IBM BPM, V8.0.1, All platforms > Programming IBM BPM > Developing client applications for BPEL processes and tasks > Queries on BPEL process and task data > Query tables in Business Process Choreographer

Query table development

Supplemental and composite query tables in Business Process Choreographer are developed during application development using the Query Table Builder. Predefined query tables cannot be developed or deployed. They are available when Business Process Choreographer is installed and provide a simple view on the artifacts in the Business Process Choreographer database schema.

The Query Table Builder is available as an Eclipse plug-in and can be downloaded on the SupportPacs site. Look for PA71 WebSphere Process Server - Query Table Builder. To access the link, see the related references section of this topic.

Query tables impact on the way applications are developed and deployed. The following steps describe the roles involved when you design and develop a Business Process Choreographer application that uses query tables.

Query table development steps
Step Who Description
1. Analysis Business analyst, client developer Analyze which query tables are needed in the client application. Questions to be answered are:

  • How many task or process lists are provided to the user? Are there task or process lists that can share the same query table?
  • What kind of authorization is used? Instance-based authorization, role-based authorization, or none?

  • Are there other query tables already defined in the system that can be reused?
  • Must the query tables provide the content in multiple languages?

    If so, the selection criteria on attached query tables.should be LOCALE=$LOCALE.

2. Query table development Client developer, business analyst Develop the query tables used in the client application. Try to specify the definition of the query tables such that the best performance is achieved with query table queries.
3. Query table deployment Administrator Query tables must be deployed to the runtime before they can be used. This step is done using the manageQueryTable.py wsadmin command.
4. Query table queries Client developer To run queries against query tables is the last step of query table development. The client developer must know the name of the query table and its attributes.

The following is sample code, which uses the query table API to query a query table. Examples 1 and 2 are provided to query the predefined query table TASK for simplicity reasons. Examples 3 and 4 query a composite query table, which is assumed to be deployed on the system. In application development, you should use composite query tables rather than directly querying the predefined query tables.


Example 1

   // get the naming context and lookup the Business    // Flow Manager Enterprise JavaBeans home; note that the Business Flow 
   // Manager Enterprise JavaBeans home should be cached for performance 
   // reasons; also, it is assumed that there's an Enterprise JavaBeans 
   // reference to the local business flow manager Enterprise JavaBeans
   Context ctx = new InitialContext();
   LocalBusinessFlowManagerHome home = 
   	(LocalBusinessFlowManagerHome) 
   	ctx.lookup("java:comp/env/ejb/BFM");
   
   // if the human task manager Enterprise JavaBeans is used, do:
   // LocalHumanTaskManagerHome home =
   //   (LocalHumanTaskManagerHome) ctx.lookup("java:comp/env/ejb/HTM");
   // assuming that a EJB reference to the human task manager EJB    // has been defined
   
   // create the business flow manager client-side stub
   LocalBusinessFlowManager bfm = home.create();
   // if the human task manager EJB is used, do:
   // LocalHumanTaskManager htm = home.create();
   // note that the human task manager Enterprise JavaBeans provides the   // same methods as the business flow manager Enterprise JavaBeans 
   // *************************************************
   // ******************* example 1 *******************
   // *************************************************
   
   // run a query against the TASK predefined query 
   // table; this relates to a simple My ToDo's task list
   EntityResultSet ers = null;
   ers = bfm.queryEntities("TASK", null, null, null);
   
   // print the result to STDOUT
   EntityInfo entityInfo = ers.getEntityInfo();
   List attList = entityInfo.getAttributeInfo();
   int attSize = attList.size();

   Iterator iter = ers.getEntities().iterator();
   while (iter.hasNext()) {
   	System.out.print("Entity: ");
   	Entity entity = (Entity) iter.next();
   	for (int i = attSize - 1; i >= 0; i--) {
   		AttributeInfo ai = (AttributeInfo) attList.get(i);
   		System.out.print(
      	entity.getAttributeValue(ai.getName()));
   	}    	System.out.println();
   }


Example 2

   // *************************************************
   // ******************* example 2 *******************
   // *************************************************
   
   // same example as example 1, but using the row-based
   // query approach    RowResultSet rrs = null;
   rrs = bfm.queryRows("TASK", null, null, null);
   
   attList = rrs.getAttributeInfo();
   attSize = attList.size();

   // print the result to STDOUT
   while (rrs.next()) {
   	System.out.print("Row: ");
   	for (int i = attSize - 1; i >= 0; i--) {
   		AttributeInfo ai = (AttributeInfo) attList.get(i);
   		System.out.print(
      	rrs.getAttributeValue(ai.getName()));
   	}    	System.out.println();
   }


Example 3

   // *************************************************
   // ******************* example 3 *******************
   // *************************************************
   
   // run a query against a composite query table
   // that has been deployed on the system before;
   // the name is assumed to be COMPANY.TASK_LIST
   ers = bfm.queryEntities(
   		"COMPANY.TASK_LIST", null, null, null);
^
   // 	print the result to STDOUT ...


Example 4

   // *************************************************
   // ******************* example 4 *******************
   // *************************************************
   
   // query against the same query table as in example 3,    // but with customized options
   FilterOptions fo = new FilterOptions();

   // return only objects which are in state ready
   fo.setQueryCondition("STATE=STATE_READY");
   
   // sort by the id of the object    fo.setSortAttributes("ID");
   
   // limit the number of entities to 50
   fo.setThreshold(50);
   
   // only get a sub-set of the defined attributes
   // on the query table
   fo.setSelectedAttributes("ID, STATE, DESCRIPTION");
   
   AuthorizationOptions ao = new AuthorizationOptions();
   
   // do not return objects that everybody is allowed
   // to see   ao.setEverybodyUsed(Boolean.FALSE);
   
   ers = bfm.queryEntities(
   		"COMPANY.TASK_LIST", fo, ao, null);

   // 	print the result to STDOUT ...

Query tables in Business Process Choreographer


Related concepts:
Query table queries
Filters and selection criteria of query tables
Predefined query tables
Supplemental query tables
Composite query tables


Related tasks:
Administer query tables
Deploying composite and supplemental query tables


Related information:

WebSphere Business Process Management SupportPacs