Some of the posts in this thread seem to view DCB attributes as a stick-on label that can be peeled off and replaced as needed. While hardware knows nothing of LRECL--hence the Logical nature--block size is very much a physical attribute of a file on CKD DASD. You may get away with output to a file with overriding attributes, but attempting to read data back via QSAM will fail for FB data if physical BLKSIZE is not an integral multiple of LRECL for any block being read.
All in all, I can't imagine what 'convenience' would be achieved by creating write-only PDS(E) members. . . JO.Skip Robinson Southern California Edison Company Electric Dragon Team Paddler SHARE MVS Program Co-Manager 626-302-7535 Office 323-715-0595 Mobile [email protected] From: Paul Gilmartin <[email protected]> To: [email protected], Date: 04/22/2013 07:59 AM Subject: Re: Files in PDS's with different LRECL's Sent by: IBM Mainframe Discussion List <[email protected]> On 2013-04-20, at 03:02, R.S. wrote: > > BTDT. > FOr PDSE you'll get an error, for PDS your file may be (will be) corrupted. > I tried this with PDSE. For RECFM=VB, I can increase the LRECL with overriding JCL with no error reported and no apparent data corruption. Attempting to decrease the LRECL results in IEC141I 013-EC. For RECFM=FB, Attempting to increase the LRECL results in IEC141I 013-EC. I like the PDSE implementation: it maintains compatibility with PDS in the safe cases, and reports an error (as PDS fails to) in the cases in which your file may be (will be) corrupted. Other cases and PDS are left as an exercise for the student. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
