MVSFORUMS.com

jacky_ying · Beginner Joined: 25 Jul 2004 Posts: 46 Topics: 20

The requirement is to get dataset name corresponding to each record in concatenated DD statements. Can this be
done with lastest DFSORT or is it possible to add this function in DFSORT.

I met this situation:say you got lots of datasets to process from different departments or branchs,the format of datasets is all some.No field in the dataset record distinguish "who is who",the only way to identify "who is who" is by dataset name.These datasets will be concatenate in one DD statements for process,so I had to write a sub program to scan the dataset and get the dataset name corresponding to each record then append the dataset name to the tail of each record.Y'know I can get TIOENTRY according together TCBTIO with DCBTIOT,and then call SWAREQ with TIOEJFCB to get JFCBDSNM.

I'm just wondering any of you guys met the same situation. It just happened to me or it occurs frequently?If latter,maybe we could suggest Frank Yaeger to add these function to the next version of DFSORT.Maybe some new reserve words DATATSETNAME MEMBERNAME like SEQNUM.haha. Of course,if you guys think it is useless or seldom needed,ignore this post plz.

If DFSORT can handle this.I'll do less coding.I am a naturally LAZY guy =^_^=

Frank Yaeger · Posted: Sat Mar 17, 2007 10:07 am Post subject:

DFSORT does not have a built-in function to do that. Nobody else has asked for it previously.
_________________
Frank Yaeger - DFSORT Development Team (IBM)
Specialties: JOINKEYS, FINDREP, WHEN=GROUP, ICETOOL, Symbols, Migration
DFSORT is on the Web at:
www.ibm.com/storage/dfsort

jacky_ying · Beginner Joined: 25 Jul 2004 Posts: 46 Topics: 20

Frank:

Thanks for rapid reply,It's OK,I just finished my program and it works.
Since nobody else asked for it previously,maybe this situation seldom happens.

Nic Clouston · Posted: Mon Mar 19, 2007 2:36 am Post subject:

I must say, I was looking at the sort (SYSOUT) messages the other day to see if the file name was in there. Although I do not need it now maybe some time in the future. What do people think? I vote FOR it - but Frank has to DO it and it is only a 'nice to have'

kolusu · Posted: Mon Mar 19, 2007 5:17 am Post subject:

IMHO it is one of the MUST have feature. However it gets complicated with the concatenated DD Lists.

Kolusu
_________________
Kolusu
www.linkedin.com/in/kolusu

Mervyn · Posted: Mon Mar 19, 2007 7:16 am Post subject:

I'd say it's certainly a "nice to have".
_________________
The day you stop learning the dinosaur becomes extinct

Alain Benveniste · Posted: Mon Mar 19, 2007 7:23 am Post subject:

Jacky,

Why not using the group key technique as an alternative :
- Each group value points to the corresponding dsname or,
- A HDR can be added manually that can be propagated by a SPLICE

Alain

Cogito-Ergo-Sum · Posted: Fri Mar 23, 2007 8:46 am Post subject: