how to get A pds member name using dfsort

sureshmurali · Posted: Sun Jan 22, 2012 12:08 am

i have a pds of 8 members and each of them is a cobol program. i want to list out paragraph names of each of them to a single ps with only the details of member name and paragraph names in it. Can anybody please tell me how to get the member name of a pds using dfsort? I am new to SORT also.

vasanthz · Posted: Sun Jan 22, 2012 1:36 am

I think REXX would be more suited & straight forward to this task than DFSORT.
But then its only my personal POV.

sureshmurali · Posted: Sun Jan 22, 2012 9:09 am

Thanks for the reply. But could you please let me know if it is possible to get the member name of a pds using SORT ?

Nic Clouston · Posted: Sun Jan 22, 2012 10:39 am

Have you looked for examples in the DFSORT documentation or in the forums? I would also look at the JCL requirements for DFSORT and sort control cards to see if anything is mentioned about PDS members. The indexes of the various manuals will narrow down your search.

Bill Woodger · Posted: Mon Jan 23, 2012 6:35 am

gcicchet · Posted: Mon Jan 23, 2012 1:38 pm

Hi,

DFSORT alone cannot give you what you want.

DFSORT can operate on a PDS(member), but not on all the members of a PDS at once.

Gerry

Skolusu · Posted: Mon Jan 23, 2012 10:23 pm

sureshmurali,

1.Flatten the PDS into a sequential file using IEBPTPCH which will put a header with the member name followed by the contents of the member.
2. Read that sequential file into DFSORT and perform the task you want.

The above 2 tasks are pretty simple , but the biggest problem for you is to identify the paragraph names.

Narayanan Annamalai · New User Joined: 20 Jan 2012 Posts: 6 Location: India

As per the coding standards, I believe the paragraph names (User defined) and few pre-defined parameter should begin in Area A (col 8-11).
Given that, I think we should be able to filter out the paragraph names using SORT Control cards.

Bill Woodger · Posted: Wed Jan 25, 2012 3:32 pm

Narayanan Annamalai · New User Joined: 20 Jan 2012 Posts: 6 Location: India

Ok. I thought a suggestion could trigger out ideas to accomplish this requirement.

I have extracted the paragraph names using REXX with the same logic and have succeeded.

May be, as you mentioned, the constraint is with COBOL Standards followed which should be optimum.

Bill Woodger · Posted: Wed Jan 25, 2012 4:14 pm

Your suggestion to use rexx instead of Sort is a good one. The point being there are better ways to accomplish the task than to use Sort, for this one.

sureshmurali · Posted: Wed Jan 25, 2012 8:14 pm

Hi everyone

I am really sorry and i should have told this before.
I already did the same task using REXX before. But interested in knowing how it could be done thru SORT.

Thanks

Bill Woodger · Posted: Wed Jan 25, 2012 8:45 pm

Yes, would have saved some...

Kolusu has shown you how to do the member followed by contents (not in Sort).

It is possible to do in Sort. If you are that interested, have a crack yourself and show us when/if you get stuck. If it is a raging success, post the output here please, for others who might through some quirk of fate have to do this in Sort.