describes available actions that are provided by the CE_Operations component.
Table 6-1 Available CE_Operations Action File objects Unfile objects Publish document Get object property values Description File one or more objects (documents and custom objects) to a specified folder. Unfile one or more objects (documents and custom objects) from a specified folder. Publish a specified document by using a publishing template. Get single or multiple property values from a supplied document, custom object, or folder. The available methods to get property values support integer, double, date, Boolean, string, and object types. Set single or multiple property values of a supplied document, custom object, or folder. The available methods to set property values support integer, double, date, Boolean, string, and object types. Change the class of a supplied document, folder, or custom object. Delete one or more objects (documents, folders, and custom objects). Check out a specified document from the repository and check in a specified document as a new major version. Copy one or more objects (documents and custom objects) within the same repository. Create a new object (documents, folders, and custom objects) instance using a specified class and property values. For document objects, MIME type and content parameters must be specified Move one or more objects (documents and custom objects) from their existing folder to a specified folder. Get an object (custom object, folder, or document) from a specified repository using a specified path.

Set object property values

Change object class Delete object Document versioning Copy object Create object

Move object Get object

Chapter 6. Advanced Process Engine API programming

169

Action Apply security template Search object

Description Apply a security template using a specified template to one or more objects (documents, folders, and custom objects). Execute a specified search template or stored search, returning the first object or an array of objects matching the search criteria. Available methods to send mail supporting the following scenarios: Text (plain) with no attachment With documents attached Using a template with no attachment Using a template with documents attached With the URLs of the attached documents appended to the end of the body of the e-mail Using a template with the URLs of the attached documents appended to the end of the body of the e-mail

Send mail

Note: All sendMail methods require that e-mail notification be properly configured by using the Process Task Manager. For further details about how to configure e-mail notification, refer to the following P8 documentation path: ECM Help System Administrator Enterprise-wide Administration Process Task Manager Process Engine Configure Process Engine Email notification.

6.1.2 Implementing a custom Java component

Java adapters handle PE calls to Java objects, which are represented to the PE as operations on queues (work items) where each operation is performed by a method of the Java class. The Java adapter executes the interface to the Java component, then automatically waits for a response from this component, updates the work item, and dispatches this work item to the next workflow step. The following steps describe the development process to implement a custom component: 1. 2. 3. 4. 5. 6. Identify the business need for the component. Define component operations. Implement the custom component. Configure component queue(s) to service the custom adaptor operations. Iteratively test and debug the custom component. Deploy custom component to production environment.

170

Developing Applications with IBM FileNet P8 APIs

Note: For multithread-safe custom components, concurrent threads may be increased to enable parallel execution of the component on a multiprocessor system, improving the component execution performance. The number of threads for the component can be increased according the number of processors on the system. The custom Java classes serving as a gateway for the component must comply with the following conditions: A constructor that has no parameters. Only public methods are available for the component operations. Public methods exposed as component operations can contain only the following parameter types (single and array): String Date Integer Float Double Boolean VWAttachment VWParticipant

Java component association

The Java component operations are defined as a public method in a Java class (gateway) compiled to a Java archive (JAR) file. The JAR file must be deployed to a network directory that the AE can connect to; however we recommend using a directory location on the AE server. The Java adaptor associates a Java component queue with a Java archive file and Login Context as shown in Figure 6-1.

Figure 6-1 Java component association

Chapter 6. Advanced Process Engine API programming

171

ITSO_Operations custom component

ITSO_Operations is a custom Java component that is implemented to perform certain business operations, which are used by this books sample applications. This custom adaptor also includes operations not used in the sample applications but implemented to show how to perform certain actions, such as custom PE API calls. In this section, we use the ITSO_Operations component package to describe certain elements that are included, in order to implement a custom Java adaptor. These elements are: The Java business classes The Java gateway class The custom Java Authentication and Authorization Service (JAAS) class and configuration file

Implementing the Java business classes

The business entities for ITSO_Operations are implemented as Data Access Object (DAO) classes that are consumed by the gateway class. The two business entities that are implemented for this custom Java adaptor are: ITSOVehicleDAO: Business entity for vehicle-related operations ITSOCustomerDAO: Business entity for customer-related operations ITSO_Operations business entities perform custom CE operations (not available in CE_Operations) and PE operations that are required for sample application. Certain actions, such as the verification of a customers credit card number (verifyCustomerCredit) are artificial, created only as placeholders with the intention to show how certain business operations can be implemented to perform calls to external systems (for example, custom database). Example 6-1 and Example 6-2 on page 174 show specific DAO methods that are implemented for both vehicle and customer business entities.
Example 6-1 Customer business entity

/** * DAO class used as business entity for customer-related operations */ public class ITSOCustomerDAO { // Retrieve the content from a ITSOCustomer // document and parse specific meta-data public String[] getCustomerDetailsFromOS(String objectStoreName, String customerID, String[] selectedFields) throws Exception { String[] customerDetails = new String[] {};

172

Developing Applications with IBM FileNet P8 APIs

ITSOUtil itsoUtil = new ITSOUtil(); SearchScope search = new SearchScope(itsoUtil.getCEConnection().fetchOS(objectStoreName)); SearchSQL sql = new SearchSQL(); sql.setSelectList("*"); sql.setFromClauseInitialValue("ITSOCustomer", "d", true); sql.setWhereClause("ITSOCustomerUserId='"+customerID+"' AND NOT (IsClass(d, CodeModule))"); DocumentSet documents = (DocumentSet)search.fetchObjects (sql,Integer.getInteger("1"),null, Boolean.valueOf(true)); Iterator itDoc = documents.iterator(); if (itDoc.hasNext()) { Document doc = (Document) itDoc.next(); ContentElementList contents = doc.get_ContentElements(); ContentElement content; Iterator itContent = contents.iterator(); if (itContent.hasNext()) { content = (ContentElement)itContent.next(); customerDetails = parseCustomerDetailsContent ((ContentTransfer)content, selectedFields); } } return customerDetails; } // Get external database connection from the // custom LoginModule and perform call public Boolean verifyCustomerCredit(String creditCardNumber, Integer requestedAmount) throws Exception { return Boolean.TRUE; } // Additional Customer DAO Methods ... }

Chapter 6. Advanced Process Engine API programming

173

Example 6-2 Vehicle business entity

/** * DAO class used as business entity for vehicle-related operations */ public class ITSOVehicleDAO { public String[] getVehicleMaintananceQueueDetails(String vehicleID, String[] selectedFields) throws Exception { String[] vehicleDetails = new String[] {}; String maintenanceQueue = "ITSO_Maintenance"; ITSOUtil itsoUtil = new ITSOUtil(); VWQueue queue = itsoUtil.getPESession(). getQueue(maintenanceQueue); String queryFilter="ITSOVehicleId=:A"; String[] substitutionVars = {new String(vehicleID)}; int queryFlags = VWQueue.QUERY_READ_LOCKED; int fetchType = VWFetchType.FETCH_TYPE_WORKOBJECT; VWQueueQuery queueQuery = queue.createQuery(null,null,null, queryFlags,queryFilter,substitutionVars,fetchType); if (queueQuery.hasNext()) { vehicleDetails = new String[selectedFields.length]; VWWorkObject workItem = (VWWorkObject) queueQuery.next(); for (int i=0;i<selectedFields.length;i++) vehicleDetails[i] = workItem.getFieldValue (selectedFields[i]).toString(); } return vehicleDetails; } // Additional Vehicle DAO Methods ... }

Implementing the Java gateway class

The Java gateway class of a custom component is used to expose the public business methods that are used by the component queue operations.

174

Developing Applications with IBM FileNet P8 APIs

The ITSO_Operations component implements the Java gateway class ITSOOperations to expose the business methods (public methods) and define the component queue operations. The gateway class has a constructor without any parameters and the public methods only contain allowable component operation parameter types. The gateway class consumes the business entities (ITSOVehicleDAO and ITSOConsumerDAO) to perform all business-related logic. See Example 6-3.
Example 6-3 Operations gateway class: ITSOOperations.java /** * Gateway class for ITSO_Operations. */ public class ITSOOperations { public ITSOOperations() { } // Validate the customer's credit from external database system public Boolean verifyCustomerCredit(String creditCardNumber, Integer requestedAmount) throws Exception { return new ITSOCustomerDAO(). verifyCustomerCredit(creditCardNumber, requestedAmount); } // Get customer details from ITSOCustomer // document content using ad-hoc query public String[] getCustomerDetailsFromOS(String objectStoreName, String customerID, String[] selectedFields) throws Exception { return new ITSOCustomerDAO(). getCustomerDetailsFromOS(objectStoreName, customerID, selectedFields); } // Get vehicle maintenance details from the // Process Engine maintenance queue public String[] getVehicleMaintDetails(String vehicleID, String[] selectedFields) throws Exception { return new ITSOVehicleDAO(). getVehicleMaintananceQueueDetails(vehicleID, selectedFields); } // Additional Business Methods ... }

Chapter 6. Advanced Process Engine API programming

175

Implementing the custom login context

A login context or configuration context in a custom component includes the login modules that provide the required authentication for all Java components. For example, a Java component must authenticate with a database for its operations, therefore a custom login module that is handling the database authentication must be defined for the login context of the component. These login contexts can be shared between many Java components. The login context for Component Manager is defined within a Java Authentication and Authorization Service (JAAS) configuration file. A JAAS configuration file is provided (taskman.login.config) including the default login contexts provided by IBM FileNet, as shown in Example 6-4.
Example 6-4 Default login contexts provided in the taskman.login.config file CELogin { filenet.vw.server.VWLoginModule required routerurl=""; com.filenet.wcm.toolkit.server.operations.util.CELoginModule required credTag=Clear; filenet.contentops.ceoperations.util.CELoginModule required credTag=Clear; }; PEWSLogin { com.filenet.wcm.toolkit.server.operations.util.CELoginModule required credTag=Clear; }; FileNetP8 { com.filenet.api.util.WSILoginModule required debug=false; };

The FileNetP8 stanza must be present for any JAAS configuration file containing the appropriate login module to connect to CE, depending on the CE transport protocol. By default, when running outside of an application server, the Content Engine Web Services (CEWS) transport protocol is recommended (com.filenet.api.util.WSILoginModule) because it is typically easier to configure. EJB transport can have performance advantages. The CELogin stanza is used by the CE_Operations component. This login context includes the following login module that provides authentication to CE: com.filenet.wcm.toolkit.server.operations.util.CELoginModule

176

Developing Applications with IBM FileNet P8 APIs

This login module is used in conjunction with the CE 3.x Java Compatibility Layer; therefore, a com.filenet.wcm.api.Session object is initiated using this module. At the minimum, a login context that is used for custom Java components must contain the filenet.vw.server.VWLoginModule module, which authenticates to a PE Isolated Region. This approach means that the provided user information should exist for both the directory server used by the IBM FileNet P8 platform and the custom component. Because ITSO_Operations performs custom CE actions that are not available within CE_Operations, the custom ITSOLogin login context is implemented for the component. The following steps describe how the custom login context is implemented for the ITSO_Operations component: 1. 2. 3. 4. 5. Create ITSOPrincipal class. Create ITSOLoginModule class. Add ITSOLogin login context to taskman.login.config file. Create ITSOUtil class to retrieve the ITSOPrincipal object. Deploy ITSO_Operations component using ITSOLogin context.

ITSOPrincipal implements the java.security.Principal interface, which is used by ITSOLoginModule class. See Example 6-5.
Example 6-5 ITSO_Operations principal class: ITSOPrincipal.java

/** * Principal class for ITSO_Operations */ public class ITSOPrincipal implements Principal, Serializable { private String m_username; private ITSOCEConnection m_ceconnection; public ITSOPrincipal( ITSOCEConnection m_ceconnection, String m_username ) { if( m_username == null ) throw new NullPointerException( "illegal null input" ); this.m_ceconnection = m_ceconnection; this.m_username = m_username; } public ITSOCEConnection getCEConnection()

Chapter 6. Advanced Process Engine API programming

177

{ return this.m_ceconnection; } public boolean equals( Object o ) { if( o == null ) return false; if( this == o ) return true; if( !(o instanceof ITSOPrincipal) ) return false; ITSOPrincipal that = (ITSOPrincipal) o; if( this.getName().equals( that.getName() ) ) return true; return false; } // Additional ITSOPrincipal methods ... } ITSOLoginModule implements the javax.security.auth.spi.LoginModule interface including the custom CE authentication and passing the connection object to ITSOPrincipal for further handling. See Example 6-6.
Example 6-6 ITSO_Operations login module class: ITSOLoginModule.java

/** * Custom Login Module class for ITSO_Operations. */ public class ITSOLoginModule implements LoginModule { // ITSO Principal private ITSOPrincipal m_principal; public void initialize(Subject subject, CallbackHandler callbackHandler, Map sharedState, Map options) { this.m_subject = subject; this.m_callbackHandler = callbackHandler;; this.m_options = options;

178

Developing Applications with IBM FileNet P8 APIs

this.m_sharedState }

= sharedState;

public boolean login() throws LoginException { if (m_callbackHandler == null) { throw new LoginException("Error: no CallbackHandler available to get authentication information"); } String password = null; m_username = (String)m_sharedState.get ("javax.security.auth.login.name"); password = (String)m_sharedState.get ("javax.security.auth.login.password"); if (m_username == null) { Callback[] callbacks = null; callbacks = new Callback[3]; callbacks[0] = new TextOutputCallback(TextOutputCallback. INFORMATION, "Custom Authentication"); callbacks[1] = new NameCallback("user name:"); callbacks[2] = new PasswordCallback("password:", false); try { m_callbackHandler.handle(callbacks); m_username = ((NameCallback)callbacks[1]).getName(); char[] tmpPassword = ((PasswordCallback)callbacks[2]). getPassword(); if (tmpPassword != null) password = new String(tmpPassword); else password = null; ((PasswordCallback)callbacks[2]).clearPassword(); m_sharedState.put("javax.security.auth.login.name", m_username); m_sharedState.put("javax.security.auth.login.password", password); }

Chapter 6. Advanced Process Engine API programming

179

catch (IOException ioe) { throw new LoginException(ioe.toString()); } catch (UnsupportedCallbackException uce) { throw new LoginException("Error: no CallbackHandler available to get authentication information"); } } try { /** * Perform custom CE authentication */ String ceURI = System.getProperty ("filenet.pe.bootstrap.ceuri"); m_ceconnection = new ITSOCEConnection(m_username, password, "FileNetP8WSI", ceURI); } catch (Exception e) { m_ceconnection = null; throw new LoginException(e.toString()); } m_validSession = true; return (m_validSession); } public boolean commit() throws LoginException { if(m_ceconnection == null) return false; m_principal = new ITSOPrincipal(m_ceconnection, m_username); if(!m_subject.getPrivateCredentials().contains(m_principal)) m_subject.getPrivateCredentials().add(m_principal); return true; } // Additional ITSOLoginModule methods ... }

180

Developing Applications with IBM FileNet P8 APIs

ITSOLogin is the custom login context appended to taskman.login.config file that is used by the ITSO_Operations component. See Example 6-7.
Example 6-7 JAAS custom login context implemented for ITSO_Operations

ITSOLogin { filenet.vw.server.VWLoginModule required; itso.autorental.components.authentication.ITSOLoginModule required debug=true; }; The ITSOUtil class is a custom class that is implemented to retrieve the ITSOPrincipal, including the custom CE authenticated object. A method to retrieve the PE session object (VWSession) is also implemented for custom PE-related operations. See Example 6-8.
Example 6-8 ITSO_Operations class to retrieve ITSOPrincipal: ITSOUtil.java

/** * Utility class for ITSO_Operations. */ public class ITSOUtil { private VWSession pesession = null; public VWSession getPESession() throws Exception { if(this.pesession == null || !(this.pesession.isLoggedOn())) { Subject subject = Subject.getSubject(AccessController.getContext()); Set creds = subject.getPrivateCredentials (Class.forName("filenet.vw.api.VWSession")); Iterator i = null; if (creds != null) i = creds.iterator(); if (i != null) this.pesession = (VWSession) i.next(); } return this.pesession; } public ITSOCEConnection getCEConnection() throws Exception { Subject subject Subject.getSubject (AccessController.getContext()); Set set = null; try {

Chapter 6. Advanced Process Engine API programming

181

set = subject.getPrivateCredentials(Class.forName ("itso.autorental.components.authentication.ITSOPrincipal")); } catch(Exception exception) { } if(set == null) throw new Exception("Failed to find principal"); Iterator iterator = set.iterator(); if(iterator == null || !iterator.hasNext()) throw new Exception("Failed to find principal"); Principal principal = (Principal)iterator.next(); if(!(principal instanceof ITSOPrincipal)) throw new Exception("Failed to find principal"); else return ((ITSOPrincipal)principal).getCEConnection(); } }

Configuring ITSO_Operations using Eclipse

Usually, the component classes and resources must be packaged into a Java archive (JAR) file to be referenced by the Process Configuration Console applet when the component queue is created. This manual step can be time-consuming and error-prone. Prior to configuring and testing the component queue by using Eclipse, see Chapter 2, Setting up development environments on page 13 to set up an Eclipse project for P8 development. The programmatic creation of the component queue is possible with the tool filenet.vw.integrator.base.PEComponentQueueHelper, which is part of the libraries that are required by PE. Table 6-2 lists the parameters for the tool.
Table 6-2 Parameters for filenet.vw.integrator.base.PEComponentQueueHelper Parameter /PEuser PEUser /PEpw PEPw /PErouter PEConnectionPointName /queue componentQueueName /JAAS JAASContext /user JAASUsername Description PE user name PE password Connection point name Name of the component queue to be created JAAS login context name JAAS user name

182

Developing Applications with IBM FileNet P8 APIs

Parameter /pw JAASPassword /class componentClassName /methods name1,name2,...

Description JAAS password Name of the component queue gateway class (Optional) Names of the methods of the classes to be imported as the component queue operations. If not specified, all public methods of the class are imported as operations for the queue

To use filenet.vw.integrator.base.PEComponentQueueHelper tool in Eclipse, perform the following steps (see Figure 6-2): 1. Create a run configuration for the tool. 2. Enter PEComponentQueueHelper for name and enter filenet.vw.integrator.base.PEComponentQueueHelper for the main class.

Figure 6-2 Run configuration for PEComponentQueueHelper

3. On the Arguments tab, set the following values by using the information in Table 6-3. Replace the parameter values with that of your environment. For an example, see Figure 6-3 on page 184.
Table 6-3 Parameters for PEComponentQueueHelper Program arguments VM arguments /PEuser administrator /PEpw filenet /PErouter CEPoint /queue ITSO_Operations /user administrator /pw filenet /JAAS ITSOLogin /class itso.autorental.components.ITSOOperations -Djava.security.auth.login.config=${workspace_loc}/P8Librarie s/CE_API/config/jaas.conf.WSI -Dwasp.location=${workspace_loc}/P8Libraries/CE_API/wsi -Dfilenet.pe.bootstrap.ceuri=http://CEServer:CEPort/wsi/FNCEW S40DIME/

Chapter 6. Advanced Process Engine API programming

183

Figure 6-3 Example for PEComponentQueueHelper parameters in Eclipse

4. Run PEComponentQueueHelper and use the Process Configuration Console to verify that the component queue is created with the specified component name (ITSO_Operations), JAAS information, and operations.

Testing and debugging the component using Eclipse

Component Manager runs as an independent Java application on the AE. It is a PE API application polling component queues for work. It can be set up to just process specific queues instead of all component queues. At initialization, Component Manager instantiates an object of the class for the custom component and sets up the necessary JAAS framework that fits the component needs for accessing its own resources, such as a Java Database Connectivity (JDBC) connection. When a work object arrives in the queue, the Component Manager invokes the corresponding method of the class on the class object. Multi-threaded custom components run in multiple threads, where each thread has its own class object. Debugging a custom component really means debugging Component Manager. An instance of Component Manager can be run in Eclipse with the command line and using filenet.vw.integrator.base.VWComponentManager class, which is part of the required PE libraries. Table 6-4 on page 185 lists Component Manager parameters.

184

Developing Applications with IBM FileNet P8 APIs

Table 6-4 Component Manager parameters description Parameter /named /routerURL=PEConnectio nPointName /username=PEUsername /password=PEPassword /registryPort=32771 Description Specifies the component manager name. Parameter is optional. Specifies the connection point name to access the PE. Specifies PE user name. Specifies PE password. Specifies the RMI registry port to register the Component Manager instance. The default value is 32771. This parameter allows the AE Process Task Manager to manage the instance The Component Manager can be set up to receive notification from the PE server when work arrives at any component queues. This approach avoids excessive polling. However, only one instance can be set up to receive notification. The default value is 32773 Specifies the Component Manager RMI server name; it must be unique within the same RMI registry. When AE Process Task Manager starts, by default, the parameter is: FileNet.VW.VWComponentManager.<connectionPointNam e>.<queueName> For example:

/eventPort=32773

/registryName=<Compone nt Manager RMI registry Name>

FileNet.VW.VWComponentManager.hqruby_600.CE_ Operations
/queues=queueName /unbind Specifies all the queues (using a value of *), a specific queue, or a list of queues separated by comma. When present, this stops the current instance that has the RMI registry object with the specified /registryName

To set up Eclipse to run Component Manager, which allows for an easy way to debug the component, perform the following steps: 1. Create a run configuration for Component Manager. 2. Enter Component Manager START for name and enter filenet.vw.integrator.base.VWComponentManager for the main class. See Figure 6-4 on page 186.

Chapter 6. Advanced Process Engine API programming

185

Figure 6-4 Run configuration for Component Manager START

3. On the Arguments tab, set the values by using the information in Table 6-5. Replace the parameter values with that of your environment. For an example, see Figure 6-5.
Table 6-5 Parameter for Component Manager START Program arguments /named /routerURL CEPoint /userName administrator /password filenet /registryPort 32771 /eventPort 32773 /registryName FileNet.VW.VWComponentManager.ITSO_Operations /queues=ITSO_Operations -Djava.security.auth.login.config=${workspace_loc}/config/t askman.login.config -Dwasp.location=${workspace_loc}/P8Libraries/CE_API/wsi -Dfilenet.pe.bootstrap.ceuri=http://CEServer:CEPort/wsi/FNC EWS40DIME/

VM arguments

Figure 6-5 Example for Component Manager START parameters in Eclipse

186

Developing Applications with IBM FileNet P8 APIs

4. Before running the Component Manager, ensure that all other Component Manager instances for the same PE connection point have been stopped. Otherwise, work in the queue is picked up and processed by those instances. Create a run configuration to stop Component Manager by duplicating the Component Manager START run configuration, and name it Component Manager STOP. Add /unbind to the program arguments as shown in Figure 6-6.

Figure 6-6 Example for Component Manager STOP parameters in Eclipse

Run the Component Manager STOP configuration to stop the running Component Manager instance. 5. Run Component Manager START in debug mode to start debugging and testing the component (ITSO_Operations) operations. Make sure to add a breakpoint to the intended component section (or sections) to be debugged.

6.2 Application space, role, and workbasket

In Process Engine (PE) 4.5.0, several concepts were introduced to the IBM FileNet P8 platform. You can leverage them to build a custom application that uses a workflow. An application space is a way of organizing the resources for a custom application. Use either Process Designer or Process Configuration Console to specify an application space to contain the roles of users who will perform the activities in the application. A role includes the access to workbaskets appropriate for the role. You can define one or more workbaskets for each role. A workbasket, also known as an In-basket, filters the work items based on your query specification defined in PE. Its definition also includes the workflow fields that will be displayed, and the custom filters that filter the work items at run time

Chapter 6. Advanced Process Engine API programming

187

In this section, we present information for using PE Java API and REST API to access the application space, role, and workbasket.

6.2.1 Retrieve role list

After user logon to PE, a list of roles will be associated to the current user. The following steps are included to retrieve a role list using the PE Java API: 1. Create a VWSession object and log on to PE. 2. Retrieve roles from the VWSession object by providing the application space name.

Using Java API

Example 6-9 shows getting role listing by using PE Java API.
Example 6-9 Get role list by using PE Java API

// Create session object and log onto Process Engine ... String appSpaceName = "GenericApproval"; // Fetch the roles from session using the application space name VWRole roles[] = session.fetchMyRoles(appSpaceName); for(int i = 0; i < roles.length; i++) { System.out.println("Role Name:" + roles[i].getName()); System.out.println("AuthoredName:" + roles[i].getAuthoredName()); }

PE REST API scenario

The following steps are required to retrieve role list for the current user: 1. Send HTTP GET using the myroles resource to retrieve the role list. 2. Process the information of each role. Example 6-10 shows how to retrieve role list with a Dojo toolkit client.
Example 6-10 Retrieve role list with a Dojo toolkit client

// Ensure to be authenticated with the Application Server container var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; var applicationSpaceName = "GenericApproval"; // Construct the URI for roles var url = baseURL + "appspaces/" + applicationSpaceName + "/myroles"; // Use GET method to retrieve the roles

188

Developing Applications with IBM FileNet P8 APIs

dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { // Callback to handle the data for(var roleName in data) { var role = data[roleName]; console.log("AuthoredName:" + roleName); console.log("Display Name:" + role.name); console.log("URI:" + role.URI); } }, error: function(data) { // The error callback console.dir(data); } } );

6.2.2 Retrieve role description and attributes

Using PE Java API, you can directly retrieve a role object by specifying the application space name and role name, then you will be able to process the workbasket that belongs to the current role. The following steps are included to retrieve role object: 1. Create a VWSession object and log on to PE. 2. Fetch VWRole object from the VWSession object by providing the role name and application space name. 3. Process each workbasket. 4. Process the role attributes.

Using Java API

Example 6-11 on page 190 shows how to retrieve role description using Java API.

Chapter 6. Advanced Process Engine API programming

189

Example 6-11 Retrieve role description by using Java API

// Create session object and log onto Process Engine ... VWRole role; String roleName = "Approver"; String appSpaceName = "GenericApproval"; role = session.fetchMyRole(roleName, appSpaceName); System.out.println("name:" + role.getName()); System.out.println("authoredName:" + role.getAuthoredName()); System.out.println("description:" + role.getDescription()); System.out.println("homePage:" + role.getHomePageURL()); filenet.vw.api.VWRole.WorkBasketReference wbs[] = role.getWorkBasketReferences(); // Process the workbasketsif(wbs != null) { for(int i = 0; i < wbs.length; i++) { System.out.println("Workbasket name:" + wbs[i].getWorkBasketName()); System.out.println("Queue name:" + wbs[i].getWorkBasketName()); System.out.println("Workbasket authoredName:" + wbs[i].getWorkBasketAuthoredName()); } } VWAttributeInfo info = role.getAttributeInfo(); String names[] = info.getAttributeNames(); for(int i = 0; i < names.length; i++) { String fieldName = names[i]; Integer filedType = info.getFieldType(fieldName); Boolean isArray = new Boolean(info.isFieldArray(fieldName)); Object object = info.getFieldValue(fieldName); // Print out the role attribute value }

REST API scenario

The following steps are required to retrieve a role object: 1. Send HTTP GET using the role resource to retrieve the role. 2. Process the information for the role.

190

Developing Applications with IBM FileNet P8 APIs

Example 6-12 shows how to retrieve role with a Dojo toolkit client.
Example 6-12 Retrieve role with a Dojo toolkit client

// Ensure to be authenticated with the Application Server container var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; var applicationSpaceName = "GenericApprovalGenericApproval"; var roleName = "Approver"; // Construct the URI for role var url = baseURL + "appspaces/" + applicationSpaceName + "/roles/" + roleName; // Use GET method to retrieve the roles dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { var role = data; console.log("HomePage:" + role.homePage); console.log("Description: " + role.description); console.log("Display Name: " + role.name); console.log("AuthoredName: " + role.authoredName); console.log("Attributes: " + role.attributes); var workbaskets = role.workbaskets; // Process the workbaskets for (var workbasketName in workbaskets) { var workbasket = workbaskets[workbasketName]; console.log("Workbasket Display Name:" + workbasket.name); console.log("Workbasket URI:" + workbasket.URI); console.log("Workbasket AuthoredName:" + workbasketName); } }, error: function(data) { // The error callback console.dir(data); } } );

Chapter 6. Advanced Process Engine API programming

191

6.2.3 Retrieve workbasket

Using PE Java API you can retrieve a workbasket object from a Queue object. The following steps are included to retrieve workbasket object: 1. Create a VWSession object and log on to PE. 2. Fetch VWQueue object from the VWSession object by providing the Queue name. 3. Fetch VWWorkBasket object from VWQueue object by providing the workbasket name.

Using Java API

Example 6-13 shows how to retrieve workbasket using Java API.
Example 6-13 Retrieve workbasket using Java API

// Create session object and log onto Process Engine ... String queueName = "GenericApprovalQueue"; String workbasketName = "Approve"; VWWorkBasket workbasket; VWQueue queue = session.getQueue(queueName); workbasket = queue.fetchWorkBasket(workbasketName); System.out.println("Workbasket name:" + workbasket.getName()); System.out.println("Workbasket authoredName:" + workbasket.getAuthoredName()); System.out.println("Workbasket description:" + workbasket.getDescription()); // retrieve the columns filenet.vw.api.VWWorkBasket.Column columns[] = workbasket.getColumns(); if(columns != null) { for(int i = 0; i < columns.length; i++) { String name = columns[i].getName(); String authoredName = columns[i].getAuthoredName(); String prompt = columns[i].getPrompt(); int type = columns[i].getType(); boolean isSortable = columns[i].isSortable(); } } // Retrieve the filers filenet.vw.api.VWWorkBasket.Filter filters[] = workbasket.getFilters();

192

Developing Applications with IBM FileNet P8 APIs

if(filters != null) { for(int i = 0; i < filters.length; i++) { String filterName = filters[i].getName(); String authoredFilterName = filters[i].getAuthoredName(); String description = filters[i].getDescription(); String prompt = filters[i].getPrompt(); int type = filters[i].getType(); int operator = filters[i].getOperator(); } }

REST API scenario

The following steps are required to retrieve Workbasket 1. Send HTTP GET using the Workbasket resource. 2. Process the information of the Workbasket. Example 6-14 shows how to retrieve workbasket using REST API.
Example 6-14 Retrieve workbasket using a Dojo toolkit client

// Ensure to be authenticated with the Application Server container var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; var queueName = "GenericApprovalQueue"; var workbasketName = "Approve"; var url = baseURL + "queues/" + queueName + "/workbaskets/" + workbasketName; // Use GET method to retrieve dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { var workbasket = data; var columns = workbasket.columns; var filters = workbasket.filters; var description = workbasket.description; var displayName = workbasket.name; var authoredName = workbasket.authoredName; // Print the columns for (var columnName in columns) { var column = columns[columnName]; console.log("Column DisplayName:" + column.name); console.log("Sortable:" + column.sortable);

Chapter 6. Advanced Process Engine API programming

193

console.log("Type:" + column.type); console.log("Prompt:" + column.prompt); console.log("Ordinal:" + column.ordinal); } // Print the filters for (var filterName in filters) { var filter = filters[filterName]; console.log("Filter DisplayName:" + filter.name); console.log("Type:" + filter.type); console.log("Description:" + filter.description); console.log("Prompt:" + filter.prompt); console.log("Operator:" + filter.operator); console.log("Attributes URL:" + filter.attributes); } }, sync: true, error: function(data) { // The error callback console.dir(data); } } );

6.2.4 Query work items from workBasket

Using PE Java API, you can search work items from a Workbasket object. The work items can be divided into pages by specifying the pageSize query parameter. The following steps are included to retrieve workbasket object: 1. Create a VWSession object and log on to PE. 2. Fetch VWQueue object from the VWSession object by providing the Queue name. 3. Fetch VWWorkBasket object from VWQueue object by providing the Workbasket name. 4. Use fetchNextBatch method and specify the query condition (such as pageSize, filter) to query first page of the work items. 5. Use fetchNextBatch method and the lastRecord object from the returned QueryResult object to query the next and following pages.

194

Developing Applications with IBM FileNet P8 APIs

Using Java API

Example 6-15 shows how to query work items from workbasket using Java API.
Example 6-15 Query work Items from workbasket using Java API

// Create session object and log onto Process Engine ... String queueName = "GenericApproval"; String workbasketName = "Index"; VWWorkBasket workbasket; // the filter condition String[] filterNames = {"CompanyName"}; String[] subsVars = {"XYZ"}; // Get the Queue VWQueue queue = session.getQueue(queueName); // Get the Workbasket workbasket = queue.fetchWorkBasket(workbasketName); int pageSize = 4; System.out.println("Workbasket name:" + workbasket.getName()); System.out.println("Workbasket authoredName:" + workbasket.getAuthoredName()); System.out.println("Workbasket description:" + workbasket.getDescription()); // Query the first page filenet.vw.api.VWWorkBasket.QueryResults results = workbasket .fetchNextBatch( VWQueue.QUERY_READ_LOCKED, // queryFlags pageSize, // buffer_size null, // sort column filterNames, // filter names subsVars, // subsitution vars null, // last record 3 // fetch type, 3 means Queue element ); Object queueElements[] = null; String lastRecord = null; int pageNumber = 1; while (results != null) {

Chapter 6. Advanced Process Engine API programming

195

// save the last record to a local variable // it will be used to query the next page lastRecord = results.getLastRecord(); queueElements = results.getWorkObjects(); System.out.println("Process page: " + pageNumber); for (int i = 0; i < queueElements.length; i++) { VWQueueElement qe = (VWQueueElement) queueElements[i]; System.out.println("WobNumber:" + qe.getWorkObjectNumber()); System.out.println("QueueName:" + qe.getQueueName()); System.out.println("StepName:" + qe.getStepName()); filenet.vw.api.VWWorkBasket.Column columns[] = workbasket.getColumns(); if (columns != null) { for (int j = 0; j < columns.length; j++) { String name = columns[j].getAuthoredName(); System.out.println("name:" + name); Object value = qe.getDataField(name).getValue(); // Print out the value System.out.println("Value:" + value.toString()); } } } results = workbasket.fetchNextBatch( VWQueue.QUERY_READ_LOCKED, // pageSize, // null, // filterNames, // subsVars, // lastRecord, // 3 // element ); System.out.println("Next page"); pageNumber++; }

queryFlags buffer_size sort column filter names subsitution vars last record fetch type, 3 means Queue

196

Developing Applications with IBM FileNet P8 APIs

REST API scenario

The following steps are required to retrieve queue elements from workbasket: 1. Send HTTP POST using the queue elements resource to get the first page of queue element. 2. Process the returned results for the first page. 3. From the returned results for the first page, get the lastRecord object and save it to a local variable. 4. Use the saved lastRecord object to construct a queryString object and put it on the POST body and send the request to get the next page. 5. Process the returned result of the next page. 6. Continue to send POST requests until the returned result is empty. Example 6-16 shows how to query queue elements from workbasket using REST API.
Example 6-16 Query queue elements from workbasket using Dojo toolkit
var var var var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; queueName = "GenericApproval"; workbasketName = "Index"; pageSize = 4;

var url = baseURL + "queues/" + queueName + "/workbaskets/" + workbasketName + "/queueelements"; // Helper method to print the queue elements var printQueueElements = function(queueElements) { dojo.forEach( queueElements, function(queueElement) { console.log("workObjectNumber:" + queueElement.workObjectNumber); console.log("queueName:" + queueElement.queueName); console.log("lockedById:" + queueElement.lockedById); console.log("lockedBy:" + queueElement.lockedBy); console.log("ETag:" + queueElement.ETag); console.log("stepElementURL:" + queueElement.stepElement); for (var columnName in queueElement.columns) { var columnValue = queueElement.columns[columnName]; console.log("columnName:" + columnName); console.log("columnValue:" + columnValue); } }, this); }; // Construct the query parameters var queryData = "pageSize=" + pageSize + "&queryFlags=1";

Chapter 6. Advanced Process Engine API programming

197

var lastRecord; var queueElements; var pageNumber = 1; // Query the first page dojo.rawXhrPost({ url: url, handleAs: "json-comment-optional", load: function(data) { lastRecord = data.lastRecord; queueElements = data.queueElements; printQueueElements(queueElements); pageNumber++; }, _raw: true, sync: true, postData: queryData, error: function(data) { // The error callback console.dir(data); } } ); // Query the next page while(queueElements && queueElements.length > 0) { // Construct the query parameters // the lastRecord will be used to fetch the next batch result queryData = "pageSize=" + pageSize + "&queryFlags=1&lastRecord=" + lastRecord; dojo.rawXhrPost({ url: url, handleAs: "json-comment-optional", load: function(data) { lastRecord = data.lastRecord; queueElements = data.queueElements; printQueueElements(queueElements); pageNumber++; }, _raw: true, sync: true, postData: queryData, error: function(data) { // The error callback console.dir(data); } } ); }

198

Developing Applications with IBM FileNet P8 APIs

6.3 Resource navigation in Process Engine REST API

When navigating between REST resources, a best practice is to follow the URIs found in the response instead of manually constructing the URIs. Generally, the client application should have no prior knowledge beyond the initial URI of the REST API.

Resource navigation example

In Example 6-17, we start from Myroles resource, which returns a list of role resource URIs. We choose the first role resource URI and use it to retrieve the detailed information. The returned role information contains the URIs of the attached workbasket resource. Similarly, we follow the first workbasket resource URI to retrieve the detailed workbasket information.
Example 6-17 Resource navigation example using REST API

var baseURL = "http://localhost:9080/WorkplaceXT/P8BPMREST/p8/bpm/v1/"; var applicationSpaceName = "GenericApproval"; // Construct the URI for the initial myroles resource var url = baseURL + "appspaces/" + applicationSpaceName + "/myroles"; var roleURI = null; var workbasketURI = null; // Use GET method to retrieve myroles, // Save the first role's resource URI to local variable dojo.xhrGet({ url: url, handleAs: "json-comment-optional", load: function(data) { // Callback to handle the data for(var roleName in data) { var role = data[roleName]; console.log("AuthoredName:" + roleName); console.log("Display Name:" + role.name); console.log("URI:" + role.URI); // construct the complete role URI roleURI = baseURL + role.URI; break; } }, sync: true, error: function(data) {

Chapter 6. Advanced Process Engine API programming

199

// The error callback console.dir(data); } } ); if (roleURI === null) { console.log("roleURI is empty."); return; } // Use the saved URI and GET method to retrieve role resource // Save the URI of the first Workbasket to a local variable dojo.xhrGet({ url: roleURI, handleAs: "json-comment-optional", load: function(data) { var role = data; console.log("HomePage:" + role.homePage); console.log("Description: " + role.description); console.log("Display Name: " + role.name); console.log("AuthoredName: " + role.authoredName); console.log("Attributes: " + role.attributes); var workbaskets = role.workbaskets; for (var workbasketName in workbaskets) { var workbasket = workbaskets[workbasketName]; console.log("Workbasket Display Name:" + workbasket.name); console.log("Workbasket URI:" + workbasket.URI); console.log("Workbasket AuthoredName:" + workbasketName); // construct the complete workbasket URI workbasketURI = baseURL + workbasket.URI; break; } }, sync: true, error: function(data) { // The error callback console.dir(data); } } ); if (workbasketURI === null) {

200

Developing Applications with IBM FileNet P8 APIs

console.log("workbasketURI is empty."); return; } // Use GET method to retrieve the information of the workbasket dojo.xhrGet({ url: workbasketURI, handleAs: "json-comment-optional", load: function(data) { var workbasket = data; var columns = workbasket.columns; var filters = workbasket.filters; var description = workbasket.description; var displayName = workbasket.name; var authoredName = workbasket.authoredName; // print out the columns and filters // ... }, sync: true, error: function(data) { // The error callback console.dir(data); } } );

6.4 ECM Widgets overview

ECM Widgets is a mashup-based framework aiming to build business process management (BPM) solutions rapidly and efficiently.
ECM Widgets is ideal for both the business and IT user. It provides a set of UI components, allowing the business user to quickly build applications just by drag and drop, without programming. It also allows the IT user to create custom widgets that can be integrated with the existing ECM Widgets. In this book, we focus on the second scenario. In this section, we first give you an introduction to ECM Widgets. After that, we show how to create a custom widget and integrate it with other ECM Widgets.

Chapter 6. Advanced Process Engine API programming

201

6.4.1 ECM Widgets concepts

Before working with ECM Widget, we first introduce the concepts of widget, WebSphere Business Space, In-basket, events, and pages. In addition, we introduce types of widgets, including In-basket widget, Step Completion widget, Attachment widget, Viewer widget, and Work Data widget.

Widget
ECM Widgets is a collection of widgets to assemble user interfaces for IBM enterprise content management and business process management applications. In the context of ECM Widgets, a widget means a user interface component that follows the iWidget specification. For information about the iWidget specification, see the IBM Mashup Center wiki: http://www.lotus.com/ldd/mashupswiki.nsf/dx/widget-programming-guide

WebSphere Business Space

WebSphere Business Space is a mashup container that hosts the ECM Widgets and other widgets. It enables the designer to drag and drop widgets onto a mashup page, lay out the widgets, set the configuration of the widgets, and wire the widgets together to define communications between the widgets through events.

In-basket
In the context of ECM Widgets, In-basket has the same meaning as the workbasket that is defined in PE.

Events
All widgets that follow the iWidget specification communicate with each other through an event mechanism. The two types of events are: published event and handled event. For example, when a user clicks a button in Widget A, an event is published to another Widget B. To enable Widget B to subscribe to events sent from Widget A, you must wire Widget A with Widget B. This process is called wiring widgets. WebSphere Business Space has a graphical user interface (GUI) that allows you to wire the widgets.

Pages
The three types of mashup pages for ECM Widgets are: My Work page, Step Processor Page, and Single Page application. A My Work page is a top-level page that displays one or more In-baskets for selecting work items. Figure 6-7 on page 203 shows an example of My Work page.

202

Developing Applications with IBM FileNet P8 APIs

Figure 6-7 My Work page

A Step Processor Page is used to process a work item selected from an In-basket. A step processor page exists for each step in a workflow. Figure 6-8 shows an example.

Figure 6-8 Step Processor Page

Chapter 6. Advanced Process Engine API programming

203

ECM Widgets also support a Single Page application. This means that all widgets are put into one mashup page, and no page-switching exists when a user opens a work item. Figure 6-9 shows an example.

Figure 6-9 Single Page application

In-basket widget
The In-basket widget displays a list of In-baskets that belong to the current role. When you open an In-basket, it displays a list of work items according to the configuration defined in PE. When you double click a work item from the In-basket widget, it triggers an event as a notification that the user wants to open the work item. The In-basket widget can be placed on the My Work page or a Single Page application.

Step Completion widget

The Step Completion widget displays the workflow response list, executes responses, and saves or closes the work item. When this widget receives a Work Item ID payload, it uses this payload to retrieve a work item from PE and

204

Developing Applications with IBM FileNet P8 APIs

broadcast the work item to other widgets, such as Attachment and Work Data widget.

Attachment widget
The Attachment widget displays attachment lists that are grouped by workflow fields. It provides document-related functions: checkout, checkin, cancel checkout, view, properties, download, and remove attachment.

Viewer widget
The Viewer widget wraps the WorkplaceXT Image Viewer applet for displaying document attachments.

Work Data widget

The Work Data widget displays customer-defined fields that are exposed on the step.

6.4.2 ECM Widgets system architecture

ECM Widgets is a framework that is based on mashup technology. Components in this framework follow the iWidget specification to communicate with each other. In ECM Widgets 4.5.1, we host the ECM Widgets in the WebSphere Business Space, which is a mashup container that provides a runtime to run the widgets, and also provides a GUI-based builder to assemble widgets to build applications. At run time, ECM Widgets reside on the browser side. All widgets are written purely in JavaScript. ECM Widgets talk to the CE and PE through the REST interface to launch workflows, process work items, and search for documents within CE. They also leverage WorkplaceXT to accomplish several CE-related functions, such as checkout and checkin documents, view document properties. Figure 6-10 on page 206 shows the system architecture of ECM Widgets.

Chapter 6. Advanced Process Engine API programming

205

Figure 6-10 ECM Widgets system architecture

6.5 Building a custom Get Next In-basket widget

ECM Widgets enables the IT user to build new custom widgets and integrate them with the existing widgets. In this section, we present instructions for building a custom Get Next In-basket widget. As its name implies, this widget automatically opens the next work item from an In-basket when the current work item is completed. It also allows the user to open the next work item directly by clicking a Get Next button. Figure 6-11 on page 207 shows a mashup page that contains the Get Next widget and integrates it with ECM Widgets. Note: Before you build a custom widget, you must have ECM Widgets 4.5.1 installed in your IBM FileNet P8 system.

206

Developing Applications with IBM FileNet P8 APIs

Figure 6-11 Get Next In-basket Widget working with the other ECM Widgets

6.5.1 Use case for the Get Next widget

When a user opens the mashup page that contains the Get Next In-basket widget, the widget retrieves the next queue element resource through PE REST service. Each queue element contains the identifier for a work item. This identifier is used to generate a work item ID payload that is sent to the Step Completion widget. In turn, the Step Completion widget uses the work item ID to fetch a complete work item object and broadcast it to the Work Data and Attachment widgets. Then, the Work Data and Attachment widgets consumes the work item payload and renders it. After the user finishes modifying the work item and completes it by clicking the Complete button in the Step Complete widget, a Close Step Processor event is sent back to the Get Next widget. That drives Get Next widget to get the next queue element and start working on the next work item. Another scenario also exists that lets the user open the next work item directly by clicking the Get Next button. When the button is clicked, the Get Next widget fetches the queue element, constructs the work item ID, AND publishes this

Chapter 6. Advanced Process Engine API programming

207

payload to the Step Completion widget. In turn, the Step Completion widget, Work Data widget, and Attachment widget will work on the next work item. Figure 6-12 illustrates the use case sequence diagram.

Figure 6-12 Use case for the Get Next widget

208

Developing Applications with IBM FileNet P8 APIs

6.5.2 Setup development environment

To build this sample widget, create a Web application in Eclipse, as follows: 1. In Eclipse, select File New to open the wizard dialog, then select the Dynamic Web Project. See Figure 6-13

Figure 6-13 Create a new Web application

2. In the next dialog, specify the project name SampleECMWidgets (Figure 6-14 on page 210) and then click Finish to create the project. In the created project, you will be able to edit the source code for this sample widget.

Chapter 6. Advanced Process Engine API programming

209

Figure 6-14 Specify the project name

6.5.3 Code structure of the Get Next widget

The Get Next widget is based on Dojo toolkit 1.0.2, which is bundled with WebSphere Business Space Web application. The widget code structure follows the normal Dojo package naming convention, which is presented in Figure 6-15 on page 211.

210

Developing Applications with IBM FileNet P8 APIs

Figure 6-15 Code Structure of Get Next Widget

As shown in Figure 6-15, the code for the Get Next widget is divided into three files: GetNext.xml This file is the XML definition for the widget. GetNext.js This file contains the JavaScript logic to control the behavior of the widget. GetNext.css This file contains the look and feel of the widget. You need to add your cascading style sheet (CSS) definitions into this file.

6.5.4 Defining the Get Next widget

According to the iWidget specification, a widget can be defined in XML style. In GetNext.xml (as shown in Example 6-18 on page 212), we define various perspectives of this widget, which are: iScope object This object is defined by the iScope attribute of iw:iwidget element, which points to the com.ibm.im.ecmwidgets.sample.GetNext Dojo class. Widget attributes In the iw:itemSet element, there are several iw:item elements that contain attribute declarations. Table 6-6 on page 212 lists these attributes.

Chapter 6. Advanced Process Engine API programming

211

Table 6-6 Widget attributes Attribute name bpmServiceBaseUrl Description The base URL for PE REST Service. To simplify the sample code, we do not provide a GUI to configure this value. If you want to change it, modify the default value stored in GetNext.xml. The application space name for retrieving role list. To simplify the sample code, we do not provide a GUI to configure this value. If you want to change it, modify the default value stored in GetNext.xml. The In-basket name used to retrieve queue elements. You can configure this value in the configuration panel of the Get Next widget. The URL of the queue elements resource. After you configure the role name and In-basket name in configuration panel, the URL of the queue elements resource for the selected In-basket will be saved into mashup page.

applicationSpaceName

inbasketName

queueElementsUrl

"Send Work Item Id" event This is a published event. The three scenarios that trigger this event are: The Get Next widget loads the first queue element. A user clicks the Get Next button. The Get Next widget receives a "Receive Close Step Processor" event. "Receive Close Step Processor" event This is a handled event. If you wire the Step Completion widget with the Get Next widget using this event, when a user clicks the Complete button in the Step Completion widget, an event will be published from the Step Completion widget to the Get Next widget. As defined in the onEvent attribute of this event, the event handler is "handleReceiveCloseStepProcessor" and which is a JavaScript function declared in the GetNext.js file.
Example 6-18 GetNext.xml

<iw:iwidget name="GetNext" xmlns:iw="http://www.ibm.com/xmlns/prod/iWidget" iScope="com.ibm.im.ecmwidgets.sample.GetNext" allowInstanceContent="true" supportedModes="view edit" mode="view"> <iw:resource uri="GetNext.js"/> <iw:resource uri="GetNext.css"/> <iw:itemSet id="attributes">

212

Developing Applications with IBM FileNet P8 APIs

<iw:item id="bpmServiceBaseUrl" value="/WorkplaceXT/P8BPMREST/p8/bpm/v1"/> <iw:item id="applicationSpaceName" value="GenericApproval"/> <iw:item id="inbasketName" value=""/> <iw:item id="queueElementsUrl" value=""/> </iw:itemSet> <iw:payloadDef name="WorkItemId"> .... </iw:payloadDef> <iw:event id="Send Work Item Id" published="true" eventDescName="SendWorkItemIdDescription"/> <iw:eventDescription id="SendWorkItemIdDescription" payloadType="WorkItemId" description="" lang="en"/> <iw:payloadDef name="WorkItem"> ... </iw:payloadDef> <iw:payloadDef name="CloseStepProcessor"></iw:payloadDef> <iw:event id="Receive Close Step Processor" handled="true" onEvent="handleReceiveCloseStepProcessor" eventDescName="ReceiveCloseStepProcessorDescription"/> <iw:eventDescription id="ReceiveCloseStepProcessorDescription" payloadType="CloseStepProcessor" description="" lang="en"/> <iw:content mode="view"> <![CDATA[ <div id="_IWID_viewModeContent" class="ecmwdgt ecmwdgtViewModeContent ecmwdgtGetNext"></div> ]]> </iw:content> <iw:content mode="edit"> <![CDATA[ <div id="_IWID_editModeContent" class="ecmwdgt ecmwdgtEditModeContent ecmwdgtGetNext"></div> ]]> </iw:content> </iw:iwidget>

6.5.5 Code skeleton for GetNext.js

The GetNext.js file is the iScope Dojo class that controls the behavior of the Get Next widget. Example 6-19 on page 214 defines its basic structure in an object-oriented way. Read the comments for each function to understand its purpose.

Chapter 6. Advanced Process Engine API programming

213

Example 6-19 Code skeleton for Get Next widget: GetNext.js

dojo.provide("com.ibm.im.ecmwidgets.sample.GetNext"); dojo.declare("com.ibm.im.ecmwidgets.sample.GetNext", null, { onLoad: function(){ // Will be invoked when mashup container loads this widget. }, onview: function(){ // Will be invoked when widget shows the view mode pane. this._renderViewModeContentNode(); }, onedit: function() { // Will be invoded when widget shows the edit mode pane. this._renderEditModeContentNode(); }, handleReceiveCloseStepProcessor: function() { // Handle the incoming Close Step Processor event. this._getNextQueueElement(); }, _getNextQueueElement: function() { // Get the next queue element and publish the // Send Work Item Id event. }, _renderEditModeContentNode: function() { // Render the edit mode content pane. }, _renderViewModeContentNode: function() { // Render the view mode content pane. }, _request: function(method, xhrArgs) { // This is the helper method that request Ajax request to PE REST // service. } });

214

Developing Applications with IBM FileNet P8 APIs

6.5.6 Rendering the Get Next widget user interface

The Get Next widget supports the view and edit modes. When a widget switches to a certain mode, the corresponding handler function is invoked. According to the iWidget specification 1.0, for the view mode, the onview() function is called; for the edit mode, the onedit() function is called. In this sample widget, the onview function calls _renderViewModeContentNode method to render the view mode GUI. Similarly, the onedit function calls _renderEditModeContentNode method to render the edit mode GUI. For the implementation of these two render methods, check the sample applications in this book. Figure 6-16 and Figure 6-17 show the Get Next widget in view and edit modes.

Figure 6-16 The Get Next widget in the view mode

Figure 6-17 The Get Next widget in the edit mode

6.5.7 Invoking PE REST service to fetch queue element

The _request function is a helper function that performs the request to PE REST service. It returns a Deferred Dojo object that allows you to register callback and errback functions. See Example 6-20.
Example 6-20 _request function

_request: function(method, xhrArgs) { if (!xhrArgs.url) {

Chapter 6. Advanced Process Engine API programming

215

console.error("No target xhr url"); var deferred = new dojo.Deferred(); deferred.errback(); return deferred; } xhrArgs.url = this.bpmServiceBaseUrl + "/" + xhrArgs.url; return dojo[method](xhrArgs); } The _getNextQueueElement function is called when the Get Next widget wants to fetch the next queue element. This method constructs the URL for the queue element resource and registers a callback function _getNextQueueElementCallback to handle the return result. It then initiates the request by calling the _request function. See Example 6-21.
Example 6-21 _getNextQueueElement method

_getNextQueueElement: function() { console.debug("[" + this.declaredClass + "." + arguments.callee.nom + "]"); var postData = "pageSize=1&queryFlags=1"; if (this.lastRecord) { postData += "&lastRecord=" + encodeURIComponent(this.lastRecord); } this._request("rawXhrPost", { url: this.queueElementsUrl, handleAs: "json", headers: { "Content-Type": "application/x-www-form-urlencoded" }, postData: postData }).addCallback(this, "_getNextQueueElementCallback"); }, The _getNextQueueElementCallback function handles the result coming from the PE REST service, getting the next queue element and publishing the data through the "Send Work Item Id" event. As a result, the target widget will be invoked to handle the incoming work item ID payload. See Example 6-22 on page 217.

216

Developing Applications with IBM FileNet P8 APIs

Example 6-22 _getNextQueueElementCallback

_getNextQueueElementCallback: function(result) { this.lastRecord = result.lastRecord; var queueElements = result.queueElements; if (!queueElements) { console.info("No more work items."); return; } var queueElement = queueElements[0]; var lockedById = queueElement.lockedById; if (lockedById !== 0 && lockedById !== this.currentUserId) { console.info("Work item is locked by another user: " + lockedById); return this._getNextQueueElement(); } var workItemId = { queueName: queueElement.queueName, workObjectNumber: queueElement.workObjectNumber }; this.iContext.iEvents.fireEvent("Send Work Item Id", "WorkItemId", workItemId); this.workItemIdNode.innerHTML = queueElement.workObjectNumber; },

6.5.8 Invoking PE REST service to fetch role and In-basket list

The steps to fetch role and In-basket list are similar to the steps for queue element. For the detailed information, see the source code in the sample widget application. Also, you can find the sample code to fetch role in Example 6-12 on page 191 and sample code for In-basket in Example 6-13 on page 192.

6.5.9 Deploying the widget

After you finish coding, deploy the custom widget by following these steps: 1. Export the project as a WAR file from Eclipse, as follows: a. Select File Export. b. In the Export dialog (Figure 6-18 on page 218), select WAR file, and save it to a location.

Chapter 6. Advanced Process Engine API programming

217

Figure 6-18 Export sample Web application

2. Deploy the WAR file to WebSphere. The context root for the deployed Web application should be SampleECMWidgets. 3. Stop the Web application of WebSphere Business Space. 4. Copy the widget registration file of the Get Next widget to the Business Spaces registry data directory. You can find the sample registration file in the sample widget application that you download from IBM Redbooks Web server. Refer to Appendix A, Additional material on page 315 to locate the information. 5. Start the Web application of Business Space. If you deploy the Get Next widget successfully, the widget appears in the widget toolbox of Business Space (Figure 6-19 on page 219) and you can drag and drop the Get Next widget into a mashup page.

218

Developing Applications with IBM FileNet P8 APIs

Figure 6-19 The Get Next widget in Business Space toolbox

6.5.10 Building the solution

As an example, to build a solution with the Get Next widget, follow these steps: 1. Open a space in WebSphere Business Space. Create a new mashup page, and name it Get Next. See Figure 6-20.

Figure 6-20 Create mashup page

2. Drag and drop the Get Next In-basket, Step Completion, Work Data, Attachment, and Viewer widgets to the page. See Figure 6-21 on page 220.

Chapter 6. Advanced Process Engine API programming

219

Figure 6-21 Place the widgets in the mashup page

3. Wire the Get Next In-basket widget with the Step Completion widget by the "Send Work Item Id" published event and "Receive Work Item Id" handled event. See Figure 6-22 for the widget wiring.

Figure 6-22 Wire Get Next and Step Completion by Send Work Item Id and Receive Work Item Id events

4. Wire the Step Completion widget with the Get Next widget by the "Send Close Step Processor" published event and "Receive Close Step Processor" received event. See Figure 6-23 for the widget wiring.

Figure 6-23 Wire Step Completion and Get Next by Send Work Item Id and Receive Work Item Id events

220

Developing Applications with IBM FileNet P8 APIs

5. Open the configuration panel of the Get Next widget, select a role name and an In-basket name. See Figure 6-24. Click Save. As a result, the Get Next widget starts to run and the other widgets display the content of current work item.

Figure 6-24 Configure the Get Next widget

Chapter 6. Advanced Process Engine API programming

221

222

Developing Applications with IBM FileNet P8 APIs

Chapter 7.

Sample applications for Fictional Auto Rental Company A

To help you better understand programming with IBM FileNet APIs, in this chapter, we describe the sample applications that are implemented for Fictional Auto Rental Company A. This chapter discusses the following topics: Introduction to sample applications Business use cases User view of the sample applications Data model Security model Workflows Internal architecture of sample applications Deployment instructions for sample applications

Copyright IBM Corp. 2009. All rights reserved.

223

7.1 Introduction to sample applications

To tie together many of the development concepts that are discussed in other chapters, we present a set of sample applications. These sample applications include: Reservation Web application Kiosk application Agent handheld application (not implemented) Fleet Status Manager Web application Billing Report application DISCLAIMER OF WARRANTIES: The accompanying code is sample code, created by IBM Corporation. This sample code is not part of any standard or IBM product and is provided to you solely for the purpose of assisting you in the development of your applications. The code is provided as is, without warranty of any kind. IBM shall not be liable for any damages arising out of your use of the sample code, even if they have been advised of the possibility of such damages. The sample applications are available for download, in both source and binary form, from the IBM Redbooks publications Web site as a companion to this book. You can download them and then deploy them in your environment with only a small amount of configuration and setup. For downloading instructions, see Appendix A, Additional material on page 315. For deployment instructions, see 7.8, Deployment instructions for sample applications on page 276. Our aim in designing these samples was to illustrate various programming and architectural techniques without trying to add them artificially. The sample applications are complete for what they are designed to do. You can easily add more features to make them a complete business solution. Although we have tested the applications to make sure that they work correctly for our own narrative purposes, and although we make the entire sample available to download for examination and study, we do not guarantee that it is free of defects. The company, business model elements, and business use cases for the sample applications are all completely fictional. We believe them to be realistic in limited circumstances, but we make no claims for applicability or fitness for any real circumstances. The sample code has hard-coded English language strings, U.S. conventions for date display, and so on. We are firm believers in globalization and localization of

224

Developing Applications with IBM FileNet P8 APIs

software, but we developed the sample without it because we believe it is somewhat easier to follow as sample code. If we were to develop this as a production application, we would definitely follow best practices for globalization and localization. Likewise, and for similar reasons, the sample applications do not adhere to best practices for accessibility. That would also be necessary and desirable for production applications.

7.2 Business use cases

Fictional Auto Rental Company A (the company) is a large, well-financed franchiser of vehicle rental operations. The company differentiates itself from the competition by having each local franchise offer a unique and varied mix of vehicles for rental to the public. As with any business, and especially for a large franchiser, the company looks for ways to keep costs low by exploiting automation. Many aspects of running the business are automated. To keep internal costs reasonable, franchisees use services from a high-end data center that is operated by the company headquarters. There are smaller computers at franchise locations, but servers that are running J2EE application servers are shared among many franchises. The primary interaction for customers to reserve vehicles is through a Web site. In contrast to mainstream rental companies, a customer of the company selects a specific vehicle. The customer can see a photo of the actual vehicle, read comments from other customers, and even see limited facts about the vehicle maintenance history. Using that same Web site, customers can register and maintain a user profile of personal factual information (for example, address and phone number) and preferences (for example, default class of vehicle). The unique nature of the company inventory of vehicles means that some rentals are done when a particular vehicle is available, so customers can sign up for alerts for specific vehicle availability and other events. Back-office applications keep track of vehicle scheduling, expected pick-ups and drop-offs, rental history, and maintenance. Maintenance can be routine (for example, an oil change, typically handled by in-house staff) or exceptional (for example, repair of body damage, typically handled by an outside service contractor). Employees of the company want to be able to check on the availability and historical records for a specific vehicle, but they also want dashboard functions, which will tell them quickly how many vehicles are in each of various status categories (for example, out for rental, in maintenance, idle, and overdue for return).

Chapter 7. Sample applications for Fictional Auto Rental Company A

225

7.3 User view of the sample applications

In this section, we discuss the applications as seen from the point of view of the users. For our sample applications, the users are employees and customers of the company and its franchisees.

7.3.1 User view: Reservation Web application

For a description of the internal architecture of this application, see 7.7.1, Architecture: Reservation Web application on page 268. For a description of the deployment instructions for this application, see 7.8.4, Deployment: Reservation Web application on page 292. Vehicle reservations are made primarily through a Web site. The site has some relatively static information (for example, information about the company and information about services available). Web site content is also available, is dynamic in nature, and is driven by ECM systems. The Reservation Web application provides the following functions. For simplicity, not all of these features are implemented in the sample code. Customers can browse the available inventory of vehicles and view information related to any particular vehicle. A customer can create an account with user ID and password. After creating an account, the customer can enter information in a user profile. A customer can rent a vehicle that is available for a specific date and location. To rent a vehicle, the customer must have an account, and must also supply certain information (for example, address, telephone number, credit card information). The information defaults to information from the user profile and gives an option to save any updates back to the profile. The customer is given the option of permanently storing or not storing the credit card information that is used for a given rental transaction. The customer can reset the password and have the new password sent through e-mail. With each confirmed vehicle reservation, a confirmation number is provided for the customer for rental pickup. Several screen captures of the Reservation Web application are shown in the remaining figures in this section.

226

Developing Applications with IBM FileNet P8 APIs

Figure 7-1 shows the login panel.

Figure 7-1 Reservation Web application login panel

If not a member, the user must register in the system. Figure 7-2 shows the registration panel. The user must enter his or her information including a new user ID and password that are required when reserving a vehicle from this application.

Figure 7-2 Reservation Web application registration panel

Chapter 7. Sample applications for Fictional Auto Rental Company A

227

Figure 7-3 shows the main reservation data panel. This is where the user can select where and when to pick up the automobile.

Figure 7-3 Reservation Web application main data panel

After the user selects where and when to rent a vehicle, the system displays a list of available vehicles to be rented during this period of time in that location, as shown in Figure 7-4. The user selects a vehicle of his or her choice. Each vehicle has a photo associated with it, a brief description, and a daily rate for rental.

Figure 7-4 Reservation Web application vehicle selection panel

228

Developing Applications with IBM FileNet P8 APIs

After selecting the rental vehicle, the user enters the information of the credit card that is used for the reservation. Figure 7-5 shows the credit card information collection panel.

Figure 7-5 Reservation Web application credit card information collection panel

Chapter 7. Sample applications for Fictional Auto Rental Company A

229

When the reservation is complete, the system displays a reservation confirmation panel. Figure 7-6 shows the confirmation panel for a successful rental reservation.

Figure 7-6 Reservation Web application rental reservation confirmation panel

7.3.2 User view: Kiosk application

For a description of the internal architecture of this application, see 7.7.2, Architecture: Kiosk application on page 270. For a description of the deployment instructions for this application, see 7.8.5, Deployment: Kiosk application on page 295. The company uses a PC-based Kiosk application for a variety of functions. Many of these functions depend on hardware, which you probably do not have, so we simulate these actions with other user interface elements. Upon approaching the kiosk, the user (who is the customer) is presented with options for either picking up or dropping off a car.

230

Developing Applications with IBM FileNet P8 APIs

At the rental counter when picking up a car, the customer enters the rental car confirmation number (provided in the confirmation panel as shown in Figure 7-6). The Kiosk application matches the confirmation number to the reservation and (in reality) physically delivers the keys and rental paperwork to the customer. In our samples however, the Kiosk application simulates delivery of the keys and paperwork by supplying two codes. One code identifies the vehicle being rented, and the other code represents this specific rental transaction. Figure 7-7 shows the interface for Rental Kiosk application where the user enters the rental car confirmation number.

Figure 7-7 Entering confirmation number

Figure 7-8 shows the rental information based on the confirmation number entered.

Figure 7-8 Rental information dialog

Chapter 7. Sample applications for Fictional Auto Rental Company A

231

When a vehicle returns to the rental lot, a sensor scans a vehicle bar code from a window sticker. We simulate this by having the user enter the code using the keyboard. See Figure 7-9.

Figure 7-9 Vehicle return dialog

7.3.3 User view: Agent handheld application

When a vehicle is returned from rental, an employee of the company accepts the vehicle from the customer. Agents use a custom-designed handheld device. The device communicates with the back-office systems through a radio link. The agent handheld application provides the following functions: Agents scan the bar code from the window sticker of the vehicle. On-board diagnostics (OBD) are read from the vehicle OBD II port and stored with the vehicles maintenance records. The diagnostics might indicate the need for a certain kind of maintenance. If it is not available from the on-board diagnostics data (which varies from vehicle to vehicle), the vehicle mileage and fuel level are manually recorded by the agent. The agent also records any exceptional conditions, such as body damage, or missing parts or accessories. In the sample application, we do not show the agent handheld application.

7.3.4 User view: Fleet Status Manager Web application

For a description of the internal architecture of this application, see 7.7.3, Architecture: Fleet Status Manager Web application on page 272. For a description of the deployment instructions for this application, see 7.8.6, Deployment: Fleet Status Manager Web application on page 296. Most minor maintenance is done by an in-house maintenance shop collocated with the rental office. The company keeps meticulous maintenance records by

232

Developing Applications with IBM FileNet P8 APIs

using IBM FileNet P8 system. As part of the sample, we do not provide an implementation of the Kiosk application used by the vehicle technicians to record low-level maintenance details (for example, parts used and breakdown of labor), but we do provide a Web application used by the maintenance supervisor to accept, schedule, and release vehicles. The same Web application is used by rental office staff because maintenance is just one aspect of tracking vehicle availability. The Fleet Status Manager Web application provides the following functions: A fleet status dashboard mode provides a summary of the status of all vehicles in the inventory (for example, out for rental, idle, in maintenance). A vehicle may be put into maintenance status by authorized employees. This can be well before the vehicle is actually moved to the maintenance area. The effect is that the vehicle is unavailable for future rentals during this period. The Kiosk application is used to scan the window sticker bar code as the vehicle is moved physically into or out of the maintenance area. We do not show that aspect in the sample application. An authorized maintenance employee may enter or update a maintenance substatus for the vehicle (for example, waiting for parts, working, sent to external service vendor). Most substatus changes require an estimated completion time for that status. An authorized maintenance employee may take a vehicle out of maintenance status, typically to return it to service. As a special case of the maintenance application, newly-acquired vehicles are added to the inventory and scheduled for an initial maintenance examination. Likewise, when a vehicle is being sold or disposed of, it gets a final maintenance inspection as it is removed from the active inventory. Several screen captures of this Fleet Status Manager Web Application are shown in the remaining figures in this section.

Chapter 7. Sample applications for Fictional Auto Rental Company A

233

Figure 7-10 shows the Fleet Status Dashboard page, which lists vehicles in inventory and their status. Supervisor users have access to this page through security roles.

Figure 7-10 Fleet Status Dashboard

234

Developing Applications with IBM FileNet P8 APIs

When the supervisor selects one of the vehicles from the inventory, the vehicle information appears, as shown in Figure 7-11. The supervisor can then send the vehicle to maintenance.

Figure 7-11 Fleet Manager page to send vehicle for maintenance

When a maintenance worker logs into this application, the worker is routed to a page that lists the vehicles in the maintenance group queue as well as those in that specific workers inbox queue. These vehicles are represented by Process Engine (PE) work items. Figure 7-12 shows a typical page that a maintenance worker sees after logging in.

Figure 7-12 Maintenance queue page

Chapter 7. Sample applications for Fictional Auto Rental Company A

235

The maintenance worker can select a vehicle from the Unassigned Work Items list, and route the item to his or her own work queue. See Figure 7-13.

Figure 7-13 Maintenance worker page: route work item to workers work queue

After the vehicle is in the maintenance workers work queue, the maintenance worker can complete the work item or send it back to the group queue. See Figure 7-14.

Figure 7-14 Maintenance worker page: process work item

236

Developing Applications with IBM FileNet P8 APIs

Figure 7-15 shows the maintenance queues page when logging in as a different maintenance worker. The purpose of this picture is to show that the group queue work items are available to all users who have access to this queue, although the users inbox only shows that users assigned work items.

Figure 7-15 Maintenance queue page for a different user

All of these work items can be tracked by using the Process Administrator applet in WorkplaceXT. Figure 7-16 shows the maintenance work items in the roster. The F_BoundUser column shows which users inbox that the work item is assigned to.

Figure 7-16 Process Administrator showing work items in roster

7.3.5 User view: Billing Report application

For a description of the internal architecture of this application, see 7.7.4, Architecture: Billing Report application on page 274. For a description of the deployment instructions for this application, see 7.8.7, Deployment: Billing Report application on page 299. Although we refer to it as the Billing Report application, it is really a collection of reports run periodically for the purpose of running the companys business. These can be run on demand, but they are typically run daily, during off hours, and cover the previous 24-hour period.

Chapter 7. Sample applications for Fictional Auto Rental Company A

237

The Billing Report application provides the following functions: Summary and details of all vehicle hours by status (for example, out for rental, idle, and in maintenance). Summary and details of all payments made (credit card charges) for completed rentals. Summary and details of all maintenance expenses (parts, labor, third party). Exception report of all unusual events (for example, vehicle not ready when promised to a customer, vehicle overdue for return, vehicle entering non-routine maintenance). In the sample application, we only show the summary and details report for vehicles hours by status. Several screen captures of this Billing report application are shown in the remaining figures in this section. Figure 7-17 shows the login dialog for this application. The user must enter an IBM FileNet P8 user ID and password to log in.

Figure 7-17 Login dialog for Billing Report application

If the user enters wrong information, the Billing Report application raises an error message, as shown in Figure 7-18.

Figure 7-18 Login failed

After login, the user provides the start date and end date to generate the report. Figure 7-19 shows the generated report displayed in the Billing Report

238

Developing Applications with IBM FileNet P8 APIs

application. The vehicle hours report contains two sections: Summary Report and Detailed Report. The Summary Report lists the total hours and total numbers of activities for the various status types of the vehicle. By clicking a row in Summary Report, the Detailed Report is refreshed and lists the corresponding detailed information.

Figure 7-19 Main window for Billing Report application

For this sample application, we did not implement the Print Report function.

7.4 Data model

Many types of data are involved in running the sample applications. We concentrate here only on the data that resides within the IBM FileNet P8 back-end systems and used by the sample applications. These consist of custom classes and properties defined within the Content Engine (CE) metadata. Figure 7-20, Figure 7-21, and Figure 7-22 provide a Unified Modeling Language (UML) overview of the classes. To keep the figures uncluttered, relationships from individual UML attributes (corresponding to CE properties) are not shown.

Chapter 7. Sample applications for Fictional Auto Rental Company A

239

They can be inferred from the attribute data types in the diagrams and also from the explanatory material in the tables in this chapter. The data file for the UML diagrams (created with IBM Rational Software Architect 7.5) is included in the downloadable material that accompanies this book, in file: sg247743-sample/CEartifacts/FictionalAutoRentalCompanyADataModel.emx In the diagrams and in the explanations that follow, we use the class and property symbolic names. These are the unique, case-insensitive, non-localizable names that are used by applications. Classes and properties also have localizable display names. A common general pattern for display names is to create the English version by removing the prefix from the symbolic name and inserting spaces between words in the symbolic name. For example, a symbolic name of ITSOVehicleActivity could have a display name of Vehicle Activity. In the data import files for the sample applications, we use display names that are the same as the symbolic names. Figure 7-20 shows the model for ITSODocument class and its subclasses.

Figure 7-20 Model for ITSODocument and subclasses

240

Developing Applications with IBM FileNet P8 APIs

Figure 7-21 shows the models for ITSOFolder, ITSOAnnotation, and their subclasses.

Figure 7-21 Models for ITSOFolder, ITSOAnnotation, and subclasses

Chapter 7. Sample applications for Fictional Auto Rental Company A

241

Figure 7-22 shows the model for ITSOCustomObject and its subclasses.

Figure 7-22 Model for ITSOCustomObject and subclasses

242

Developing Applications with IBM FileNet P8 APIs

7.4.1 Base classes

As an organizational technique, we first define a series of base classes. There is a custom base class for each of the types of objects used elsewhere in the data model. Aside from showing our intent explicitly in the object hierarchy, these base classes also serve as a point where we can apply any common custom properties and perform broad queries. If we did that instead with the system base classes, we might affect or include classes that are not part of our application. The custom base classes include: ITSODocument is a subclass of system class Document. ITSOFolder is a subclass of system class Folder. ITSOCustomObject is a subclass of system class CustomObject. ITSOAnnotation is a subclass of system class Annotation. Properties common to all custom base classes are listed in Table 7-1 on page 244. Note: In the data model tables, the following notational conventions apply: String(n) means a string-valued property with a maximum length of n. Binary(n) means a binary value of at most n bytes. Long String means a string-valued property with its UsesLongColumn metaproperty set to true. OVP(xyz) means an object-valued property with a required type of xyz. MVOVP(xyz) means a multi-valued object-valued property with a required type of xyz. Date-valued properties are always date and time combinations with a resolution accurate to whatever the underlying database supports.

Chapter 7. Sample applications for Fictional Auto Rental Company A

243

Table 7-1 ITSODocument, ITSOFolder, ITSOCustomObject, ITSOAnnotation Class: ITSODocument, ITSOFolder, ITSOCustomObject, ITSOAnnotation Property ITSOFranchiseCode Type String(10) Comment Value required. Franchise identity code is always normalized to lowercase. A pseudo-code of itsoautohq is for the company headquarters. Most objects have a single pointer to an ITSORole object for security inheritance purposes. Even objects which are publicly visible have a value here for simplified administration.

ITSOSecurityProxy

OVP (ITSORole)

Folders must exist at a single place in a strictly hierarchical namespace. All folders of any kind for these applications are under the top-level folder /ITSOAutoRental/. A franchise has a franchise identity code, and folders specific to a particular franchise are placed under a subfolder named after the franchise code. For example, a franchise with identity code ought2 would have all its folders under /ITSOAutoRental/ought2/.

7.4.2 ITSOVechicle
This class acts as a collector of information about many things that are related to a specific physical vehicle. Examples of these assorted things include scanned images of legal documents related to ownership and registration, photographs, and accident reports. Because it acts as a collecting point for many different types of things, it is modelled as a folder object. Various documents of unpredictable types may be filed into a folder for a specific vehicle. A large franchise may have hundreds of vehicles in inventory, so browsing folders as a way of searching for vehicles is not very practical. We instead expect the applications to identify a particular vehicle through a search and then use the corresponding folder as a paradigm for navigating to related documents. Each vehicle has a unique identity code. That identity code is used as the name of the ITSOVehicle folder. To facilitate occasional browsing navigation, the ITSOVehicle folders are filed into subfolders that are named for the character in the ninth position of the identity code (all of which are filed under a vehicles subfolder). If no ninth character exists, the last non-space character is used. If that character is a letter, it is converted to lowercase. For example, a vehicle with

244

Developing Applications with IBM FileNet P8 APIs

an identity code of JT4TN12D8V0030519 could be represented by a folder with the following path: /ITSOAutoRental/ought2/vehicles/8/JT4TN12D8V0030519 Allowing at least 17 characters for the identity code allows for the direct use of the industry standard Vehicle Identification Number (VIN). Position 9 of the VIN is a check digit and thus well-distributed; it has eleven possible values (digits 0 - 9, letter X). Because of the unique inventories of the company franchisees, they might have older vehicles without a VIN, or with a VIN in a different format. Much of the provenance information for a vehicle (for example, year, make, and model of manufacture) can be extracted by decoding a standard VIN, but it must be explicitly provided for other types of identity codes. For purposes of the sample application, we assume that all vehicles have a VIN and the ITSOVehicleId property always matches the ITSOVehicleVIN property. In other words, ITSOVehicleIdIsVIN is always true for the sample application. By convention, there are documents filed into an ITSOVehicle folder with specific containment names to indicate specific purposes. See, for example, 7.4.3, ITSOPhotoGallery on page 247. Note the following information: Not shown elsewhere in this book is an ITSODocument with a containment name of obd. Its content is an XML document holding OBD II diagnostics data that is collected from the vehicle. Each time OBD II data is collected, a new version of the document is created so that old data sets are preserved. Not shown elsewhere in this book is an ITSODocument with a containment name of description. It contains the full description of the vehicle. The actual descriptive narrative is contained in the first content element, and it must be in a format understood by the rendering application. For our samples, we use text or HTML descriptions. Rental charges for a vehicle are abstracted to a separate ITSOChargesSchedule object, which can be shared by many vehicles. For our sample application, we do not use or further describe the ITSOChargesSchedule (vehicle rental charges are supplied by hard-coded application logic). After it is created, ITSOChargesSchedule objects are not modified because they are still pointed to by ITSORentalActivity objects from the past. A DeletionAction metaproperty value of DeletionAction.PREVENT on the MVOVP that is pointing from the ITSOChargesSchedule to ITSOVehicle objects can be used to ensure referential integrity for attempted deletions.

Chapter 7. Sample applications for Fictional Auto Rental Company A

245

Data
Table 7-2 shows the properties of ITSOVehicle.
Table 7-2 ITSOVehicle Class: ITSOVehicle (subclass of ITSOFolder) Property ITSOVehicleId ITSOIsAvailable Type String(25) Boolean Comment Value required. Standard VIN is 17 characters. As a performance measure, this field can be manually set to false when a vehicle is in maintenance for an unusually long time. It is also set to false when the vehicle is disposed of. Value required. VIN decoding can be used if this value is true. Manufacturers model year Manufacturer name Manufacturers model name Same codes as used in a VIN Intended for searching. More detailed and nuanced color descriptions can be provided in the vehicle description. Value required. Short description that is suitable for display in search summaries, and other things. This property avoids an immediate fetch of the full description document that is contained in the folder. Daily rate Total mileage of car Reflective of ITSOVehicleActivity property ITSOVehicle.

ITSOVehicleIdIsVIN ITSOVehicleYear ITSOVehicleMake ITSOVehicleModel ITSOVehicleType ITSOColor

Boolean String(4) String(10) String(10) String(1) String(15)

ITSOShortDescription

Long String(100)

ITSODailyRate ITSOMileage ITSOVehicleActivities

String(10) Integer MVOVP (ITSOVehicleActivity)

246

Developing Applications with IBM FileNet P8 APIs

Security
ITSOVehicle security is defined as follows: ITSOVehicle can be read by customers and all employees. ITSOVehicle can be created, updated, and deleted by franchise supervisory employees. The OBD II document can be created, read, and updated by any franchise employee. An update for the OBD II document is a checkout and checkin cycle. The OBD II document versions can be deleted by franchise supervisory employees. ITSOChargesSchedule can be read by customers and all employees. ITSOChargesSchedule can be created, updated, and deleted by franchise supervisory employees.

7.4.3 ITSOPhotoGallery
A document filed in an ITSOVehicle folder with a containment name of photogallery contains photographs of the vehicle. All photographs are in either JPEG, GIF, or PNG format. The primary photograph is in the first content element. Additional photographs are in other content elements.

Data
The ITSOPhotoGallery is a subclass of ITSODocument and has no specific subclass custom properties.

Security
An ITSOPhotoGallery completely inherits its security from the containing ITSOVehicle, and is implemented through the normal dynamic security inheritance of the SecurityFolder property.

7.4.4 ITSOThumbnail
By convention, each content element in an ITSOPhotoGallery document has an ITSOThumbnail annotation containing a smaller version of the image. The thumbnail image is contained in the first and only content element of the annotation and is also in JPEG, GIF, or PNG format.

Chapter 7. Sample applications for Fictional Auto Rental Company A

247

Data
The ITSOThumbnail is a subclass of ITSOAnnotation and has no specific subclass custom properties.

Security
An ITSOThumbnail completely inherits its security from the annotated ITSOPhotoGallery, and is implemented through the normal dynamic security inheritance of the AnnotatedObject property.

7.4.5 ITSOVehicleActivity
A vehicle is always in some kind of activity state; the default state is idle. All activity records indicate a timespan, expressed as start and end date properties. In some cases, those dates are adjusted over the lifetime of the activity record. For example, the start and end dates are estimates for a reservation and are adjusted to actual rental period dates when the vehicle is picked up and returned. We can look both backward and forward in time to see when a vehicle was in a state or when it is expected to be in a state. For example, when a vehicle is reserved for rental by a customer, an activity record with a future timespan is created. Any period of time that is not covered by an activity record is considered idle time, and the vehicle is available for rental for that period (although business rules enforce a buffer period around some types of activities to recognize that the vehicle might not become available exactly when expected). The major states for vehicles are reflected in subclasses of ITSOVehicleActivity (see 7.4.7, ITSORentalActivity on page 250; 7.4.8, ITSOMaintenanceActivity on page 251; and 7.4.9, ITSODisposalActivity on page 253). Several other minor activities exist, but we do not show them here (for example, when a vehicle has been returned and is being washed and prepared for the next rental).

248

Developing Applications with IBM FileNet P8 APIs

Data
Table 7-3 shows the properties of ITSOVehicleActivity.
Table 7-3 ITSOVehicleActivity Class: ITSOVehicleActivity (subclass of ITSOCustomObject) Property ITSOVehicle ITSOStartDate ITSOEndDate ITSODisposalActivity Type OVP (ITSOVehicle) Date Date MVOVP (ITSODisposalActivity) Comment If this activity led directly to a disposal, then this property points to the disposal activity object. Although for referential integrity purposes this is a reflective property corresponding to the ITSODispoalActivity property ITSODisposingActivity, we expect it to have either no value or a single value.

Security
ITSOVehicleActivitys security is defined as follows: All ITSOVehicleActivity records can be read by customers and all employees. Customers must read all records for the purpose of doing reservation activity; however, application logic restricts display to only ITSORentalActivity records which are affiliated with that customer through the ITSOCustomer property. Further access is described for each ITSOVehicleActivity subclass.

7.4.6 ITSOSingleton
ITSOSingleton is used to maintain the last confirmation ID that is used for a car rental. See Table 7-4.
Table 7-4 ITSOSingleton Class: ITSOSingleton (subclass of ITSOCustomObject) Property ITSOIntegerValue Type Integer Comment Used to generate the unique ITSOConfirmationID for rental reservations.

Chapter 7. Sample applications for Fictional Auto Rental Company A

249

7.4.7 ITSORentalActivity
This type of activity record is used both for reservations, in-progress rentals, and completed rentals. Payment details, including credit card number and itemized charges, are recorded in an ITSOChargesRecord. For simplicity in our sample application, we do not use or further describe the ITSOChargesRecord. Instead, we directly record an artificial credit card-like number in the ITSOCreditCardSimulation property.

Data
Table 7-5 shows the properties of ITSORentalActivity.
Table 7-5 ITSORentalActivity Class: ITSORentalActivity (subclass of ITSOVehicleActivity) Property ITSOConfirmationID Type Integer Comment Fetch the ITSOSingleton object and get the ITSOIntegerValue and increment it by 1 for generating ITSOConfirmationID. Save the ITSOSingleton object with incremented value. Value required. It can be one of the following values: reserved, active, historical Is used for historical and auditing purposes. Cancelled reservations do not participate in vehicle activity. Daily rate Artificial credit card-like payment indicator Pointer to a separate payment record containing credit card and other details

ITSORentalStatus

String(10)

ITSORentalIsCancelled

Boolean

ITSOCustomer ITSODailyRate ITSOCreditCardSimulation ITSOChargesRecord

OVP (ITSOCustomer) Integer String(12) OVP (ITSOChargesRecord)

250

Developing Applications with IBM FileNet P8 APIs

Class: ITSORentalActivity (subclass of ITSOVehicleActivity) Property ITSOCurrency Type String(3) Comment The currency type for money amounts. Values are usd, cad, eu, and so on. Total estimated or actual rental charges rolled up from the ITSOChargesRecord Net estimated or actual rental charges (after direct costs like taxes and fees are deducted) that are rolled up from the ITSOChargesRecord Is used to demonstrate versioning Mileage at the beginning of rental Miles driven during rental period.

ITSOChargesTotal

String

ITSOChargesNet

String

ITSORentalAgreement ITSOStartingMileage ITSOMilesDriven

OVP (ITSODocument) Integer Integer

Security
ITSORentalActivity security is defined as follows: ITSORentalActivity objects may be created, read, and updated by any customer, but application logic restricts read access to objects associated with that customer. All create and update activities are mediated by an application. ITSORentalActivity objects may be created, read, updated, and deleted by any employee of the franchise to which the vehicle belongs.

7.4.8 ITSOMaintenanceActivity
This type of activity records a single maintenance event, generally meaning a single contiguous trip to the maintenance area. There are additional ITSOMaintenanceDetail objects showing time spent waiting, sent to outside service contractors, and so on. The details are aggregated to a single ITSOMaintenanceActivity record. The sample application does not show the ITSOMaintenanceDetail objects.

Chapter 7. Sample applications for Fictional Auto Rental Company A

251

Data
Table 7-6 shows the properties of ITSOMaintenanceActivity.
Table 7-6 ITSOMaintenanceActivity Class: ITSOMaintenanceActivity (subclass of ITSOVehicleActivity) Property ITSOMaintenanceCode Type String (10) Comment Required value. One of a set of fixed values including oilchange, tirerepair, bodydamage, and so on. Unusual maintenance can be described in:
ITSOMaintenanceComment

ITSOMaintenanceComment

Long String(5000)

Manually entered text describing the nature of the maintenance Reflective from ITSOMaintenanceDetail property ITSOMaintenanceActivity Currency type for money amounts. Values are usd, cad, eu, and so on. Cost of parts charged by an outside service contractor. Money amounts are multiplied by 100 to preserve accuracy. Cost of parts from in-house sources. Money amounts are multiplied by 100 to preserve accuracy. Labor hours from outside service contractors Labor hours from in-house technicians

ITSOMaintenanceDetails

MVOVP (ITSOMaintenanceDetail)

ITSOCurrency

String(3)

ITSOPartsExternal

Integer

ITSOPartsInternal

Integer

ITSOLaborExternal ITSOLaborInternal

Float Float

252

Developing Applications with IBM FileNet P8 APIs

Security
ITSOMaintenanceActivity security is defined as follows: ITSOMaintenanceActivity objects may be read by any employee of the franchise to which the vehicle belongs. ITSOMaintenanceDetail objects may be read by any maintenance employee or supervisory employee of the franchise to which the vehicle belongs. ITSOMaintenanceActivity objects may be created, updated, or deleted by any supervisory employee of the franchise to which the vehicle belongs. There is typically at least one supervisory employee in the maintenance department. ITSOMaintenanceDetail objects may be created, updated, or deleted by any maintenance employee of the franchise to which the vehicle belongs.

7.4.9 ITSODisposalActivity
When a vehicle is sold or removed permanently from inventory, the vehicle records are kept online for a period of time. To prevent an unavailable vehicle from being scheduled for any real activities, it is given an ITSODisposalActivity record with an end date in the distant future. The sample application does not show the process of disposing of vehicles.

Data
Table 7-7 shows the properties of ITSODisposalActivity.
Table 7-7 ITSODisposalActivity Class: ITSODisposalActivity (subclass of ITSOVehicleActivity) Property ITSODisposalType Type String(10) Comment Value required. One of the values sold, scrapped, stolen. Manually entered text describing the nature of the disposal. An optional pointer to a rental or maintenance activity that led to the disposal.

ITSODisposalComment

Long String(5000)

ITSODisposingActivity

OVP (ITSOVehicleActivity)

Chapter 7. Sample applications for Fictional Auto Rental Company A

253

Security
ITSODisposalActivity is defined as follows: ITSODisposalActivity objects may be read by any employee of the franchise to which the vehicle belongs. ITSODisposalActivity objects may be created/updated/deleted by any supervisory employee of the franchise to which the vehicle belongs.

7.4.10 ITSOIdleActivity
When a vehicle has no explicit activity of some other type, it is said to be idle. It is possible to deduce the idle periods by noting the gaps between other types of activities. However, it is difficult to perform a query for that gap, so the repository uses explicit ITSOIdleActivity objects to keep track of the gaps. To find vehicles available for rental, the application queries for ITSOIdleActivity records for the applicable time frame.

Data
The ITSOIdleActivity is a subclass of ITSOVehicleActivity and has no specific subclass custom properties.

Security
ITSOIdleActivity security is defined as follows: ITSOIdleActivity objects may be read by any customer or employee. Because ITSOIdleActivity objects are managed completely by event action handlers (which run in system context), they are read-only to all customers and employees.

7.4.11 ITSOCustomer
Employees have entries in the enterprise directory, but customers do not. Customer information is kept in the repository. Customer information can be coarsely divided into two types: customer factual information (for example, address) and customer preference information. Information that is not typically needed for routine customer searches is kept as XML as the document content. A new document version is created when the information is revised so that prior data sets can be kept.

254

Developing Applications with IBM FileNet P8 APIs

Data
Table 7-8 shows the properties of ITSOCustomer.
Table 7-8 ITSOCustomer Class: ITSOCustomer (subclass of ITSODocument) Property ITSOCustomerUserid ITSOCustomerPassword Type String (20) Binary (20) Comment Used by customer for login identity SHA-1 160-bit hashed password value. Used for authentication. Reflective from ITSORentalActivity property ITSOCustomer Reflective from ITSOCommentary property ITSOCustomer

ITSORentalActivities

MVOVP (ITSORentalActivity) MVOVP (ITSOCommentary)

ITSOCommentaries

Security
ITSOCustomer security is defined as follows: ITSOCustomer objects may be created, read, updated, or deleted by customers or any employee at headquarters or at a franchise location. Application logic restricts a customer to viewing or acting on the single applicable ITSOCustomer object.

7.4.12 ITSOCommentary
As an aspect of social networking, the company allows customers to make comments and provide ratings of specific vehicles. These comments are visible to customers as they browse the vehicle inventory. A comment is always tied to a particular ITSOCustomer, but it is not a requirement that a customer has rented that specific vehicle in the past. For example, customers might like to make comments that are related only to the type or model of vehicle. ITSOCommentary is implemented as a custom annotation class annotating the ITSOVehicle objects. Although the company does not censor negative comments, all comments added go through an editorial moderation process to screen for obscenities, spam, legal risks, and so on. Moderators may also make minor formatting, grammar, and spelling corrections.

Chapter 7. Sample applications for Fictional Auto Rental Company A

255

Data
Table 7-9 shows the properties of ITSOCommentary.
Table 7-9 ITSOCommentary Class: ITSOCommentary (subclass of ITSOAnnotation) Property ITSOCustomer ITSOCommentaryDate Type OVP (ITSOCustomer) Date Comment The moderation process is prone to disrupt the DateCreated and DateLastModified properties, so this explicit date reflects when the commentary was originally created by the customer. Values are restricted to 0 - 10, where 0 means not rated. A conventional five-star scale with the possibility of half-stars can be created by dividing this value in half. Value is assumed to be HTML text. Moderation life cycle status of a particular comment. Values are raw (as received from the author), edit (being edited by a moderator), and ready (available for viewing by customers)

ITSORating

Integer

ITSOCommentaryText ITSOCommentaryStatus

Long String(5000) String(5)

256

Developing Applications with IBM FileNet P8 APIs

Security
ITSOCommentary security is defined as follows: When a commentary is initially created, it has an explicit access control entry of AccessType.DENY for all access rights for the customers. As part of the moderator approval process, that access control entry is removed. Thus, normal access control mechanisms will render unmoderated commentaries invisible to customers. ITSOCommentary objects are visible to customers (after the moderation process) and any headquarters or franchise employee. ITSOCommentary objects may be created, updated, or deleted by a customer or any non-maintenance employee of the franchise to which the vehicle belongs. Application logic restricts customers to updating or deleting comments that they created.

7.4.13 ITSOFranchise
ITSOFranchise objects provide details for Fictional Auto Rental Company A franchise locations. The inherited ITSOFranchiseCode is used by many other classes to narrow those objects to a particular franchise location. Note: The sample application assumes one location per city. In reality, one city could have multiple locations.

Data
Table 7-10 shows the properties of ITSOFranchise.
Table 7-10 ITSOFranchise Class: ITSOFranchise (subclass of ITSOCustomObject) Property ITSOFranchiseName ITSOFranchiseCity ITSOFranchiseStateProvince ITSOFranchisePostalCode ITSOFranchiseDescription ITSOFranchiseTelephone ITSOFranchiseDirections Type String(20) String(20) String(3) String(10) String(30) String(12) String(30)

Chapter 7. Sample applications for Fictional Auto Rental Company A

257

7.4.14 ITSORole
ITSORole objects identify functional types of users. The sample application considers the following distinct types of users, each represented by a distinct ITSORole object with the indicated ITSORoleName: Customer (customer) Headquarters employee (hq) Franchise supervisory employee (supervisor) Franchise front-office (customer-facing) employee (frontoffice) Franchise maintenance employee (maintenance) An important role is customer, which is the role for the single account that is used by the back-end systems for all customer interactions. The corresponding ITSORole object has an ITSOFranchiseCode of itsoautohq, as does the ITSORole object for headquarters employees. Objects that represent roles serve two distinct purposes in the sample application. It is not a requirement that the same set of ITSORole objects be used for both purposes. Because a large amount of overlap exists, we do not specialize it further into subclasses. Instead, we distinguish the type of individual ITSORole object through custom properties and business rules, which use those properties. A franchise supervisory employee has access to everything to which either a franchise front-office employee or a franchise maintenance employee has access. We model this with ITSORole objects by having both the frontoffice and maintenance ITSORole objects for a given franchise inheriting security (through the ITSOSecurityProxy property) from the supervisor ITSORole object for that same franchise.

Roles for application customization

Roles are used to control the visibility of various features in the user interfaces of application components. This is not a form of security access control; it is merely user experience policy guidance to the application logic. Someone could write a different application and completely bypass this role-based activity. The specific mechanism is that if a user can see a particular ITSORole object (that is, AccessRight.READ), then that user is allowed to see the related functions in the user interface. All maintenance workers in a particular franchise would be listed in the ACL for the ITSORole object with that ITSOFranchiseCode and an ITOSRoleName of maintenance. Thus, the ACL on the ITSORole object is used to indirectly control the application. For example, screens that can logically only be used by maintenance employees can be hidden from users who do not have access to at least one

258

Developing Applications with IBM FileNet P8 APIs

maintenance ITSORole object. To make its decision, the application issues a query of the form: SELECT TOP 1 Id from ITSORole WHERE ITSORoleName = 'maintenance' If at least one result is returned from that query, the application knows that the user has access to one or more maintenance ITSORole objects. In this particular example, where we are controlling the display of application features visible only to maintenance employees, matching on the ITSOFranchiseCode is not necessary. Security constraints on the maintenance-related objects ensure that users see only objects that are associated with a particular franchise.

Roles for dynamic security inheritance

Roles are used to control access to other types of objects. The specific mechanism that is used is dynamic security inheritance. Dynamic security inheritance is a feature whereby an object can inherit some or all of its security access control from one or more proxy objects. In our data model, the ITSORole objects are the security proxy objects. The use of the ITSORole objects for this purpose is described in more detail in the 6.2, Application space, role, and workbasket on page 187.

Data
Table 7-11 on page 260 shows the properties defined for ITSORole.

Chapter 7. Sample applications for Fictional Auto Rental Company A

259

Table 7-11 ITSORole Class: ITSORole (subclass of ITSOCustomObject) Property ITSORoleName Type String(20) Comment Short but meaningful name for this ITSORole object, as described in the text. Uniqueness is enforced by convention when combined with the ITSOFranchiseCode for this object. Value is true if this ITSORole is used for security access control. Value is true if this ITSORole is used for controlling user interface elements. Additional security proxy property. Used when combining two roles together. See ITSOSecurityProxy2 property.

ITSORoleDescription ITSOIsSecurityRole ITSOIsFunctionalRole

Long String(250) Boolean Boolean

ITSOSecurityProxy2

OVP (ITSORole)

ITSOSecurityProxy3

OVP (ITSORole)

Security
The security of the ITSORole objects (of both types) is dictated by how each individual ITSORole object is used. Unlike most of the other objects in our data model, no users (other than the system administrators) inherently have access to the ITSORole objects.

7.5 Security model

For the purposes of our sample application, we divide users into the distinct populations given in 7.4.14, ITSORole on page 258. In a real application, much greater granularity would likely exist, and we would also have to account for users being members of more than one functional group. In addition to user populations with corresponding ITSORole objects, there are also headquarters system administrators who fill various functional roles, which we do not address in our sample application.

260

Developing Applications with IBM FileNet P8 APIs

Note: For simplicity in the sample application, we have made the customer pseudo-identity slightly more powerful than we would like, and we enforce additional controls at the application level. For example, to be able to make reservations, the customer must have security access to all ITSOIdleActivity records, but our policy is that they can see only their own ITSORentalActivity objects and no other ITSOActivity objects. Furthermore, in some parts of the sample application, we have to supply explicit credentials to authenticate as the customer pseudo-identity. Both of these situations are undesirable as security best practices. A more secure (but, unfortunately, too complex to show in our sample application) implementation would pass all customer access through an EJB method with a configured RunAs role. That is, each EJB method is configured (in the deployment descriptor) to run as the customer pseudo-identity when further J2EE calls are made. This has the advantage of consolidating the logic which enforces additional restrictions on access to objects. We could probably even configure the customer pseudo-identity in the enterprise directory to prohibit logins. EJB RunAs roles is a standard J2EE feature. Actually configuring it has aspects that are specific to application server and is beyond the scope of this book. Nonetheless, it is an important security implementation pattern, and you should consult your application server documentation to learn more about it. Customers do not have records in the the company enterprise directory, whereas all headquarters and franchise employees do. A single customer pseudo-identity is present in the enterprise directory and is used for all access across all franchises. Customer account information is stored as objects in the Content Manager repository. The enterprise directory uses a group structure to organize employees, but it is static and relatively flat. There is a top-level directory group for headquarters and a top-level directory group for each franchise. Within those top-level groups, the enterprise directory does not use further subgroups. Access control security for objects is implemented according to these principals: Customers can see only vehicle information that is intended to be part of Web pages. They cannot see other vehicle information or other business objects. Customers can see all vehicles from all franchises. The application logic allows them to limit or expand their vehicle searches. An employee in one franchise location cannot view the business objects of another franchise location.

Chapter 7. Sample applications for Fictional Auto Rental Company A

261

Different groups of employees may have different types of access to different business objects for the franchise. Appropriate employees of the company headquarters have access to selected business objects for any franchise. Back-end system administrators have access to all objects in the system. All system administrators are part of the company headquarters. Specific mechanisms that are used for security access include direct access control entries (Aces) on objects, standard security inheritance mechanisms (for example, object Aces inherited by an annotation object), and a collection of ITSORole objects acting as security proxies through dynamic security inheritance. The dynamic security inheritance embodied in the ITSORole objects is transitive, so an ITSORole object can itself inherit ACEs from another ITSORole object. Cascading combinations of inheritance can be used to control access to objects. The advantages to this are: ITSORole objects can stand in for a more granular group structure than is available from the enterprise directory. When an employee moves from one role to another (for example, a franchise line employee becoming a franchise supervisory employee), the security for any number of objects may be instantly updated by changing the ACLs on applicable ITSORole objects.

7.6 Workflows
In this section, we describe the workflows implemented for the the company management solution. For a complete business solution, additional elements, such as new workflows, can be identified and implemented across the applications that comprise this solution; however, for the sample application, we provide the following workflow definitions: Vehicle reservation workflow Vehicle maintenance workflow These workflow definitions use a specific isolated region configuration, including a custom roster, event log, and queues. A custom component queue (ITSO_Operations) is also included as part of this configuration to perform certain custom calls within the workflows.

262

Developing Applications with IBM FileNet P8 APIs

7.6.1 Isolated region configuration

An isolated region is a logical subdivision of the workflow database that contains the queues for the work items, event logs, rosters, and other configuration information. It also contains all transferred workflows and running work items. Table 7-12 shows the isolated region configuration implemented for the company management application.
Table 7-12 Isolated region configuration used by the sample application Isolated region element Rosters Event logs Work queues Name ITSORoster ITSOEventLog ITSO_Reservation ITSO_Maintenance ITSO_Malfunction Fields exposed ITSOConfirmationId(30) ITSOVehicleStatus(10) ITSOVehicleId(30) ITSOConfirmationId(30) ITSOVehicleStatus(10) ITSOVehicleId(30) -

Component queues

CE_Operations ITSO_Operations

7.6.2 Component queues

Two Java adaptors are used by the sample application workflows. The following component queues are implemented to service these Java adaptors: CE_Operations Built-in component used to integrate the CE and PE facilitating the management of documents and other objects in an object store from within a workflow. This component is provided, by default. Most of the component operations used by the sample application are included in CE_Operations. Note: The sendMail operation from CE_Operations is commonly used within the sample application workflows to send plain text e-mail notifications. In order for these notifications to work, e-mail settings must be properly configured using the Process Task Manager.

Chapter 7. Sample applications for Fictional Auto Rental Company A

263

ITSO_Operations Implemented to service a custom Java component to perform certain business operations used by the sample application workflows. This component is also provided as a sample to show how to implement a custom Java component by using Component Integrator. Table 7-13 describes the operations included in this custom Java adaptor.
Table 7-13 ITSO_Operations component queue operations Operation name verifyCustomerCredit Description Performs customer credit card validation by calling a third-party system. The operation simulates the call to the external system, always returning a true value. Retrieves and parses the ITSOCustomer document content to extract certain metadata values and assign them to the workflow data fields. This operation demonstrates how to perform a custom call to CE. Uses certain business rules related to a vehicle maintenance activity and launches the vehicle maintenance workflow, passing all required objects. This operation demonstrates how to perform a custom call to PE.

getCustomerDetailsFromOS

sendVehicleToMaintenance

7.6.3 Vehicle reservation workflow

This workflow interacts with both the Reservation Web application and the Kiosk application. It is used to manage the vehicle reservation life cycle, from the moment a vehicle reservation is requested by a customer until the moment the vehicle is returned to the rental agency.

264

Developing Applications with IBM FileNet P8 APIs

Figure 7-23 shows the main diagram that is used by this workflow definition.

Figure 7-23 ITSO vehicle reservation main workflow diagram

Although workflows can be launched by using workflow subscription events, the ITSO vehicle reservation workflow is launched programmatically by the Reservation Web application in order to pass the required objects as attachment references and avoid unnecessary calls within the workflow. The Kiosk application interacts with this workflow when a customer picks up and returns a vehicle. The vehicle reservation workflow definition is comprised of system, component, and work steps. All system and component steps are executed without human intervention to either perform a task within the workflow or call to an external system.

Chapter 7. Sample applications for Fictional Auto Rental Company A

265

Table 7-14 describes the work steps defined in this workflow.

Table 7-14 ITSO vehicle reservation work steps Work step Invalid credit review Work queue ITSO_Reservation Description When the customers credit validation is unsuccessful, the vehicle item is routed to this step for review. At this stage, the reservation can be either cancelled or approved. The vehicle reservation is approved and waiting to be picked up and processed from the kiosk Web application. The vehicle is rented and waiting to be returned by the customer and processed from the kiosk Web application. The malfunction step is to handle unexpected workflow exceptions.

Waiting for pickup

ITSO_Reservation

Out for rental

ITSO_Reservation

Review

ITSO_Malfunction

7.6.4 Vehicle maintenance workflow

This workflow is used to manage the vehicle maintenance life cycle from the Fleet Status Manager Web application.

266

Developing Applications with IBM FileNet P8 APIs

Figure 7-24 shows the main diagram that is used by this workflow definition.

Figure 7-24 ITSO vehicle maintenance main workflow diagram

The vehicle maintenance workflow can be launched either from the vehicle reservation workflow when the vehicle is returned and must be sent to maintenance, or from the Fleet Status Manager Web application when a vehicle is manually sent to maintenance. The Fleet Status Manager Web application is also used to process all vehicles sent to maintenance. Table 7-15 describes the work step defined in the vehicle maintenance workflow definition.
Table 7-15 ITSO vehicle maintenance work steps Work step Maintenance queue Work queue ITSO_Maintenance Description The vehicle is sent to maintenance and waits in the queue until it gets assigned and processed. Malfunction step to handle unexpected workflow exceptions.

Review

ITSO_Malfunction

Chapter 7. Sample applications for Fictional Auto Rental Company A

267

7.7 Internal architecture of sample applications

In this section, we describe the technical architecture of various application components. By technical architecture, we mean a high-level view of the implementation technologies used and how the application components interact.

7.7.1 Architecture: Reservation Web application

For a description of the user view of this application, see 7.3.1, User view: Reservation Web application on page 226. For a description of the deployment instructions for this application, see 7.8.4, Deployment: Reservation Web application on page 292. The Reservation Web application is a J2EE application that is written by using Java Server Faces (JSF). It communicates with the CE (using the CE Java API) and the PE (using the PE Java API). The overall program flow is illustrated in Figure 7-25.

268

Developing Applications with IBM FileNet P8 APIs

Registration Success

Login

Register

Register

No

Failed Registration

Valid User?

Yes

ITSOCarRentalHome

Cars Available

ITSOCarSelectionPage1

Car Selected

ITSOCarRentalConfirm

Rental Success

ITSOVehicleReserve

Figure 7-25 Program flow for Reservation Web application

This application consists of the following Java classes: com.itso.MetadataConstants: The MetadataConstants class contains CE and PE property and class names. itso.autorental.ITSOUtil: This class contains various convenience methods. The intent of these methods is mainly to isolate various common interactions with the CE and PE so that it is not spread throughout the other classes. itso.autorental.process.ITSOAutoRentalWorkflow: This class contains methods related to interaction with the PE API. com.itso.bean package: This package consists of JavaBean classes for data objects.

Chapter 7. Sample applications for Fictional Auto Rental Company A

269

itso.autorental.RentalActivityHandler: Convenience methods including creating and deleting various vehicle activities, as follows: createMaintenanceActivity(): This method takes a vehicle ID and franchise code and creates a maintenance activity. createRentalActivity(): This method creates a rental activity for a confirmed reservation. deleteVehicleActivity(): This method removes a vehicle activity (for example, maintenance activity) given the activity ID. The pagecode package contains the JSF backing classes for each JSP.

7.7.2 Architecture: Kiosk application

For a description of the user view of this application, see 7.3.2, User view: Kiosk application on page 230. For a description of the deployment instructions for this application, see 7.8.5, Deployment: Kiosk application on page 295. The Kiosk application is a Microsoft .NET 2.0 application that is written in C# and communicates directly with the CE (using the CE .NET API) and the PE (using the Process Engine Web Services API). See Figure 7-26 for the application flow diagram.

270

Developing Applications with IBM FileNet P8 APIs

Startup (ApplicationWizard.cs)

Prompt for credentials ApplicationWizard.ApplicationWizard_Shown() P8Prompt.ShowDialog() Connect to CE & PE ApplicationWizard.ConnectToP8() ContentEngineConnector.Connect() ProcessEngineConnector.Connect() Choose Activity: Pickup or Return MainWizardPage Pickup Return Restart

Restart

Prompt For Confirmation Number PickupPage1.cs Failure Lookup Confirmation Number PickupPage1.Lookup() ContentEngineConnector.GetRentalRecord()

Prompt For Vehicle ID ReturnPage1.cs

Complete Return ReturnPage1.CompleteReturn() ProcessEngineConnector.CompleteReturn()

Display Rental Information PickupPage2.cs

Display Final Instructions ReturnCompletedPage.cs

Complete Pickup PickupPage2.PickupPage2_MoveNextWizard() ProcessEngineConnector.CompletePickup()

Display Final Instructions CompletePickupPage.cs

Figure 7-26 Program flow for Kiosk application

The application is responsible for: Pickup Prompting the customer for the rental confirmation ID Finding appropriate rental and displaying rental information to customer Completing the rental by completing the workflow step Return Prompting the customer for the vehicle ID Completing the rental workflow step

Chapter 7. Sample applications for Fictional Auto Rental Company A

271

Content Engine connectivity

The class ContentEngineConnector provides the primary connection to the CE and offers two primary public methods: Connect The Connect method performs the initial connection and authorization to the CE by taking the user name, password, and connection string and retrieving the configured object store. GetRentalRecord The GetRentalRecord method fetches a rental record and all associated information from the CE and returns a RentalRecord object. The RentalRecord object contains the vehicle information (including image) and general information about the rental.

Process Engine connectivity

The class ProcessEngineConnector provides the primary connection to the PE and offers three primary public methods: Connect The Connect method performs the initial connection and authorization to the PE by taking the user name, password, connection point and connection string and executing a getQueues() call to confirm authentication. CompletePickup The CompletePickup method takes the confirmation number, finds the appropriate workflow, and completes that step, moving the workflow on to the next step. CompleteReturn The CompleteReturn method takes the confirmation number, finds the appropriate workflow, and completes that step, moving the workflow on to the next step.

7.7.3 Architecture: Fleet Status Manager Web application

For a description of the user view of this application, see 7.3.4, User view: Fleet Status Manager Web application on page 232. For a description of the deployment instructions for this application, see 7.8.6, Deployment: Fleet Status Manager Web application on page 296.

272

Developing Applications with IBM FileNet P8 APIs

The Fleet Status Manager application is a J2EE application written using Java Server Faces (JSF). It communicates with the CE (using the CE Java API) and the PE (using the PE Java API). Overall program flow is illustrated in Figure 7-27.

login.jsp

Supervisor fleetstatus.jsp Select car in fleet Create work request carinfo.jsp

Login Type

Mechanic

maintqueue.jsp Assign work to my queue

Select work in group queue

groupqueueitem.jsp

Select work in my queue Return work to group queue Complete Work

No

Work complete?

myqueueitem.jsp

Yes

Figure 7-27 Program flow for Fleet Status Manager application

This application consists of the following Java classes: itso.autorental.MetadataConstants: The Constants class contains CE and PE property and class names. itso.autorental.ITSOUtil: This class contains various convenience methods, including: loginToFileNet(): This method establishes connections to CE and PE. loadProperties(): This method loads configuration data from a properties file. getFleetStatus(): This method takes a franchise code and object store and retrieves all cars and their statuses within that franchise. This method returns an array which is used for populating the JSF HtmlDataTable. getQueueWorkItems(): This method retrieves all maintenance activities in a particular franchise code and group queue. This method also returns an array which is used for populating the JSF HtmlDataTable.

Chapter 7. Sample applications for Fictional Auto Rental Company A

273

 assignWorkItemToMyQueue(): This method takes a queue name, wob number, and a user name. It assigns the work item to this users queue. reassignWorkItemToGroupQueue(): This method takes a wob number and reassigns the work item back to the group queue. completeWorkItem(): This method takes a wob number and completes the workflow. itso.autorental.ITSOVehicle: This class represents an Vehicle object for this application. itso.autorental.UserRole: This class represents a user role for the logged in user. User can be a maintenance worker or a supervisor. itso.autorental.RentalActivityHandler: This class has convenience methods, including creating and deleting various vehicle activities: createMaintenanceActivity(): This method takes a vehicle id and franchise code and creates a maintenance activity. deleteVehicleActivity(): This method removes a vehicle activity (for example, maintenance activity) given the activity ID. The itso.autorental.fleetstatusmanager.beans package contains the JSF base classes for each JSP.

7.7.4 Architecture: Billing Report application

For a description of the user view of this application, see 7.3.5, User view: Billing Report application on page 237. For a description of the deployment instructions for this application, see 7.8.7, Deployment: Billing Report application on page 299. The Billing Report application is a stand-alone Java application based on the Standard Widget Toolkit (SWT). It uses real P8 credentials to log in to P8 system. When a user clicks the Search button to execute a query, the system calls the CE API to search the ITSOVehicleActivity objects based on the search criteria provided by the user. The application processes the returned search result to calculate the total hours and total numbers for different vehicle activity types. After the calculation process finishes, the calculated result is converted into a JavaScript file. The application loads an HTML file to render this JavaScript file and displays the rendered result in an embedded browser window.

274

Developing Applications with IBM FileNet P8 APIs

Content Engine connectivity

The classes CEHelper and VehicleHoursReportGenerator provide the primary connection to the CE and offers several primary methods: CEHelper.login(): This method performs the initial connection and authorization to the CE by taking the user name, password, and connection string. CEHelper.validateConnection(): This method validates the connection between the client application and CE. If the connection is not valid, it will throw an exception. CEHelper.getObjectStore(): This method retrieves the object store according to the object store name. VehicleHoursReportGenerator.queryVehicleHoursSummary(): This method fetches ITSOVehicleActivity from the CE and calculates the total hours and total numbers for different vehicle activity types.

Chapter 7. Sample applications for Fictional Auto Rental Company A

275

Figure 7-28 shows the complete application flow diagram.

Figure 7-28 Program flow for Billing Report application

7.8 Deployment instructions for sample applications

In this section, we describe how to download, install, and configure the artifacts that comprise the the company application. The installation and configuration procedures can vary depending on the destination environment. Certain procedures that are described in this section assume that the deployment is performed on a Microsoft Windows-based P8 platform, but should easily be able to adjust for other platforms. Most steps are Java-oriented and platform-independent. In most cases, we use forward-slash (/) notation for paths because it is generally interpreted properly on both Windows and non-Windows platforms. In some cases, the application server is assumed to be WebSphere

276

Developing Applications with IBM FileNet P8 APIs

Application Server 6.1. Because the steps are relatively simple, adjusting for WebLogic or JBoss should be straightforward. The samples are provided in source form, and we also provide prebuilt binaries enabling you to more easily run them before setting up your development environment. The instructions here are focused mainly on how to run the prebuilt binaries, but, where applicable, we offer guidance for using the sources to re-create those binaries. Note: Installing and running these sample applications in a production environment is not advisable. Part of the setup procedure involves introducing changes to the CE and PE servers, and for that reason we strongly recommend using a development environment or other non-critical environment. Getting the sample applications up and running requires the following steps: 1. Download and unpack the samples. The IBM Redbooks publications Web site for this book contains a download link to a file named sg247743-sample.zip. (Refer to Appendix A, Additional material on page 315 for locating the information.) Download that file and extract it in a convenient location on your local system. Depending on the tools you use locally for some of the steps, you might have to ensure that no spaces exist in the path leading to that location. The top-level directory is called sg247743-sample/, and all of our paths in these instructions are relative to that top-level directory. The overall structure is shown graphically in 7.8.1, Application package structure on page 278. 2. Prepare your CE and PE. The sample applications work in coordination with an established environment for the CE (for example, custom classes and properties) and PE (for example, workflow definitions). The downloaded sample applications include artifacts that are used to establish those environments. Specific instructions are provided in 7.8.2, Content Engine artifacts on page 279 and 7.8.3, Process Engine artifacts on page 285. 3. Add P8 and third-party modules. These modules will be added to the environment in locations under sg247743-sample/. Where necessary, specific instructions are provided in sections for specific components. We assume that you have a working P8 installation available locally; P8 components can be copied from that environment. You might also find convenience in running the P8 client installer to obtain a local copy of P8 components. In any case, you should use P8 client-side components that are compatible with the P8 CE and PE servers that you will be using.

Chapter 7. Sample applications for Fictional Auto Rental Company A

277

4. Modify configuration files. The sample applications use configuration files for URIs pointing to the CE and PE, among other things, so that you can use the binary files in the samples. Specific instructions are provided in sections for specific components where necessary. 5. Deploy each sample component. Because we use a variety of technologies to demonstrate various things, there are also a variety of steps needed to install and deploy components of the sample applications.

7.8.1 Application package structure

Figure 7-29 depicts the overall directory structure of the downloadable sample package, and Table 7-16 on page 279 briefly describes several of those specific directories. Additional information is given, where applicable, in the sections for specific sample applications.

Figure 7-29 Directory structure of sg247743-sample

278

Developing Applications with IBM FileNet P8 APIs

Table 7-16 describes the high-level directory structure for the downloaded sample applications.
Table 7-16 The sample application directory structure Directory CEartifacts/ Description Information related to the metadata used in the CE. Directory includes a set of files to be imported into the CE to establish custom classes, custom properties, and some instance data. Process definitions and other configuration items for the PE. Also includes source files and precompiled binaries for the ITSOOperations connector for Component Integrator. Source files and precompiled binaries for most of the Java-based components of the sample applications. For ITSOReservationSystem and ITSOFleetStatusManager directories, the source files are included in the EAR or WAR files, along with the compiled class files. For the other applications, the sources are in an obvious subdirectory. Source files and precompiled binaries for the Windows-only Kiosk application. Prebuilt WAR file for the sample widget. Because the widget is implemented in JavaScript, the WAR file also contains the source files.

PEartifacts/

Java/

Kiosk/ Widget/

7.8.2 Content Engine artifacts

You should have a working P8 system before trying to work with the sample applications. Even so, certain metadata (class and property definitions) and instance data pieces should be added to the CE environment, specifically in support of the sample applications. The sample applications operate with a single ObjectStore of your choosing. Although it is not a requirement, you might want to create an empty ObjectStore just for the purposes of experimenting with the sample applications. Note: Although problems can sometimes occur when importing data into an ObjectStore at a different CE release than the ObjectStore from which it was exported, the constructs that we use in the data files are straightforward. You are unlikely to have any problem if you are using any IBM FileNet Content Manager 4.5.x or IBM FileNet Business Process Manager 4.5.x release. The metadata customization for the sample applications consists of custom subclasses. There are no changes to standard system classes. The custom

Chapter 7. Sample applications for Fictional Auto Rental Company A

279

classes and properties are described in 7.4, Data model on page 239. We also provide sample instance data for various CE classes that are used by the sample applications. Importing this sample data is not strictly necessary, but you would otherwise have to manually create certain data if you want your exploration of the sample applications to be meaningful. Although the structure of our sample data correctly matches the data model description, it is only sparsely populated. Other than in structure, the data is not intended to be particularly realistic. Even after you have imported the sample instance data, you might want to create additional instance data of your own for the purposes of experimenting with the sample applications. You can add this custom metadata and sample instance data to your ObjectStore through the standard import procedure, as follows: 1. Start the IBM FileNet Enterprise Manager application and log on to the P8 domain. 2. Navigate to the ObjectStore you wish to use for the sample applications, right-click on it, and select All Tasks Import All. See Figure 7-30.

280

Developing Applications with IBM FileNet P8 APIs

Figure 7-30 CE import task

3. Enterprise Manager presents an import helper panel. On the Import Options tab (see Figure 7-31), enter the following information: a. In the Import Manifest File field, browse to and select the file:
sg247743-sample/CEartifacts/import/farcasample_CEExport_Manifest.xml

b. Leave the External Content Path field empty. c. For Storage Location for Imported Objects, select Default Database Storage Policy from the drop-down menu. d. For Standard Options, check only Import Object ID and Ignore duplicate error messages during import. Deselect all other boxes. The reason for ignoring duplicates is that a few things in the import data might (or might not) already be present in your ObjectStore. It is also helpful if you have to re-run the import process.

Chapter 7. Sample applications for Fictional Auto Rental Company A

281

Figure 7-31 CE import helper panel Import Options

282

Developing Applications with IBM FileNet P8 APIs

4. On the Import Items tab (in Figure 7-32), check only Property Templates and Class Definitions. Deselect all other boxes. This step causes only metadata objects to be imported. Instance data will be imported in a later step.

Figure 7-32 CE import helper panel Import Items for metadata

5. On the Advanced tab, make sure all available checkboxes are checked. 6. Click the Import button. The import of metadata should proceed without errors or warnings. If problems occur, consult Enterprise Manager documentation and resolve those problems before going further. 7. Repeat these steps, but on the Import Items tab, deselect only Property Templates and Class Definitions check boxes. Check all other boxes. See Figure 7-33. This step causes instance data to be imported. Depending on your specific version of Enterprise Manager, you might have to repeat the import of instance data to resolve all dependencies. Warnings about duplicated items can be ignored, but errors for missing items must be resolved by repeating the import. You are looking for a summary that says the import was successful, not merely partially successful.

Chapter 7. Sample applications for Fictional Auto Rental Company A

283

Figure 7-33 CE import helper panel Import Items for instance data

284

Developing Applications with IBM FileNet P8 APIs

After the instance import, you should see a structure like Figure 7-34 under the /ITSOAutoRental/ folder in your ObjectStore.

Figure 7-34 ITSOAutoRental and subfolders

7.8.3 Process Engine artifacts

The following steps describe how to deploy and configure the PE artifacts required for the sample applications: 1. Ensure that the PE notification settings are properly configured as follows: a. Log onto the Application Server. b. Open Process Task Manager.

Chapter 7. Sample applications for Fictional Auto Rental Company A

285

c. Stop all component manager instances; see Figure 7-35.

Figure 7-35 Stop all component manager instances

d. Ensure that the PE notification settings are properly configured in the Process Task Manager (Figure 7-36). These notification settings are required in order to send e-mail by using the CE_Operations component.

Figure 7-36 Ensure correct notification settings on Process Task Manager

2. Import the ITSOAutoRentalIsolatedRegion.xml file as follows: a. Log onto the Workplace or WorkplaceXT application using an administrator account. b. Open the Process Configuration Console. See Figure 7-37. a. Open a connection on the connection point where the new configuration will be imported. b. Right-click on the selected connection point node and select the Import from XML file option.

286

Developing Applications with IBM FileNet P8 APIs

c. Browse to and select the following file to import: sg247743-sample/PEartifacts/ITSOAutoRentalIsolatedRegion/ITSOAuto RentalIsolatedRegion.xml d. Select Merge as the import type. e. Click the Import button and select Yes from the warning message.

Figure 7-37 Import isolated region configuration

3. Set component queues adaptor security (Figure 7-38) as follows: a. Select the ITSO_Operations component queue from the Component Queues node. b. Open the component properties and navigate to the Adaptor tab. c. Enter a valid user name and password for the JAAS Credentials section and click OK. d. Repeat steps a - c for the CE_Operations component adaptor. e. Commit your changes.

Chapter 7. Sample applications for Fictional Auto Rental Company A

287

Figure 7-38 Set JAAS credentials on component queues

4. Add the workflow definitions (.pep files) to the object store by using the Workplace or WorkplaceXT application as follows: a. Create a folder where the workflow definitions are to be added. For example, ITSOAutoRentalWorkflows. b. Add the following workflow definition in the created folder, specifying the document class as Workflow Definition: sg247743-sample/PEartifacts/ITSOAutoRentalWorkflows/ITSO Vehicle Reservation Workflow.pep c. Right-click on the added workflow definition and select the Transfer Workflow action. d. Leave the default prompted name to transfer the workflow definition and click the Transfer button. e. Repeat steps a to d for the ITSO Vehicle Maintenance Workflow.pep workflow definition.

288

Developing Applications with IBM FileNet P8 APIs

5. Copy the contents of the sg247743-sample/PEartifacts/ITSOOperations/ directory to the Application Engine server and configure general settings as follows: a. Log onto the Application Server and create a custom filesystem folder where the ITSOOperations PE artifacts are to be copied. For example, ITSOAutoRental. b. Copy ITSOOperations.jar and log4j.xml files from the sample to the created folder. c. Open the ITSOOperations.jar file with any compression program (for example, WinZip or the Java jar command). d. Locate and open the ITSOAutoRental.properties file with a text editor, and enter the correct property values for your environment. e. Save and update changes performed to ITSOAutoRental.properties (see Example 7-1).
Example 7-1 ITSOAutoRental.properties sample

##########the company General Settings########## # Object Store Name ITSOObjectStore=DemoOS ##########the company Email Settings############ # Sender Email Address [email protected] # Customer Service Email Address [email protected] # Vehicle Maintenance Manager Email Address [email protected] ##############ITSO Malfunction Settings########################## # Debug malfunction flag (default: false) ITSOMalfunctionDebug=false # Retry malfunction flag (default: false) ITSOMalfunctionRetry=false 6. Configure ITSOOperations custom logging. The log4j.xml file provided with the ITSOOperations package is preconfigured on debug level for the specified packages and creates a daily log file at the following location, by default: C:/itso_autorental_ops.log

Chapter 7. Sample applications for Fictional Auto Rental Company A

289

Logging parameters may be changed within the log4j.xml file according to your environment (Example 7-2). If you are not running on Windows, you should at least change the daily log file name.
Example 7-2 Provided log4j.xml file for the ITSO_Operations component .. <!-- ITSO_Operations Trace Dated Log File --> <appender name="ITSOOpsTraceDatedLogFile" class="org.apache.log4j.DailyRollingFileAppender"> <param name="File" value="c:/itso_autorental_ops.log"/> <param name="DatePattern" value=".dd-MM-yyyy"/> <layout class="org.apache.log4j.PatternLayout"> <param name="ConversionPattern" value="%d %p [%t] %c (%M:%L) %m%n"/> </layout> </appender> ... <!-- ITSO_Operations package logging --> <logger name="itso.autorental.components" additivity="false"> <level value="debug"/> <appender-ref ref="ITSOOpsTraceDatedLogFile"/> <appender-ref ref="console"/> </logger> ...

7. Add ITSOLogin login context to the taskman.login.config file: a. ITSO_Operations component uses a custom login context named ITSOLogin. The following sample file is provided, to show you how this custom login is implemented: sg247743-sample/PEartifacts/ITSOOperations/taskman.login.config.s ample b. Navigate to the taskman.login.config file that is used by the Component Manager. For example, the file might be in this location: E:\Program Files\FileNet\WebClient\Router\taskman.login.config c. Append the ITSOLogin login context, as shown in Example 7-3.
Example 7-3 ITSOLogin login context ITSOLogin { filenet.vw.server.VWLoginModule required; itso.autorental.components.authentication.ITSOLoginModule required debug=true; };

290

Developing Applications with IBM FileNet P8 APIs

8. Add ITSO_Operations required library and configure the JRE parameters: a. Open the Process Task Manager and select the Required Libraries tab on the Component Manager node. See Figure 7-39. b. Add the following file to the list of Component Manager required libraries: sg247743-sample/PEartifacts/ITSOOperations/ITSOOperations.jar c. Click Apply.

Figure 7-39 ITSOOperations.jar required library

d. Select the Advanced tab (Figure 7-40) on the Component Manager node and add the following required JRE parameter, which is the location of the ITSO_Operations log4j.xml file: itsolog4j.configuration For example: -Ditsolog4j.configuration=E:/Program Files/FileNet/ITSOAutoRental/log4j.xml e. Click Apply.

Figure 7-40 ITSO_Operations JRE parameters sample

Chapter 7. Sample applications for Fictional Auto Rental Company A

291

9. Start the Component Manager and verify that all component queues have started, with no errors. See Figure 7-41.

Figure 7-41 Component Manager started with no errors

7.8.4 Deployment: Reservation Web application

For a description of the user view of this application, see 7.3.1, User view: Reservation Web application on page 226. For a description of the internal architecture of this application, see 7.7.1, Architecture: Reservation Web application on page 268. In this section, we give details of preparing and deploying the Reservation Web application. These instructions are very similar to those in 7.8.6, Deployment: Fleet Status Manager Web application on page 296.

EAR file preparation

Preparation steps include adding P8 libraries to the EAR file and making other configuration changes. Because JavaServer Faces (JSF) is directly supported by most application servers, adding JSF-related JAR files to the EAR or WAR might not be necessary. Consult your application server documentation for applicable JSF configuration for a Web application. The preparation steps are as follows: 1. Locate the EAR file: sg247743-sample/Java/ITSOReservationSystem/ITSOCarRentalApp.ear 2. Unpack the EAR file using a tool that understands ZIP format (for example, WinZip or the Java jar command). For the purposes of this explanation, we assume you unpacked it to the ITSOCarRentalApp.ear directory. 3. Navigate to the ITSOFleetStatusManager.ear directory and locate the following WAR file: ITSOCarRentalApp.ear/ITSOCarRentalApp.war

292

Developing Applications with IBM FileNet P8 APIs

4. Unpack the WAR file by using a tool that understands ZIP format (for example, WinZip or the Java jar command). For the purposes of this explanation, we assume you unpacked it to the ITSOCarRentalApp.war directory. 5. Navigate to the ITSOCarRentalApp.war/WEB-INF/classes/ directory. 6. Edit the SampleCarRental.properties file and set each parameter appropriately for your FileNet environment. Because this application runs inside a J2EE application server, the simplest approach is to configure the CE connection URI to use EJB transport. 7. Navigate to the ITSOCarRentalApp.war/WEB-INF/lib directory. You might have to create the lib subdirectory. 8. Add the CE and PE API JAR files from your P8 environment, as described in Chapter 2, Setting up development environments on page 13. Because you are using CE EJB transport, the CE JAR files are limited to: Jace.jar log4j.jar The PE JAR files are: pe.jar pe3pt.jar peResources.jar 9. Re-create ITSOCarRentalApp.war, including the changes to the properties file and the addition of the API JAR files. 10.Re-create ITSOCarRentalApp.ear, including the changes to the ITSOCarRentalApp.war file. 11.Continue with deployment, as described in the Deployment section. In those deployment instructions, we assume that you replaced the original ITSOCarRentalApp.ear file with the modified version in the original location.

Deployment
Details of deploying a Web application vary from application server to application server, and details can be different for different releases of the same application server brand. To simplify the explanation, we give the steps to deploy the Reservation Web application in WebSphere Application Server 6.1, as follows: 1. Make sure that you have performed the preparation steps as described in the EAR file preparation on page 292. 2. Log on to the WebSphere administrative console

Chapter 7. Sample applications for Fictional Auto Rental Company A

293

3. Install the ITSOCarRentalApp.ear file as follows: a. Select Applications Install New Application. b. Click Browse and navigate to the file: sg247743-sample/Java/ITSOReservationSystem/ITSOCarRentalApp.ear c. Click Next. Keep all defaults for each dialog page by clicking Next, and click Finish on the last dialog page. Save all your changes. 4. Configure the class loading and updating information: a. Select Applications Enterprise Applications. b. Select the ITSOCarRentalApp link. c. Select Class loading and update detection. d. For the Class loader order field, select Classes loaded with application class loader first. e. Accept the default value for the Polling interval for updated files field. f. Click OK. Click Save. 5. Configure class loader order information: a. Select Applications Enterprise Applications. b. Select the ITSOCarRentalApp link. c. Click Manage Modules. d. Select the ITSOCarRentalWeb link. e. For the Class loader order field, select Classes loaded with application class loader first. f. Click OK. Click Save. 6. If you are using CE EJB transport, no extra JVM parameters are required. If you are using CEWS transport with a CE release before 4.5.1, set up the JVM parameters: a. Go to the WebSphere administrative console. b. Select Servers Application servers. c. Click on your server (for example, server1). d. Select Java and Process Management Process Definition Java Virtual Machine. e. On the Generic JVM Arguments tab, enter the JVM parameters that are applicable to CEWS transport, as described in Chapter 2, Setting up development environments on page 13. f. Click OK. Click Save to save your changes.

294

Developing Applications with IBM FileNet P8 APIs

7. Start your application in WebSphere: a. Select Applications Enterprise Applications. b. Select the ITSOCarRentalApp check box. c. Click Start. Your application should start properly. 8. Launch your application from a browser: a. Open a Web browser. b. Navigate to: http://<YourAppServer:port>/ITSOCarRentalWeb/Login.faces c. Log in using a customer account. The first time, you have to follow the registration link to create a customer account.

7.8.5 Deployment: Kiosk application

For a description of the user view of this application, see 7.3.2, User view: Kiosk application on page 230. For a description of the internal architecture of this application, see 7.7.2, Architecture: Kiosk application on page 270. Because it is implemented as a .NET application, the Kiosk application only runs on Windows. To deploy and run the Kiosk application: 1. Ensure the Microsoft .NET 2.0 framework is installed. 2. Ensure the Microsoft Web Services Enhancements 3.0 is installed. 3. Ensure the IBM FileNet P8 .NET client is installed or copy FileNet.Api.dll from a Windows CE client environment (for example, where your IBM FileNet Enterprise Manager is installed) to the same directory as the Kiosk application. 4. Change to the directory of the Kiosk application: sg247743-sample/Kiosk/ 5. Run ITSO.Application.exe, either from the command line or by double-clicking it from Windows Explorer. 6. On the initial screen, enter the user name, password, CE connection URI (for example, http://<CEServer:port>/wsi/FNCEWS40MTOM), object store, and Process Engine connection point. 7. Click Connect.

Chapter 7. Sample applications for Fictional Auto Rental Company A

295

7.8.6 Deployment: Fleet Status Manager Web application

For a description of the user view of this application, see 7.3.4, User view: Fleet Status Manager Web application on page 232. For a description of the internal architecture of this application, see 7.7.3, Architecture: Fleet Status Manager Web application on page 272. In this section, we give details of preparing and deploying the Fleet Status Manager Web application. These instructions are very similar to those in 7.8.4, Deployment: Reservation Web application on page 292.

EAR file preparation

Preparation steps include adding P8 libraries to the EAR file and making other configuration changes. Because JavaServer Faces (JSF) is directly supported by most application servers, adding JSF-related JAR files to the EAR or WAR might not be necessary. Consult your application server documentation for applicable JSF configuration for a Web application. The preparation steps are as follows: 1. Locate the EAR file:
sg247743-sample/Java/ITSOFleetStatusManager/ITSOFleetStatusManagerApp.ear

2. Unpack the EAR file using a tool that understands ZIP format (for example, WinZip or the Java jar command). For the purposes of this explanation, we assume you unpacked it to the following directory: ITSOFleetStatusManagerApp.ear 3. Navigate to the ITSOFleetStatusManager.ear directory and locate the following WAR file: ITSOFleetStatusManager.ear/ITSOFleetStatusManager.war 4. Unpack the WAR file by using a tool that understands ZIP format (for example, WinZip or the Java jar command). For the purposes of this explanation, we assume you unpacked it to the following directory: ITSOFleetStatusManagerApp.war 5. Navigate to the ITSOFleetStatusManager.war/WEB-INF/classes/ directory. 6. Edit the SampleCarRental.properties file and set each parameter appropriately for your FileNet environment. Because this application runs inside a J2EE application server, the simplest approach is to configure the CE connection URI to use EJB transport. 7. Navigate to ITSOFleetStatusManager.war/WEB-INF/lib directory. You might have to create the lib subdirectory.

296

Developing Applications with IBM FileNet P8 APIs

8. Add the CE and PE API JAR files from your P8 environment, as described in Chapter 2, Setting up development environments on page 13. Because you are using CE EJB transport, the CE JAR files are limited to: Jace.jar log4j.jar The PE JAR file are: pe.jar pe3pt.jar peResources.jar 9. Re-create the ITSOFleetStatusManagerApp.war file, including the changes to the properties file and the addition of the API JAR files. 10.Re-create the ITSOFleetStatusManagerApp.ear file, including the changes to ITSOFleetStatusManagerApp.war file. 11.Continue with deployment, as described in the section Deployment. In those deployment instructions, we assume that you replaced the original ITSOFleetStatusManagerApp.ear with the modified version in the original location.

Deployment
Details of deploying a Web application vary from application server to application server, and details can be different for different releases of the same application server brand. To simplify the explanation, we give the steps to deploy the FleetStatusManager Web application in WebSphere Application Server 6.1. 1. Make sure that you have performed the preparation steps as described in the section EAR file preparation on page 296. 2. Log on to the WebSphere administrative console 3. Install the ITSOFleetStatusManager.ear file: a. Select Applications Install New Application. b. Click Browse and navigate to the file:
sg247743-sample/Java/ITSOFleetStatusManager/ITSOFleetStatusManager.ear

c. Click Next. Keep all defaults for each dialog page by clicking Next, and click Finish on the last dialog page. Save all your changes. 4. Configure the class loading and updating information: a. Select Applications Enterprise Applications. b. Select the ITSOFleetStatusManagerApp link. c. Select Class loading and update detection.

Chapter 7. Sample applications for Fictional Auto Rental Company A

297

d. For the Class loader order field, select Classes loaded with application class loader first. e. Accept the default value for Polling interval for updated files field. Click OK. Click Save. 5. Configure class loader order information: a. Select Applications Enterprise Applications. b. Select the ITSOFleetStatusManagerApp link. c. Click Manage Modules. d. Select the ITSOFleetStatusManager link. e. For the Class loader order field, select Classes loaded with application class loader first. f. Click OK. Click Save. 6. If you are using CE EJB transport, no extra JVM parameters are required. If you are using CEWS transport with a CE release before 4.5.1, set up the JVM parameters: a. Go to the WebSphere administrative console. b. Select Server Application servers. c. Click on your server (for example, server1). d. Select Java and Process Management Process Definition Java Virtual Machine. e. On the Generic JVM Arguments tab, enter the JVM parameters that are applicable to CEWS transport, as described in Chapter 2, Setting up development environments on page 13. f. Click OK. Click Save to save your changes. 7. Start your application in WebSphere: a. Select Applications Enterprise Applications. b. Select the ITSOFleetStatusManager check box. c. Click Start. Your application starts. 8. Launch your application from a browser: a. Open a Web browser. b. Navigate to: http://<YourAppServer:port>/ITSOFleetStatusManager/login.faces c. Log in by using a valid LDAP account.

298

Developing Applications with IBM FileNet P8 APIs

7.8.7 Deployment: Billing Report application

For a description of the user view of this application, see 7.3.5, User view: Billing Report application on page 237. For a description of the internal architecture of this application, see 7.7.4, Architecture: Billing Report application on page 274. The Billing Report is a thick desktop Java application with a graphical user interface (GUI). It can be run from the command line (which we describe briefly); it can also be run from within the Eclipse IDE, which we describe in detail. Eclipse is a popular integrated development environment for Java and other application types. It is available from: http://www.eclipse.org These deployment instructions were developed by using Eclipse 3.5, but any recent Eclipse release is similar. Both the command line and the Eclipse IDE methods share a common directory structure, and for both methods, you will need a JAR file from Eclipse Standard Widget Toolkit (SWT). Download the SWT package from the following location, and select the version for your operating system: http://www.eclipse.org/swt/ Although the Billing Report application was developed with SWT 3.4, it is not especially sensitive to the SWT version, so any recent version should work. Save the .zip file in a convenient location in your environment.

Preparing the application environment

The first steps for deploying the Billing Report application are to add P8 and third-party components to the environment, as follows: 1. Change directory to:
sg247743-sample/Java/ITSOBillingReport/BillingReportApplication/deploy/

2. Extract the swt.jar file (from the downloaded SWT ZIP archive) into the lib/ subdirectory. 3. Copy CE client JAR files to the lib/ subdirectory. JAR files include Jace.jar, log4j.jar, and other transport-dependent JAR files. For our sample setup, we use CEWS transport, assuming a P8 4.5.1 environment. Therefore, copy the applicable JAR files into the lib/ subdirectory. JAR files include stax-api.jar, xlxpScanner.jar, and xlxpScannerUtils.jar file. If you want to use EJB transport or use CEWS transport with a pre-4.5.1 environment, consult general setup instructions for those environments in Chapter 2, Setting up development environments on page 13.

Chapter 7. Sample applications for Fictional Auto Rental Company A

299

Running from the command line

To run the application from the command line: 1. If you are not already there, change directory to:
sg247743-sample/Java/ITSOBillingReport/BillingReportApplication/deploy/

2. Edit file billing_wsi.bat file. Ensure that the paths and environment variable values are correct for your environment. The script assumes that your current directory is the deploy/ directory and that dependent JAR files are located in the lib/ subdirectory. At a minimum, you should supply appropriate values for an ObjectStore name and the CE connection URI. 3. Run the billing_wsi.bat file. Further details of the running application are given in 7.3.5, User view: Billing Report application on page 237.

Importing into Eclipse

As a preparatory step for deploying the Billing Report application, first import the Billing Report application as a project into Eclipse. 1. Run Eclipse. If you have multiple workspaces, select the workspace where you want to place the Billing Report application. 2. Select File Import. 3. In the pop-up dialog, select Existing Projects into Workspace. 4. In the next window, select Select root directory, and browse to the following directory: sg247743-sample/Java/ITSOBillingReport/BillingReportApplication/ If you were to visit that directory outside of eclipse, you would see that it contains files .project and .classpath to provide project and dependency data to Eclipse. 5. Make sure that the check box Copy project into workspace is not checked. 6. Click Finish to import the application. See Figure 7-42.

300

Developing Applications with IBM FileNet P8 APIs

Figure 7-42 Importing the BillingReportApplication project into Eclipse

If you have chosen to follow our example and prepared the deploy/lib/ subdirectory with CEWS transport JARs for a P8 4.5.1 environment, all dependencies in the Eclipse project will probably be resolved automatically. Otherwise, you will have to make adjustments to the Eclipse build path for the project. We do not describe that here. Your view of the BillingReportApplication project in the Eclipse project explorer should look like Figure 7-43. Depending on your exact Eclipse version and your personal preference settings, the view could be somewhat different.

Chapter 7. Sample applications for Fictional Auto Rental Company A

301

Figure 7-43 BillingReportApplication structure in Eclipse

The BillingReportApplication project in Eclipse is set up to find source files in the src/ directory and compile them into the deploy/classes/ directory (which does not ordinarily show up in the Eclipse project explorer view). Dependent JAR files are found in the deploy/lib/ subdirectory. From the project explorer view, you can examine the source files at your leisure.

Running from Eclipse

To run the application from Eclipse: 1. Create a new run configuration in Eclipse. There are a few ways to do this, but perhaps the simplest way is to right-click on ReportApplication.java and select Run as Java application. See Figure 7-44. The application either fails immediately or the GUI presents you with a login window. If the latter happens, exit the application. Even if the application does not launch correctly, it still creates a configuration that has many correct settings that you can use later.

302

Developing Applications with IBM FileNet P8 APIs

Figure 7-44 Run As Java Application

2. From the Eclipse menu bar, select Run Run Configurations. 3. The ReportApplication project should already be selected. If it is not, navigate to it. On the Main tab, the following Main class should already be filled in: itso.autorental.billingreport.ReportApplication If it is not filled in, enter that value. 4. Click the Arguments tab and provide the VM arguments setting, as shown in Figure 7-45. Be careful to put these in the VM arguments box and not the Program arguments box. Notice that these arguments are very similar to those used in the BAT file for the command line procedure in Running from the command line on page 300. They include (with values that must be adjusted for your environment): -Ddefault.objectstore.name=MyObjectStore -Djaas.stanza=FileNetP8WSI -Dce.uri=http://server.example.com:9080/wsi/FNCEWS40MTOM -Djava.security.auth.login.config="jaas.conf.WSI" 5. In the Working directory subpanel, select Other, click the Workspace button, and navigate to the BillingReportApplication/deploy/ directory. This is the same working directory we used when running from the command line.

Chapter 7. Sample applications for Fictional Auto Rental Company A

303

Figure 7-45 Billing application JVM requirements

6. Click Apply and click Close. You now have the correct run configuration for the Billing Report application. 7. Back in the Eclipse project, right-click the ReportApplication.java and select Run as Java application to run the Billing Report application. You should now see the main window of the Billing Report application. Further details of the running application are given in 7.3.5, User view: Billing Report application on page 237.

304

Developing Applications with IBM FileNet P8 APIs

Chapter 8.

Logging and troubleshooting

In this chapter, we discuss how to enable and interpret the various logging options for the IBM FileNet P8 Platform application programming interfaces (APIs). We also direct you to several technical troubleshooting articles and provide important data gathering points when troubleshooting IBM FileNet programming issues. This chapter discusses logging only from the perspective of the client-side API components. The topic of server-side logs that are generated by Content Engine (CE) and Process Engine (PE) is a larger topic for overall troubleshooting and is not within the scope of this book. This chapter discusses the following topics: How to enable logging for the various IBM FileNet APIs How to read the IBM FileNet API log files Pointers to technical articles for troubleshooting Data to gather when reporting IBM FileNet API issues

Copyright IBM Corp. 2009. All rights reserved.

305

8.1 Logging
In this section, we describe how to enable logging for various FileNet APIs, including CE and PE Java API, CE .NET API, and Content Engine Web Services (CEWS). We also present several examples of interpreting the client-side API log files.

8.1.1 CE Java API

The CE Java API uses Apache log4j for a configurable set of logging options. You can obtain details of log4j configuration and options on Apaches log4j Web site: http://logging.apache.org/log4j

Setting up the CE Java API log4j environment

You can enable logging in the CE Java API in more than one way. One way is mentioned in FileNet ECM online help documentation. The documentation mentions packaging a log4j.properties file into a custom JAR file and deploying it to the applications classpath. Here are those steps: 1. Make a copy of the following file and copy it to the client application server: FileNet\ContentEngine\config\samples\log4j.properties.client 2. Create an empty JAR file with an easily identifiable name such as debugCElog4j.jar. Making this new JAR enables you to perform a simple process later to disable CE Java API logging by simply removing this debugCElog4j.jar file. 3. Remove the .client extension from the log4j.properties.client file. Edit this log4j.properties file, and add it to the root directory of the debugCElog4j.jar file. We discuss the specifics of modifying the log4j.properties file in Modifying the CE log4j.properties file on page 307. 4. Copy the debugCElog4j.jar file to either the applications WEB-INF\lib folder or a specified location in the JVM CLASSPATH parameter. 5. Restart the application. Executing the CE Java API code generates two files: p8_api_trace.log p8_api_error.log. Another way to enable log4j logging is to configure a JVM argument that specifies the location of the log4j.properties file (on MicroSoft Windows): -Dlog4j.configuration=file:c:\<YourPath>\log4j.properties

306

Developing Applications with IBM FileNet P8 APIs

The benefit of this JVM argument method is the ability to change log4j log levels dynamically by simply modifying the log4j.properties file and does not require restarting the JVM.

Modifying the CE log4j.properties file

The CE log4j.properties file contains subcomponents of the tracing API that can be configured. If you are interested only in enabling full CE Java API tracing, follow these steps: 1. Uncomment the following line to enable the client API error logger: #log4j.logger.filenet_error.api = warn 2. Change the following line from off to debug: #log4j.logger.filenet_tracing = off, FileNetTraceRollingAppender log4j.logger.filenet_tracing = debug, FileNetTraceRollingAppender 3. Uncomment the following line for filenet_trace.api logging: log4j.logger.filenet_tracing.api = debug More options for modifying the log4j.properties file can be found by selecting: ECM_help Developer Help Content Engine Development Java and .NET Developers Guide Trace Logging Working with the Apache log4j Configuration File Sample log4j Configuration Files As mentioned in the previous section, you can disable tracing dynamically if you use the JVM argument method, which removes the necessity to restart the JVM.

Log file size and rollover

By default, the various log4j appenders in the log4j.properties file have the following file size and file count settings: MaxFileSize=100MB MaxBackupIndex=1 These settings specify that only one log file will be generated and overwritten when the log file size exceeds 100 MB. It might be necessary to increase the MaxFileSize and MaxBackupIndex values so that logs are not being overwritten, resulting in lost data. We recommend that you increase the MaxBackupIndex when tracing data over a lengthy amount of time.

Reading the p8_api_trace.log file

Reading the p8_api_trace.log file takes some experience and familiarity before fully understanding these logs. We provide several tips to get you started.

Chapter 8. Logging and troubleshooting

307

CE Java API version

Example 8-1 displays the build number for the client CE Java API library. This build number can be cross referenced against the P8 Patch Dependency spreadsheet found on the IBM FileNet P8 support site.
Example 8-1 BuildVersion output in CE Java API trace file

DEBUG [WebContainer : 1] - Configuration for:BuildVersion value:dap440.247 applied This BuildVersion is useful to determine the exact CE API patch level you are using, and which helps when you are trying to identify a particular patch level that you must be on for a particular bug fix. This BuildVersion is also useful in determining whether your client CE APIs are of the same version as your CE server. CE server also outputs the build number when logging is enabled. You also see a message similar to Example 8-2 in your CE servers p8_server_trace.log file if the client and servers build versions do not match. The local version is always the version (from the point of view of the component) that is producing the log line.
Example 8-2 Mismatch build version message in CE Server trace file

2008-04-24T20:56:49.314Z B129CD3B API - The transport build version does not match the local build version: transport=dap000.441, local=dap435.029

Thread Identifiers
Because client applications are likely to be multithreaded, multi-user environments, the CE Java API logs can become inundated with data across various threads and users. To follow the log entries for a particular thread, look at the thread ID, such as thread ID 3 in Example 8-3. The thread ID, which appears in all log lines, is the part inside [square brackets]. Formats of thread IDs vary from one application server to another.
Example 8-3 Thread identifier

DEBUG [WebContainer : 3] - UserContext.set() invoked in its creator Thread <885535944> for UserContext <null>.

Request / Response
The request and data that are sent to the CE server, as well as the resulting response, can be useful when troubleshooting issues. Example 8-4 on page 309 shows an example of an outgoing executeSearchRequest call and Example 8-5 on page 309 shows the corresponding result. The format of the logged data is

308

Developing Applications with IBM FileNet P8 APIs

logically similar to, but not the same as, the corresponding CEWS data on the wire. When similar data is logged on the CE server, the hard line breaks are replaced with the character sequence \n so that logging stays on a single line.
Example 8-4 executeSearchRequest call

2009-03-25 14:08:16,078 DEBUG [WebContainer : 3] <executeSearchRequest clientCaller="com.filenet.ae.bean.datasource.CeQuery.executeQuery(CeQue ry.java:191)" maxElements="200" continueable="true" rid="1"> <searchMode type="SearchMode" value="1" rid="2"/> <searchScope type="ObjectStoreScope" rid="3"> <objectStore type="GlobalIdentity" classIdentity="ObjectStore" identity="{5B4ADE2D-2FCC-4D8E-AFFC-B406A7123235}" rid="4"/> </searchScope> <search type="RepositorySearch" searchSQL="SELECT [VersionStatus], [IsReserved], [Id], [MajorVersionNumber], [DateLastModified], [DocumentTitle], [MinorVersionNumber], [ContentSize], [CompoundDocumentState], [ClassDescription], [IsCurrentVersion], [VersionSeries], [LastModifier], [MimeType] FROM Document WHERE THIS INFOLDER '{4DC50869-7E4E-4EFF-9CEA-0D413B0E48EB}' ORDER BY DocumentTitle" rid="5"> <propertyFilter type="PropertyFilter" rid="6"> <includePropertyList size="0" rid="7"/> <includeTypeList size="0" rid="8"/> <excludePropertyList size="0" rid="9"/> </propertyFilter> </search> </executeSearchRequest>
Example 8-5 Corresponding executeSearchResponse

2009-03-25 14:08:16,078 DEBUG [WebContainer : 3] <executeSearchResponse elapsed="0" rid="1"> <value type="RepositoryRowCollection" stale="false" rid="2"> <list size="2" rid="3"> <entry type="RepositoryRow" rid="4"> <properties type="RepositoryRowProperties" rid="5"> <property type="PropertyInteger32" name="VersionStatus" dirty="false" access="1" rid="6"> <value>1</value> </property> <property type="PropertyBoolean" name="IsReserved" dirty="false" access="1" rid="7"> <value>false</value> </property>

Chapter 8. Logging and troubleshooting

309

<property type="PropertyId" name="Id" dirty="false" access="1" rid="8"> <value type="Id">{C9712786-4B17-4512-9E13-9F4154B35FC2}</value> </property> <property type="PropertyInteger32" name="MajorVersionNumber" dirty="false" access="1" rid="9"> <value>1</value> </property> <property type="PropertyDateTime" name="DateLastModified" dirty="false" access="1" rid="10"> <value>2009-03-19T20:16:31.406Z</value> </property> <!-- ... portions omitted ... --> <paging type="DefaultPaging" rid="44"> <connection type="Connection" uri="iiop://hqdemo1:2810/FileNet/Engine" asURI="iiop://hqdemo1:2810/FileNet/Engine" rid="45"> <parameters type="ConfigurationParameters" rid="46"> <values size="0" rid="47"/> </parameters> </connection> </paging> </value>
</executeSearchResponse>

8.1.2 CE .NET API

As of the current P8 4.5 release, the CE .NET API does not have client-side logging. You can enable CE Server logging in FileNet Enterprise Manager to trace the requests from the client. Because the .NET API always uses CEWS transport, you can also use network tracing to capture the actual requests and responses.

8.1.3 Content Engine Web Services

CEWS clients are stub classes that are generated from the server WSDLs. As such, these stub classes do not contain client-side logging unless it is provided by the client Web services toolkit that you are using. You can enable CE Server logging in FileNet Enterprise Manager to trace the requests from the client. You can also use network tracing to capture the actual requests and responses.

310

Developing Applications with IBM FileNet P8 APIs

8.1.4 PE Java API

The PE Java API also uses log4j for a configurable set of logging options.

Setting up the PE Java API log4j environment

To enable the PE Java API, follow these steps: 1. Make a copy of the following file and copy it to the client application servers JRE lib folder: \fnsw_loc\sd\fnlog4j.properties.sample 2. Remove the .sample extension from the fnlog4j.properties.sample file. Edit this fnlog4j.properties file. 3. Modify in the fnlog4j.properties file, which is discussed in the next section (Modifying the fnlog4j.properties file).

Modifying the fnlog4j.properties file

The PE fnlog4j.properties file contains various subcomponents that can be configured. If you are interested in enabling full PE Java API client tracing, set the loggers to DEBUG mode in the PE fnlog4j.properties file: log4j.debug=true log4j.logger.filenet.vw=DEBUG, CON, TXT log4j.logger.filenet.pe=DEBUG, CON, TXT You can disable full PE Java API tracing dynamically by changing the following loggers back to INFO mode in the file: log4j.logger.filenet.vw=INFO, CON, TXT log4j.logger.filenet.pe=INFO, CON, TXT

8.1.5 Process Engine Web Services

Process Engine Web Services (PEWS) clients are stub classes that are generated from the server WSDLs. As such, these stub classes do not contain any client side logging unless it is provided by the client Web services toolkit you are using. You can enable the PEWS listener on the CE and PE server by setting the following logger: log4j.logger.filenet.ws=DEBUG,CON,TXT

Chapter 8. Logging and troubleshooting

311

8.1.6 PE REST Service

You can enable the REST Service trace logging on the Application Engine and WorkplaceXT server by setting the loggers as shown in Example 8-6.
Example 8-6 Enable PE REST Service tracing # for tracking REST accesses-----------------------------------------log4j.appender.REST=org.apache.log4j.DailyRollingFileAppender log4j.appender.REST.layout=org.apache.log4j.PatternLayout log4j.appender.REST.layout.ConversionPattern=%d{yyyy/MM/dd HH:mm:ss} [%t] %m%n log4j.appender.REST.File=c:\\RESTLog.txt log4j.appender.REST.Append=true # ---------- filenet.pe.rest.* loggers--------log4j.logger.filenet.pe.rest=ERROR,TXT #-------------------------------------------------log4j.logger.filenet.pe.rest.servlet=ERROR log4j.logger.filenet.pe.rest.request=ERROR log4j.logger.filenet.pe.rest.response=ERROR log4j.logger.filenet.pe.rest.handler=ERROR log4j.logger.filenet.pe.rest.utils=ERROR

8.2 Troubleshooting
In this section, we describe troubleshooting techniques for CE and PE. We also discuss the data that you must gather so that the IBM support team can help you with troubleshooting.

8.2.1 log4j debugging

If client-side API logging is enabled using the steps in 8.1, Logging on page 306, but no logs are generated, these additional steps can help further isolate the issue. When no logging is configured for the CE Java API, a warning message is written to the System.err file. See Example 8-7.
Example 8-7 Message when log4j is not configured log4j:WARN The log4j system is not properly configured! log4j:WARN All ERROR messages will be sent to the system console until proper configuration has been detected

312

Developing Applications with IBM FileNet P8 APIs

If you do not see these lines, most likely log4j is configured but the log4j.properties file is being picked up in a different location. This can occur when a custom application uses its own packaging of log4j. To determine that information, you can enable log4js debug mode by setting the following JVM parameter: -Dlog4j.debug=true This setting generates output in System.out similar to Example 8-8.
Example 8-8 log4j debug output log4j: Trying to find [log4j.xml] using context classloader sun.misc.Launcher$AppClassLoader@a39137. log4j: Trying to find [log4j.xml] using sun.misc.Launcher$AppClassLoader@a39137 class loader. log4j: Trying to find [log4j.xml] using ClassLoader.getSystemResource(). log4j: Trying to find [log4j.properties] using context classloader sun.misc.Launcher$AppClassLoader@a39137. log4j: Using URL [file:/C:/classpath/log4j.properties] for automatic log4j configuration. log4j: Reading configuration from URL file:/C:/whatever/log4j.properties ...

8.2.2 Content Engine troubleshooting techniques

The following technical articles provide tips about programming with FileNet P8 CE APIs and troubleshooting techniques: Writing great code with the IBM FileNet P8 APIs, Part 1: Hello, Document! Getting started with your first FileNet P8 program
http://www.ibm.com/developerworks/data/library/techarticle/dm-0810carpenter

Writing great code with the IBM FileNet P8 APIs, Part 2: Spying on your handiwork Techniques for seeing what you're putting on the wire
http://www.ibm.com/developerworks/data/library/techarticle/dm-0902carpenter

8.2.3 Process Engine troubleshooting techniques

The following article discusses various techniques to troubleshoot PE to CE connectivity issues: http://www.ibm.com/support/docview.wss?uid=swg21328045

Chapter 8. Logging and troubleshooting

313

8.2.4 Data must be gathered for troubleshooting

When you report FileNet programming issues to IBM FileNet support, obtain the necessary information as described in the following technical articles: Content Engine 4.5 http://www.ibm.com/support/docview.wss?uid=swg21308231 Process Engine 4.5 http://www.ibm.com/support/docview.wss?uid=swg21327304

314

Developing Applications with IBM FileNet P8 APIs

Appendix A.

Additional material
This book refers to additional material that can be downloaded from the Internet as described below.

Locating the Web material

The Web material associated with this book is available in softcopy on the Internet from the IBM Redbooks Web server. Point your Web browser at: ftp://www.redbooks.ibm.com/redbooks/SG247743 Alternatively, you can go to the IBM Redbooks Web site at: ibm.com/redbooks Select the Additional materials (and then Access additional materials) and open the directory that corresponds with the IBM Redbooks form number, SG247743.

Copyright IBM Corp. 2009. All rights reserved.

315

Using the Web material

The additional Web material that accompanies this book includes the following files: File name SG247743-sample.zip Description Code for the sample application used in the book (the file is compressed in .zip file format)

System requirements for downloading the Web material

The following system configuration is recommended: Hard disk space: Operating System: Processor: Memory: 10 GB minimum Microsoft Windows XP Intel Core Duo processor 1 GB minimum

How to use the Web material

Create a subdirectory (folder) on your workstation, and extract the contents of the Web material .zip file into this folder.

316

Developing Applications with IBM FileNet P8 APIs

Related publications
The publications listed in this section are considered particularly suitable for a more detailed discussion of the topics covered in this book.

IBM Redbooks
For information about ordering these publications, see How to get Redbooks on page 318. Note that some of the documents referenced here may be available in softcopy only. IBM FileNet Content Manager Implementation Best Practices and Recommendations, SG24-7547 Introducing IBM FileNet Business Process Manager, SG24-7509 IBM FileNet P8 Platform and Architecture, SG24-7667 Understanding IBM FileNet Records Manager, SG24-7623 IBM Content Manager OnDemand Web Enablement Kit Java APIs: The Basics and Beyond, SG24-7646

Online resources
These Web sites are also relevant as further information sources: IBM FileNet P8 Platform main information page
http://www.ibm.com/software/data/content-management/filenet-p8-platform

IBM FileNet P8 Platform product documentation http://www.ibm.com/support/docview.wss?rs=3247&uid=swg27010422 The above URL includes links to all expansion IBM FileNet P8 products. IBM FileNet Content Manager
http://www.ibm.com/software/data/content-management/filenet-content-manager

Copyright IBM Corp. 2009. All rights reserved.

317

IBM FileNet Business Process Manager http://www.ibm.com/software/data/content-management/filenet-business -process-manager IBM InfoSphere Enterprise Records http://www.ibm.com/software/data/content-management/filenet-recordsmanager/

How to get Redbooks

You can search for, view, or download Redbooks, Redpapers, Technotes, draft publications and Additional materials, as well as order hardcopy Redbooks publications, at this Web site: ibm.com/redbooks

Help from IBM

IBM Support and downloads ibm.com/support IBM Global Services ibm.com/services

318

Developing Applications with IBM FileNet P8 APIs

Index
Symbols
.NET environment setup 29 .NET API 34, 40, 48, 55, 270, 306, 310 126, 132, 188189, 192 Java Compatibility Layer, Content Engine (CE) 6 Java, Content Engine 14 Java, IBM InfoSphere Enterprise Records 8 Java, Process Engine 7, 114, 121, 124 log file, IBM FileNet 305 PE API 269, 293, 297 PE API classes, functional relationship 118 PE API, core classes 116 PE error handling class 117 PE REST API 112113, 140, 144, 150 PE REST API, scenario 188 PE REST API, service 207, 215 PE retrieve work status classes 117 PE search work classes 117 PEWS API 114 Process Engine (PE) 114115, 119 REST API 7, 112, 114, 123, 188, 190, 193 REST Services, Process Engine 7 workflow definition related 113 API set 4, 67 Application Engine (AE) 2, 9, 100, 168, 312 application environment 299 application programming interface See API application server 10, 4445, 48, 119120, 123, 127, 142, 146, 176, 188, 191, 276, 285, 289, 306, 308, 311 container 154, 160, 193 documentation 261, 292, 296 application space 187 architecture ECM Widgets 205 sample applications 268 association property 94 Attachment widget 202, 205, 208 authentication in .NET 45 in Java 44 PE 120 authorization 88 accessing REST operations 123

A
access 88 access control entry 88, 108, 257 access control list (ACL) 88, 258 access permission manipulation in C# 89 manipulation in Java 88 accessing Document content in C# 63 in Java 62 AccessPermission 88 AccessRight.DELETE permission 8889 active content 102 Admin package CE API 41 administration API API administration related 113 advanced document recognition (ADR) 9 annotation 99100 annotations property 94 API 1, 34, 13, 3941, 80, 88, 94, 96, 111114 .NET 34, 40, 48, 55, 270 .NET API 306, 310 CE .NET API, installing 30 CE API, patch level, 308 CE Java 268, 273 CE Java API 306308 CE Java API code 306 CE Java API log 308 CE Java API log4j environment 306 CE Java API trace file 308 COM Compatibility Layer 67 configuration 113 eForms 8 exception 123 Java 34, 67, 40, 44, 48, 104, 112114, 124,

Copyright IBM Corp. 2009. All rights reserved.

319

B
batch refresh versus no refresh 81 retrieval 81 retrieving items in batch 83 transactional operation 81 updates 81 updating objects in batch 82 Billing Report application 224, 237239 architecture 274 user view 237 binary 279 BOOTSTRAP_ADDRESS 17 building SQL using IBM FileNet Enterprise Manager 72

C
cache, metadata 106 callback 103, 127128, 142143, 179, 189, 191, 194 cascading style sheet (CSS) 97, 211 CE 25, 13, 3940, 72, 87, 119, 121, 168, 176, 239, 263264, 272, 275, 305 API class model 40 gather information for troubleshooting 314 Java API 4, 7, 14 Java Compatibility Layer API 6 Query Builder 73 sample Java application setup 22 URI 120, 125 Web service 5 WSDL 5 CE .NET API installing 30 CE API class model 40 Core package 41 Events package 42 patch level 308 type 40 CE Java API 268, 273, 306308 code 306 log 308 log4j environment 306 trace file 308 CE_Operations 176, 263 CEWS 31, 306 CEWS transport 6, 14, 21, 23, 4546, 294,

298299, 310 requirements 1819 CFS-IS 9, 100 check in document 90 document in C# 93 document in Java 92 check out document in C# 93 document in Java 92 class model CE API 40 client thick versus thin 15 Collection package CE API 42 collection properties 54 COM Compatibility Layer API 67 programming 7 component association 171 document 97 operation 171 package 172 Component Manager 176, 184186, 286, 290291 configuration 185 parameters 185 component queue 168, 170171, 182, 184, 263, 287288 configuring and testing 182 programmatic creation 182 ComponentRelationship 97 compound document manipulation 97 configurable set 306, 311 configuration API 113 connection 43 initial connection 43 initial connection in C# 49 initial connection with EJB transport in Java 47 initial connection with Web services transport in Java 47 making initial connection 44 connection point 119 connectivity troubleshooting 21 containment relationship 95 CONTAINS 76

320

Developing Applications with IBM FileNet P8 APIs

content active content 102 retrieving 6162 content element 6163, 65, 99, 247 metadata items 62 volatile position number 99 working with, in C# 65 working with, in Java 63 Content Engine See CE Content Engine (CE) troubleshooting 313 Content Engine (CE) API Admin package 41 Collection package 42 Meta package 41 Property package 42 Query package 42 Security package 41 Content Engine Web Services See CEWS Content Federation Services for Image Services See CFS-IS content repository IBM FileNet 8 content search 76 content size 62 content type 62 content-based retrieval (CBR) 76 ContentElement using 63 contents.iterator 173 ContentUrl 79 context root deployed Web application 218 convenience method 44, 65, 269, 273 convertMethod() 116 Core package CE API 41 corresponding ITSORole object user populations 260 create instance 82 createMethod() 116 creating object in C# 53 in Java 53 custom class 181 custom database 172 custom inheritance 110

Custom object 44, 52, 58, 94, 149, 169170 properties collection 57 custom widget building 206

D
Data Access Object (DAO) 172 data type 105, 144, 149150, 168 definition 106 deleting object 60 in C# 61 in Java 60 Demo_Java.zip 23 deployed Web application context root 218 deploying widget 217 deployment, sample applications 276 description 106 development setup for PE REST API sample code 33 setup, ECM Widgets 38 development environment setup,Java 13 discover metadata at runtime 105 DNS 17 document check in and out in C# 93 check in and out in Java 92 checked-in 90 compound document, manipulation 97 filing in a folder in C# 97 filing in a folder in Java 96 retrieve a document 92 viewing 78 Document content direct accessing in C# 63 direct accessing in Java 62 Document doc 60, 77, 84, 173 document file type 78 Document object 4, 62, 77, 90, 169 DocumentSet document 173 Dojo library 33 Dojo toolkit 210 client 127, 142, 146, 188, 191, 193 domain 44 doMethod() 116 DynamicReferentialContainmentRelationship (DRCR) 96

Index

321

E
Eclipse 22 ECM Widget 38, 201202, 205 development setup 38 mashup page 202 system architecture 205 eForms API 8 EJB transport 14, 20, 47, 176, 293294, 296 requirement 20 element sequence number 62 EngineRuntimeException 49 Enterprise Information System (EIS) 168 Enterprise JavaBeans See EJB transport environment setup for .NET 29 setup, Java 13 error callback 128, 143, 189, 191, 194 error recovery 123 event handler 104 asynchronous 103 synchronous 103 event log 117, 129, 137, 162, 262 optimizing query 139 retrieving work items 138 work items 139 event logging categories 137 event subscription 103 Events package CE API 42 exception API 123 runtime 49 exception handling in C# 50 in Java 50 Process Engine (PE) 124 exception handling, Process Engine (PE) 123

FetchInstance() 71 fetchless instantiation 71 performance advantage 71 fetchMethod() 116 fetchProperties 58 filing a document in folder in C# 97 in folder in Java 96 filter.addIncludeProperty 6768, 84 Fleet Status Manager Web application architecture 272 deployment 296 fnlog4j.properties 311 folder 95 filing a document in C# 97 filing a document in Java 96 franchise location 109, 225, 257 business objects 261 maintenance employees 110 maintenance records 110 FREETEXT 76 full-text search 76

G
Get Next widget 208, 210 building 206 building solution with 219 defining 211 rendering 215 use case 207 getInstance 4649, 58, 70 getMethod() 116 graphical user interface (GUI) 202, 302 thick desktop Java application 299

H
HTTP GET 154 HTTP PUT 154

F
Factory.Domain.getInstance 4648, 6061, 84, 9697 fetching object in C# 59 in Java 59 fetching queue element 215 fetchInstance 58

I
IBM FileNet API 223, 305 log file 305 API library 80 Business Process Manager 7 Capture 2, 8 Content Manager 46

322

Developing Applications with IBM FileNet P8 APIs

content repository 8 Enterprise Manager 2, 168 Image Services 9 P8 deployment 1 Platform 2, 9, 168, 187 IBM FileNet Business Process Manager 111, 125 IBM FileNet Content Manager 67, 90, 108 IBM FileNet Enterprise Manager 280, 295, 310 building SQL 72 CE server logging 310 Query Builder 73 IBM FileNet P8 5, 7, 233, 238239 deployment 1 object Id 53 Platform 12, 5, 177 Platform documentation 72 product 3 resource 44 support site 308 IBM InfoSphere Enterprise Records 2, 4, 8 Java API 8 identity code 244245 ninth position 244 other types 245 vehicle 244 IdleActivity 53 Image Services (IS) 9 CFS-IS in-basket 187, 202 In-basket widget 204 IndependentlyPersistableObject 81 IndependentObject 81 inheritance custom inheritance, security 110 security 108 initial connection 44, 47, 121, 272, 275 integrated development environment (IDE) 299 INTERSECTION merged scope search 75 iScope Dojo class 213 isMethod() 116 isolated region 137 ITSO_Operations 264 ITSOChargesSchedule 245 ITSOCustomObject 249, 257 ITSOFranchiseCode 244, 257258 ITSORentalActivity 245 ITSORole object 110, 244, 258260, 262

same set 258 ITSOVehicle 244 ITSOVehicleActivity 248250, 252253 iWidget specification 202, 205, 211

J
J2EE application 4, 10, 168, 225, 268, 273 server 293, 296 server container 103 JAAS 4, 44, 46, 48, 172, 176, 182 JAAS stanza 46 Jace.jar 14 Java adapter 170, 172 Java API 34, 6, 40, 44, 48, 104, 112114, 124, 126, 132, 188189, 192 Java application setup for sample, Content Engine 22 stand-alone 15 Java archive (JAR) resource 182 Java Authentication and Authorization Service See JAAS Java business class 172 Java component 170 Java development environment setup 13 Process Engine 28 setup 27 Java gateway class 175 Java Messaging System (JMS) 168 Java virtual machine See JVM JavaScript Object Notation See JSON JOIN, merged scope search 75 JSON MIME type 112 object 127, 142 JVM 28, 121, 294, 298, 304, 306307, 313 JVM parameter required for thick client application 16

K
Kiosk application 62, 224, 231 user view 230

L
LDAP registry 123

Index

323

local cache property 58 log file API 305 size and rollover 307 thread identifier 308 log4j environment 306 log4j.jar 14 log4j.properties 306307, 313 login 176

M
maintenance employee 109110, 233, 253, 257 major version 90, 92 manipulation of compound document 97 mashup container 202 mashup page 202, 204, 206 widget 218 merged scope search 75 Meta package CE API 41 metadata 53 cache 106 discover at runtime 105 object 41, 283 minor version 90 multi-page file 61 multi-page scanned document 61 multi-valued property 54 property, setting 55 property, setting in C# 56 property, setting in Java 56 multi-valued object-valued property See MVOVP MVOVP 94, 243, 245246 My Work Page 202

deleting object 60 fetching, in C# 59 fetching, in Java 59 retrieving 42, 58 object store 4, 4344, 46, 72, 74, 104, 149, 151, 153, 168, 263, 272273, 275, 288289 search cross object stores 75 ObjectChangeEvent event (ONEVENT) 104 object-valued property (OVP) 54, 94, 109, 243244, 249 optimizing event log query 139 optimizing queue query 135 optimizing, queue 135 ORDER BY, merged scope search 75

P
p8_api_trace.log 307 paging support, result set 76 PE 13, 105, 111113, 126, 132, 135, 167168, 170, 187, 235, 270, 272, 305, 311 authentication 120 connection point 295 exception handling 123124 gather information for troubleshooting 314 Java API 7, 114, 121, 124 Java development project 28 Java development setup 27 maintenance queue 175 object 163 REST API 112, 199 REST Services API 7 server 121 session 119, 121, 125, 139 troubleshooting 313 Web service 7 Web Service Policy object 122 Web services 112 PE API 114115, 119, 121, 124, 269, 293, 297 application polling component queue 184 core classes 116 object 119 object method 115 other sets 112 usage tip 119 PE API classes functional relationships 118 PE error handling class 117 PE query, optimizing 130

N
Next widget 207, 210211

O
object creating object 52 creating object in C# 53 creating object in Java 53 deleting in C# 61 deleting in Java 60

324

Developing Applications with IBM FileNet P8 APIs

PE REST API 113, 140, 144, 150 PE REST API sample code development setup 33 scenario 188 PE REST sample application 33 PE REST service 207, 215, 312 PE retrieve work status classes 117 PE search work classes 117 PE session class 116 establish 119 object 181 PEComponentQueueHelper 183 pending action 52 performance boost, property filter 66 eliminating round trips 68 eliminating round trips in C# 69 eliminating round trips in Java 69 getInstance() versus fetchInstance() 71 limiting returned properties in C# 68 limiting returned properties in Java 67 optimizing PE queries 130 permission AccessRight.DELETE 8889 PEWS 32, 112, 119, 131, 158, 311 PEWS API 114 PEWS client 126, 132, 135 PEWS RequestSoapContext header creating 122 Post Once Exactly (POE) 127128 precompiled binary 279 process application 143 Process Configuration Console 137, 187 Process Designer 187 Process Engine See PE Process Engine Web Service See PEWS process history, retrieving 162 process status 162 Process Task Manager 170 programming COM Compatibility Layer 7 programming language 94 properties collection 54, 95, 104 property 53 annotations 94

association property 94 limiting returned properties for performance in C# 68 limiting returned properties for performance in Java 67 local cache 58 multi-valued 54 reflective property 94 single-valued 54 property filter 58, 6669, 71 boost performance 66 eliminate round trips for performance 68 eliminate round trips for performance in C# 69 eliminate round trips for performance in Java 69 limiting returned properties for performance in C# 68 limiting returned properties for performance in Java 67 Property package CE API 42 property value 42 retrieving 56 retrieving in C# 57 retrieving in Java 57 PropertyFilter 58

Q
query 135 building with SearchSQL 72 building with SearchSQL in C# 72 building with SearchSQL in Java 72 optimizing event log 139 optimizing, PE 130 queues 134 Query Builder Content Engine (CE) 73 IBM FileNet Enterprise Manager 73 Query package CE API 42 queue optimizing query 135 query 134 queue element 114, 195197, 207, 212 fetching 215 first page 197 queue name 114, 119, 135136, 141142, 190, 192, 194

Index

325

R
RCR 96 Redbooks Web site 318 Contact us xviii ReferentialContainmentRelationship (RCR) 96 reflective property 94 refresh batch 81 relationship containment relationship 95 repository IBM FileNet 8 Representational State Transfer (REST) 7, 112 request URI 127, 142, 146 reservation 91 reservation state 90 Reservation Web application 224, 264, 268 program flow 268 user view 226 ReservationType.EXCL USIVE 92 reserved state 91 resource navigation 199 REST API 7, 112, 114, 123, 188, 190, 193 resource 112 resource request 123 scenario 154, 188, 197 REST operation authorization for accessing 123 result set, paging support 76 retrieval name 62 retrieve items in batch 83 retrieving batch 81 retrieving content 6162 using Document object 62 retrieving object 42, 58 retrieving property value 56 in C# 57 in Java 57 retrieving work items 138 role 187 retrieving role description 189 retrieving role list 188 role-based access control (RBAC) 109 round trips eliminating for performance 68 eliminating for performance in C# 69 eliminating for performance in Java 69 runtime exception 49

S
sample application 96, 100, 104, 106, 122, 172, 223224, 226, 232233, 238239, 245, 250, 316 architecture 268 Billing Report application 224, 237239 Billing Report application, architecture 274 Billing Report application, user view 237 business use cases 224 data import files 240 deployment 276 distinct purposes 258 Fleet Status Manager Web application, architecture 272 Fleet Status Manager Web application, deployment 296 Java-based components 279 Kiosk application 62, 224, 231 Kiosk application, user view 230 metadata customization 279 PE REST sample application 33 Reservation Web application 224, 264 Reservation Web application, program flow 268 Reservation Web application, user view 226 security 249 security model 261 workflow 262 sample code 212, 215, 217, 224, 226 search 71 content search 76 cross object stores 75 example in C# 74 example in Java 73 INTERSECTION 75 JOIN 75 merged scope search 75 ORDER BY 75 result set, paging support 76 UNION 75 WHERE 75 work items from workbasket 194 search object 170 search operation 129 search scope 7475 SearchScope class 71 SearchSQL 69, 7174, 77, 107108, 173 security 90 custom inheritance 110 dynamic inheritance 108 sample applications 249

326

Developing Applications with IBM FileNet P8 APIs

security model, sample applications 261 Security package, CE API 41 security proxy property 110 security template 170 send mail 170 session 116 PE session, establish 119 session object 126, 132, 135, 141, 144, 150, 177, 181, 188, 190, 192, 195 setMethod() 116 SetProcessSecurityToken 45 SetThreadSecurityToken 45 setting property value multi-valued property 55 multi-valued property, in C# 56 multi-valued property, in Java 56 single-valued property 54 single-valued property, in C# 55 single-valued property, in Java 55 Single Page 202 single-valued property 54 property, setting 54 property, setting in C# 55 property, setting in Java 55 SOAP header request 119, 122 source object 103 SQL building using IBM FileNet Enterprise Manager 72 SQL statement 7173 Standard Widget Toolkit (SWT) 274, 299 Step Completion widget 204 step element 114, 140, 142, 148 parameter values updating 154 parameters 143 parameters retrieving 144, 146 retrieving 140141 step processor 113, 138, 140, 156 Step Processor Page 203 String password 179 subscription event 103 workflow 105 synchronous event handler 103 system queue 134

CEWS transport requirements 1819 required JVM parameters 16 thin client 15, 22 CEWS transport library requirements 22 CEWS transport requirement 21 EJB transport requirement 20 thread Identifier log file 308 trace log p8_api_trace 307 REST Service, Application Engine 312 transactional operation, batch 81 transport 2021, 46 CEWS requirements 1819 CEWS transport 21, 294, 298299, 310 EJB transport 14, 20, 45, 47, 176, 293294, 296 protocols 14 troubleshooting 312 CE server logging 310 connectivity 21 Content Engine (CE) 313314 Process Engine (PE) 313314

U
unfile object 169 Uniform Resource Identifier (URI) 15, 43, 119120, 188189, 191 UNION, merged scope search 75 update objects in batch 82 updating batch 81 user queue 134 user token 80 UserContext 44

V
vehicle barcode 232 identity code 244 See also VIN version 90 version series object 90 Viewer widget 202, 205 VIN 245246 VisualStudio.NET, configuring 30 VWAttachmentType 115 VWFetchType 115 VWQueue 115 VWQueue queue 135, 141, 174, 192, 195

T
thick client 15, 22

Index

327

VWQueueDefinition 115 VWQueueElement 115 VWRoster 115 VWRosterDefinition 115 VWRosterElement 115 VWSession object 117, 119120, 188189, 192 VWQueue object 192, 194 VWRole object 189 VWWorkObjectNumber 132, 135, 141

W
WcmApiConfig.properties 28, 120 Web application 7, 9, 209210, 218, 233, 292, 296 applicable JSF configuration 292, 296 Web service 3, 56, 4546, 48, 112, 119120, 168, 176 Content Engine (CE) 5 Process Engine (PE) 7 Web Service Definition Language See WSDL Web services interface (WSI) 5, 46 WebSphere Business Space 202 WHERE, merged scope search 75 Widget ECM 202 ECM Widget 201, 205 ECM Widgets, development setup 38 widget 202, 221, 274, 279, 299 Attachment widget 202, 205, 208 building custom widget 206 deploying 217 Get Next 215 Get Next widget 208, 210 Get Next widget defining 211 Get Next widget use case 207 Get Next widget, building solution 219 In-basket widget 204 mashup page 218 Next widget 207, 210211 Step Completion widget 204 Viewer widget 202, 205 wiring widget 202 Work Data widget 202, 205 WOB Number 274 Work Data widget 202, 205 displays customer-defined field 205 work item 170, 187, 204, 221, 235, 263 completing 156, 160

helper data 130 query first page 194 query from workbasket 194 retrieving 138 work object 113, 127, 132, 184 participant information 162 resource URL 127 VWProcess object 162, 165 work object number 142 work performer 140 work queue 134, 236, 266267 work search operation 129 workbasket 187, 191, 193, 200202 retrieving 192 workbasketName 191, 193, 197 workbasketURI 199 workflow 129, 139, 162, 262263 activity tracking 162 attachments 148 logs 139 participants 148 response list 204 roster 129 sample applications 262 workflow definition 105, 112113, 125, 140, 148, 168, 262, 265, 267, 288 data fields 168 specific step 140 workflow definition API 113 workflow event action 105 tracking 137 workflow history 162164 step history objects 162 workflow milestone 162 retrieving 165 workflow process 117, 125127 milestones definition 165 retrieving and launching 125 step history data 117 workflow subscription 105 WorkplaceRootUrl 79 WSDL 5, 29, 112, 121 Content Engine (CE) 5

328

Developing Applications with IBM FileNet P8 APIs

Developing Applications with IBM FileNet P8 APIs

(0.5 spine) 0.475<->0.875 250 <-> 459 pages

Back cover

Developing Applications with IBM FileNet P8 APIs

Content Engine basic and advanced APIs Process Engine basic and advanced APIs REST API, ECM Widgets, sample applications, and more

This IBM Redbooks publication can help you develop content and process management applications with IBM FileNet APIs. The IBM FileNet P8 suite of products contain a set of robust APIs that range from core platform APIs to supporting application APIs. This book focuses specifically on Content Engine and Process Engine APIs. Content Engine API topics that we discuss include creating, retrieving, updating, and deleting objects; querying and viewing documents; and batching and batch execution. We also explore more complex topics, including permissions and authorization, versioning, relationships, annotations, workflow subscriptions and event actions, metadata discovery, and dynamic security inheritance. Process Engine API topics that we discuss include launching a workflow, searching for and processing work items, and working with process status. The more complex topics we cover include, Component Integrator application space, role, workbasket, resource navigation in Process Engine REST API, ECM Widgets, and building a custom Get Next In-basket widget. To help you better understand programming with IBM FileNet APIs, we provide a sample application implemented for a fictional company. We include the data model, security model, workflows, and various applications developed for the sample. You can download them for your reference. This book is intended for IBM FileNet P8 application developers. We recommend using this book in conjunction with the online ECM help.

INTERNATIONAL TECHNICAL SUPPORT ORGANIZATION

BUILDING TECHNICAL INFORMATION BASED ON PRACTICAL EXPERIENCE IBM Redbooks are developed by the IBM International Technical Support Organization. Experts from IBM, Customers and Partners from around the world create timely technical information based on realistic scenarios. Specific recommendations are provided to help you implement IT solutions more effectively in your environment.

For more information: ibm.com/redbooks

SG24-7743-00 ISBN 0738433721