Grouping records

This section describes how to group records.

Usage

The record key for reduced data contains a number of OMEGAMON for Db2 PE identifiers. The identifiers contained in the key vary according to the report set.

When records are grouped, the set name of the records relating to grouped items is substituted for the original value of the specified identifier in the record key. When the substitution has been made, the records are not available for processing by using the original key values.

The following examples illustrate how sets are processed.

Example 1:

The following records are processed:

Table 1. Record set processing
Location	Connection ID	Correlation Name	Plan Name	Primary Authorization ID
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_2	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_3	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_4	ACCOUNTS

The following OMEGAMON for Db2 PE command stream example groups three of the four plans:

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
 
EXEC

This establishes a set named PLANGRP that you can use with INCLUDE or EXCLUDE on REDUCE and REPORT.

The stage of processing when the key value substitution takes place depends on where you include the set:

You can use sets on GLOBAL(INCLUDE) to include records for grouped items. No substitution takes place during preprocessing, but the input records for the grouped items are included as if you entered each item.

Note: When using sets on GLOBAL, remember that the GLOBAL INCLUDE or EXCLUDE specification becomes the default for all other commands in the job step. GLOBAL INCLUDE is illustrated in the following example:

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
 
EXEC

The following records are available for further processing:

Table 2. Records grouped using GLOBAL(INCLUDE)
Location	Connection ID	Correlation Name	Plan Name	Primary Authorization ID
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_2	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_3	ACCOUNTS

When you use the set name on REDUCE(INCLUDE), the set name is substituted for the original value of the identifier in the key during REDUCE processing. The reduced data contains only the grouped records. You cannot process data by using the original key values in REPORT, and only the records for the set are stored in the save data set.
```
 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
ACCOUNTING
   REDUCE
      INCLUDE (PLANNAME (G(PLANGRP)))
   SAVE
 
EXEC
```

When you use the set name on GLOBAL(INCLUDE), it acts as the default for subcommands without an INCLUDE specification. The result for the following example is the same as the result for the preceding example. The records are grouped during REDUCE processing by using the GLOBAL(INCLUDE) default.

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
ACCOUNTING
   REDUCE
   SAVE
 
EXEC

The reduced data and the save data set contain the following record:

Table 3. Records grouped during REDUCE processing using the GLOBAL(INCLUDE)
Location	Connection ID	Correlation Name	Correlation Number	Plan Name	Primary Authorization ID
LOCATION_1	TSO	USER_1	0	PLANGRP	ACCOUNTS

When you use the set name on REPORT(INCLUDE), the records for grouped items are consolidated during report processing. In the following examples, the REDUCE subcommand specification overrides the GLOBAL default. All records that pass data filtering are included in the save data set because they are not grouped during REDUCE processing.

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		    )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
ACCOUNTING
   REDUCE
      INCLUDE (PLANNAME(*))
   REPORT
      INCLUDE (PLANNAME(G(PLANGRP)))
   SAVE
 
EXEC

In the following example, the set name on GLOBAL(INCLUDE) acts as the default for the REPORT subcommand. The result is the same as for the previous example.

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
ACCOUNTING
   REDUCE
      INCLUDE (PLANNAME(*))
   REPORT
   SAVE
 
EXEC

The Accounting report contains the following entry:

Table 4. Accounting report set
Location	Connection ID	Correlation Name	Correlation Number	Plan Name	Primary Authorization ID
LOCATION_1	TSO	USER_1	0	PLANGRP	ACCOUNTS

The save data set contains the following records:

Table 5. Save data for the accounting report set
Location	Connection ID	Correlation Name	Plan Name	Primary Authorization ID
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_1	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_2	ACCOUNTS
LOCATION_1	TSO	USER_1	PLAN_3	ACCOUNTS

Example 2

In this example, two reports are generated. The first presents set items individually. In the second report, records are grouped. The REDUCE subcommand specification again overrides the GLOBAL default, so all records that pass preprocessing are included in the save data set.

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME (G(PLANGRP))
		)
ACCOUNTING
   REDUCE
      INCLUDE (PLANNAME(*))
   REPORT
      INCLUDE (PLANNAME(*))
   REPORT
      INCLUDE (PLANNAME(G(PLANGRP)))
   SAVE
 
EXEC

In the following example, the set name on GLOBAL(INCLUDE) acts as the default for the second REPORT subcommand. The result is the same as for the previous example.

 
//SYSIN DD *
GROUP (
	PLANNAME (
	   PLANGRP (
		    PLAN_1
		    PLAN_2
		    PLAN_3
		   )
		 )
	 )
GLOBAL
	INCLUDE (
	   PLANNAME(G(PLANGRP))
		)
ACCOUNTING
   REDUCE
      INCLUDE (PLANNAME(*))
   REPORT
      INCLUDE (PLANNAME(*))
   REPORT
   SAVE
 
EXEC