IBM Mainframe Forum

by **mfrookie** » Tue Apr 26, 2011 10:21 am

I am using LISTDS to list all the members of some of the PDSs. My input looks like as shown and I want it to be formatted as shown in output below. I tried a lot but couldn't it as expected. I can do it in multiple passes but was wondering if there a better way to do it in minimum passes.

Input -

Select all

READY
LISTDS 'USERID.COBSRC' MEMBERS
USERID.COBSRC
--RECFM-LRECL-BLKSIZE-DSORG
FB 80 32720 PO

--VOLUMES--
LILTAN
--MEMBERS--
ABI00001
ABI00002
ABI00003
ABI00004
READY
LISTDS 'USERID.CPYLIB' MEMBERS
USERID.CPYLIB
--RECFM-LRECL-BLKSIZE-DSORG
FB 80 27920 PO

--VOLUMES--
LILTAL
--MEMBERS--
ABC00001
DGC00001
POC00001
READY
LISTDS 'USERID.JCLLIB' MEMBERS
USERID.JCLLIB
--RECFM-LRECL-BLKSIZE-DSORG
FB 80 27920 PO

--VOLUMES--
LIXLAD
--MEMBERS--
SDJCL001
SDJCL002
READY
LISTDS 'USERID.PROCLIB' MEMBERS
USERID.PROCLIB
--RECFM-LRECL-BLKSIZE-DSORG
FB 80 27920 PO

--VOLUMES--
LILTAL
--MEMBERS--
SDPRC001
SDPRC002
READY
END

Expected Output -

Select all

ABI00001 USERID.COBSRC
ABI00002 USERID.COBSRC
ABI00003 USERID.COBSRC
ABI00004 USERID.COBSRC
ABC00001 USERID.CPYLIB
DGC00001 USERID.CPYLIB
POC00001 USERID.CPYLIB
SDJCL001 USERID.JCLLIB
SDJCL002 USERID.JCLLIB
SDPRC001 USERID.PROCLIB
SDPRC002 USERID.PROCLIB

In short, besides the member name, I want the PDS name to be appear to which the member belongs. this output will further be used in building multiple FTP scripts.

Thanks

by **skolusu** » Tue Apr 26, 2011 9:42 pm

mfrookie,

Use the following DFSORT JCL which will give you the results in one pass. I am also showing you the LISTDS step. This will create the output of 80 bytes with member name of 8 bytes and PDS name of 36 bytes and the rest padded with spaces.

Select all

//STEP0050 EXEC PGM=IKJEFT01
//SYSTSPRT DD DSN=&&L,
// DISP=(NEW,CATLG,DELETE),
// SPACE=(CYL,(15,5),RLSE),
// DCB=(LRECL=80,RECFM=FB,BLKSIZE=27920)
//SYSTSIN DD *
LISTDS 'USERID.COBSRC' MEMBERS
LISTDS 'USERID.CPYLIB' MEMBERS
LISTDS 'USERID.JCLLIB' MEMBERS
LISTDS 'USERID.PROCLIB' MEMBERS
//*
//STEP0100 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD DSN=&&L,DISP=SHR
//SORTOUT DD SYSOUT=*
//SYSIN DD *
SORT FIELDS=COPY
OMIT COND=(1,3,CH,EQ,C'END')
INREC IFTHEN=(WHEN=GROUP,BEGIN=(3,6,CH,EQ,C'LISTDS'),
PUSH=(81:SEQ=1),RECORDS=2),
IFTHEN=(WHEN=GROUP,BEGIN=(81,1,ZD,EQ,2),PUSH=(82:1,36)),
IFTHEN=(WHEN=GROUP,BEGIN=(01,11,CH,EQ,C'--MEMBERS--'),
END=(1,5,CH,EQ,C'READY'),PUSH=(118:1,1))

OUTFIL BUILD=(3,8,X,82,36,80:X),
OMIT=(118,01,CH,EQ,C' ',OR,
001,05,CH,EQ,C'READY',OR,
001,11,CH,EQ,C'--MEMBERS--')
//*

by **mfrookie** » Wed Apr 27, 2011 11:16 am

Hi Kolusu,

Thanks a lot for the solution. Its working perfectly.

It seems my understanding of WHEN=GROUP clause wasn't correct. I was under the impression that at a time only one WHEN=GROUP (the first one that satisfies the condition, similar to that of EVALUATE clause in COBOL) is applied and then all following GROUP clauses are ignored. And I always thought why don't we have HIT=NEXT option available on WHEN=GROUP clause whereas its already available on WHEN=(condition) clause.

But now I am clear on how it works, should have carefully read the fine manual.

Thanks lot once again.

IBM Mainframe Forum

Format LISTDS output using DFSORT

Format LISTDS output using DFSORT

Re: Format LISTDS output using DFSORT

Re: Format LISTDS output using DFSORT