On Wed, 13 Apr 2016 23:54:58 +0000, Savor, Thomas (Alpharetta) wrote: >I've run into another PDSe problem. >When I was reading through a PDS file of Source Code, I would define DCB as >RECFM=U, >After the OPEN, DCBLRECL = Record Length. DCBBLKSI = File Block Size >After issuing a READ, DCBLRECL would be the Actual Block Size. >My PDS : LRECL=80, BLKSIZE=27920 Number of records per block = 349 >So if a program has 500 lines of code: >First read DCBLRECL would = 27920, Number of lines would = 349 >Next read DCBLRECL would = 12080, Number of lines would = 151 >Maybe it was an accident, but this has worked for me forever. > >Now same program using same DCB definition, determines than it's a PDSe before >the OPEN. >If it's a PDSe, program changes RECFM to be FB....which returned a full block >of records. >How do I tell when the difference between a full block and a short block on a >PDSe ?? >Hope I explained this right. > I don't think block boundaries mean much to a PDSE which keeps everything in 4KiB pages. You can override with any DCBBLKSI from 80 to 27920 and a READ will fetch enough records to fill that block.
Customarily, RECFM=U indicates a Program Object. Perhaps PDSE enforces this. What are you trying to do? -- gil > >Thanks, > >Tom Savor > > > > >---------------------------------------------------------------------- >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
