I have two RECFM=FBA,LRECL=164 files with identical layouts. The first 8 bytes after the CC is a label. The first record is a header, with CC 1. I want to sort and copy the second file, dropping every record that doesn't have a match on the first file. I think that for DFSORT something like this will work: JOINKEYS FILE=XREFS,FIELDS=(1,1,D,2,9,A) JOINKEYS FILE=LAYOUTS,FIELDS=(1,1,D,2,9,A) SUM FIELDS=NONE
-- Shmuel (Seymour J.) Metz http://mason.gmu.edu/~smetz3 עַם יִשְׂרָאֵל חַי נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר ________________________________________ From: IBM Mainframe Discussion List <[email protected]> on behalf of Alain Benvéniste <[email protected]> Sent: Saturday, January 31, 2026 4:50 AM To: [email protected] <[email protected]> Subject: Re: SyncSort join of two files External Message: Use Caution I don’t know what SyncSort is able to do. But if we consider that he has the same abilities as dfsort, I can help you. I would need a concret example of what you have in input and how you want to see the result. Alain Kyndryl Resiliency Services on Z Mainframe (Paris, France) [email protected] > Le 30 janv. 2026 à 18:51, Seymour J Metz <[email protected]> a écrit : > > I have two files extracted from ADATA. One has all of the symbolic > references, one has selected field definitions. I want to sort both files on > the field name and select only those field names that appear in the cross > reference file. I only want one copy of the field name definition even if > there are multiple XREF records for the field names. Can this be done in a > ingle SyncSort invocation? > > Thanks. > > -- > Shmuel (Seymour J.) Metz > http://mason.gmu.edu/~smetz3 > עַם יִשְׂרָאֵל חַי > נֵ֣צַח יִשְׂרָאֵ֔ל לֹ֥א יְשַׁקֵּ֖ר > > > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
