IBM Mainframe Forum

by **rmd3003** » Thu May 23, 2013 10:45 pm

Hello there. I was curious if it's possible to reformat multiple records with the same key into single one using ICETOOL?
Input file has multiple records with the same key (they can change). No more than 25 records per key.

Output file will have single record per key with occurences (25)

I also need to populate counter on record on output record in position 15

Input file:
-

Select all

---+----1----+----2----+----3----+----4----+----5
AAAA00001 R01X001
AAAA00001 R01X002
AAAA00222 R02X001
AAAA00222 R02X002
AAAA00222 R02X003
AAAA33333 R03X001
AAAA33333 R03X002
AAAA33333 R03X003
AAAA33333 R03X004
AAAA33333 R03X005

Output file will be like this:
Counter is highlighted in yellow.

Select all

---+----1----+----2----+----3----+----4----+----5----+--
AAAAA00001 002 R01X001 R01X002
AAAAA00222 003 R02X001 R02X002 R02X003
AAAAA33333 005 R03X001 R03X002 R03X003 R03X004 R03X004

If it's possible maybe somebody has similar code to show me.
Thank you very much in advance.....

BTW - SYNCSORT FOR Z/OS 1.4.0.1R

by **BillyBoyo** » Fri May 24, 2013 5:11 am

Well, there is a similar question today in the DFSORT forum on the Sister site, http://ibmmainframes.com You'll need to use SPLICE with several IFTHENs (25 at least to arrange the data, perhaps a few more to do the count).

by **Thampy** » Sat May 25, 2013 4:27 pm

Please find below the Syncsort JCL for the single record per key with occurences (5). You can modify the jcl to add another 20 IFTHEN's to suit your requirement
The OUTFIL statement has to be modfied to the record length you are looking for

//STEP010 EXEC PGM=SORT
//SYSOUT DD SYSOUT=*
//SORTIN DD *
AAAA00001 R01X001
AAAA00001 R01X002
AAAA00222 R02X001
AAAA00222 R02X002
AAAA00222 R02X003
AAAA33333 R03X001
AAAA33333 R03X002
AAAA33333 R03X003
AAAA33333 R03X004
AAAA33333 R03X005
//SORTOUT DD SYSOUT=*
//SYSIN DD *
INREC IFTHEN=(WHEN=INIT,OVERLAY=(15:C'001',28:8Z,36:8Z,
44:8Z,52:8Z,
300:SEQNUM,3,ZD,RESTART=(1,9))),
IFTHEN=(WHEN=(300,3,ZD,EQ,2),OVERLAY=(28:20,8,20:8Z)),
IFTHEN=(WHEN=(300,3,ZD,EQ,3),OVERLAY=(36:20,8,20:8Z)),
IFTHEN=(WHEN=(300,3,ZD,EQ,4),OVERLAY=(44:20,8,20:8Z)),
IFTHEN=(WHEN=(300,3,ZD,EQ,5),OVERLAY=(52:20,8,20:8Z))
SORT FIELDS=(1,9,CH,A)
SUM FIELDS=(15,3,ZD,20,8,BI,28,8,BI,36,8,BI,44,8,BI,
52,8,BI)
OUTREC FINDREP=(IN=X'00',OUT=X'40')
OUTFIL BUILD=(1,80)

Sample Output
AAAA00001 002 R01X001 R01X002
AAAA00222 003 R02X001 R02X002 R02X003
AAAA33333 005 R03X001 R03X002 R03X003 R03X004 R03X005

by **NicC** » Sun May 26, 2013 8:02 pm

As IFTHEN is a sort control keyword modifying the JCL is impossible as it does not have IFTHEN. Presumably you meant modify the sort control cards.

by **rmd3003** » Tue May 28, 2013 8:07 pm

Thanks everybody.

by **Thampy** » Wed May 29, 2013 12:45 am

I meant modify the SORT control cards to add 20 IFTHEN's. Sorry for the confusion caused.

IBM Mainframe Forum

Reformat multiple records into single with occurences????

Reformat multiple records into single with occurences????

Re: Reformat multiple records into single with occurences???

Re: Reformat multiple records into single with occurences???

Re: Reformat multiple records into single with occurences???

Re: Reformat multiple records into single with occurences???

Re: Reformat multiple records into single with occurences???