MVSFORUMS.com

[tiggy]

So here's my dilemma. I have a file which is too large to simply edit. I have a single record in that file in which I need to change one value. I have the file layout, and criteria on which to find the file (three fields). Based on a match of these three fields, I need to change the value in a fourth field from x to y.
I was wondering if there is any way to do this in a single sort, and if so, how? This is something that I need to have done today to several files and it would be much easier to use a sort than to write a COBOL program to do this. It would also be a useful tool for the future for my colleagues & I.

Thanks,
_________________
Robert Tighe
Legacy Systems Analyst
Keane Canada, Inc.
Robert_F_Tighe@keane.com

[Frank Yaeger]

You can do this quite easily with the new IFTHEN and OVERLAY parameters of DFSORT's INREC, OUTREC or OUTFIL statement available with z/OS DFSORT V1R5 PTF UQ95214 or DFSORT R14 PTF UQ95213 (Dec, 2004). For complete information on all of the new DFSORT and ICETOOL functions available with the Dec, 2004 PTF, see:

www.ibm.com/servers/storage/support/software/sort/mvs/pdug/

You didn't give any details, so I'll set up an example to give you the idea.

If you want to find a record that has 'XX' in positions 5-6 and 'Y' in position 10 and 'ZAB' in positions 15-17, and change positions 20-22 in that record to 'YES', you can use the following DFSORT job:

[tiggy]

This is awesome, that's exactly what I'm looking for. Thanks so much. I'll be sure to put more detail next time.

Thanks again.
_________________
Robert Tighe
Legacy Systems Analyst
Keane Canada, Inc.
Robert_F_Tighe@keane.com

[tiggy]

I think I spoke too soon. I tried what you gave me:

[kolusu]

Tiggy,

I guess you don't have the latest ptf installed in your shop. here is an alternate way of doing it.

I assuming your input file is 80 bytes in length and is of FB recfm.

Using OUTREC fileds we put the first 17 bytes and 8 bytes from 26 byte and 5 bytes from 56th byte at the end of every record.

Now using a change command on OUTFIL we look it as a single string and change the position at the 34 and copy rest of the bytes upto 80 bytes.

[Frank Yaeger]

Tiggy,

The error messages indicate you don't have DFSORT's Dec, 2004 PTF installed. Ask your System Programmer to install it (it's free).

Note that the method using CHANGE that Kolusu shows works when the criteria involves matching a set of character strings, but becomes problematical when the criteria is more complex such as when it involves numeric comparisons (ZD field gt +250), or when OR must be used ((A OR B OR C) AND D), etc). IFTHEN can handle all of these cases!
_________________
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

[tiggy]

Great, thanks for all your time and help, the two of you. This has been a great help and very educational. I really appreciate it.

Thanks,
_________________
Robert Tighe
Legacy Systems Analyst
Keane Canada, Inc.
Robert_F_Tighe@keane.com

[rishikesh_u]

Kolusu,

Tried the sort option as given by you and it's working fine. But i am not very clear how this is acheived.

What is the need of "Using OUTREC fileds we put the first 17 bytes and 8 bytes from 26 byte and 5 bytes from 56th byte at the end of every record."

and can you also explain how the below works and how did you arrive at
C'5TDZA22C05S2575262005020432097'.

OUTFIL OUTREC=(1,33,
34:81,30,CHANGE=(3,C'5TDZA22C05S2575262005020432097',
C'999'),NOMATCH=(34,3),
37,44)

your reply is highly appreciable..

Thanks,

[kolusu]

Rishikesh_u,

The orginal poster wanted to validate 3 fileds at different locations and change another field. Using Outrec fields I put the 3 fields together at the end of every record, so that he can validate the 3 fields as one single field.