IBM Mainframe Forum

by **Robert Alford** » Fri Jan 27, 2012 11:09 am

Hi,

I've read the manuals about GROUPing with BEGIN and KEYBEGIN and how they can be used together. What I am struggling with is how I can achieve my expected results below.

I know what BEGINs my group, but I can't determine the END of the group, other than change of key.

Input Data:

Select all

111 DATA ...... 1
111 HEADING AAA 2
111 DATA ...... 3
111 HEADING BBB 4
111 DATA ...... 5
111 DATA ...... 6
222 DATA ...... 1
222 DATA ...... 2
222 HEADING AAA 3
222 DATA ...... 4
222 DATA ...... 5
222 HEADING BBB 6
222 DATA ...... 7

Expected Results I want to acheive:

Select all

111 DATA ...... 1
111 HEADING AAA 2 111AAA
111 DATA ...... 3 111AAA
111 HEADING BBB 4 111BBB
111 DATA ...... 5 111BBB
111 DATA ...... 6 111BBB
222 DATA ...... 1
222 DATA ...... 2
222 HEADING AAA 3 222AAA
222 DATA ...... 4 222AAA
222 DATA ...... 5 222AAA
222 HEADING BBB 6 222BBB
222 DATA ...... 7 222BBB

Code I've attempted, but is not quite there for what I wanted:

Select all

//TESTSORT EXEC ICETOOL
//SYSIN DD *
COPY FROM(SORTIN) TO(SORTOUT) USING(IND1)
/*
//SORTIN DD *
111 DATA ...... 1
111 HEADING AAA 2
111 DATA ...... 3
111 HEADING BBB 4
111 DATA ...... 5
111 DATA ...... 6
222 DATA ...... 1
222 DATA ...... 2
222 HEADING AAA 3
222 DATA ...... 4
222 DATA ...... 5
222 HEADING BBB 6
222 DATA ...... 7
/*
//SORTOUT DD SYSOUT=*
//IND1CNTL DD *
INREC IFTHEN=(WHEN=GROUP,
KEYBEGIN=(001,003),
BEGIN=(005,007,CH,EQ,C'HEADING'),
PUSH=(020:001,003,
013,003))
/*
//*

with undesired result:

Select all

111 DATA ...... 1 111...
111 HEADING AAA 2 111AAA
111 DATA ...... 3 111AAA
111 HEADING BBB 4 111BBB
111 DATA ...... 5 111BBB
111 DATA ...... 6 111BBB
222 DATA ...... 1 222...
222 DATA ...... 2 222...
222 HEADING AAA 3 222AAA
222 DATA ...... 4 222AAA
222 DATA ...... 5 222AAA
222 HEADING BBB 6 222BBB
222 DATA ...... 7 222BBB

Any assistance would be great. Thanks.

by **BillyBoyo** » Fri Jan 27, 2012 3:48 pm

Are you saying if there is no heading you want blanks in the additional fields?

by **skolusu** » Fri Jan 27, 2012 9:51 pm

Robert Alford,

If your understand your requirement correctly , you can just do it with Begin itself . Use the following DFSORT JCL which will give you the desired results.

Select all

//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
----+----1----+----2----+----3----+----4----+----5----+--
111 DATA ...... 1
111 HEADING AAA 2
111 DATA ...... 3
111 HEADING BBB 4
111 DATA ...... 5
111 DATA ...... 6
222 DATA ...... 1
222 DATA ...... 2
222 HEADING AAA 3
222 DATA ...... 4
222 DATA ...... 5
222 HEADING BBB 6
222 DATA ...... 7
//SORTOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=COPY
INREC IFTHEN=(WHEN=GROUP,BEGIN=(5,7,CH,EQ,C'HEADING'),
PUSH=(20:1,3,13,3)),
IFTHEN=(WHEN=(1,3,CH,NE,20,3,CH),OVERLAY=(20:6X))
//*

by **Robert Alford** » Mon Jan 30, 2012 3:56 am

Hi,

Skolusu, simple and elagent solution I admit I overlooked. Was trying to be unnecessarily fancy.

Thank you for your wisdom. It will do the job nicely.

Cheers,
Rob.

IBM Mainframe Forum

Using GROUP with BEGIN 'AND' KEYBEGIN

Using GROUP with BEGIN 'AND' KEYBEGIN

Re: Using GROUP with BEGIN 'AND' KEYBEGIN

Re: Using GROUP with BEGIN 'AND' KEYBEGIN

Re: Using GROUP with BEGIN 'AND' KEYBEGIN