Hi Gil,
Some software vendors distribute their CLIST/Rexx PDSs/PDSEs only as FB/80.
This will cause you grief because you cannot concatenate VB and FB.
That is why I have never converted to VB CLISTs/Rexx. It's just not
worth the extra work.
(Yes, I know that you can ALTLIB, but, that is not always possible.)
I agree that BLKSIZE=3120 (or 3200) is dumb and outdated for card-image
Datasets (e.g. MACLIBs, PROCLIBs, JCLLIBs, Source Code).
One has to keep in mind, though, that cleaning up JCL does not make
money for the software vendor and introduces the need to test the new
Allocations.
There is no return on investment and there is more cost.
Regards,
David
On 2022-09-19 08:22, Paul Gilmartin wrote:
On Mon, 19 Sep 2022 04:00:05 +0000, Seymour J Metz wrote:
WTF? TSO has supported LRECL=255 for over half a century. Don't be confused by
IBM shipping code in an archaic form factor.
Alas, I am confounded, not confused, by IBM's shipping SYSLIB, SYSPROC,
etc. in an archaic form factor making concatenation of user libraries in a more
rational form factor. I somewhat circumvented that absurdity by making my
SYSEXEC RECFM=VB.
When I first encountered MVS I wondered that RECFM=VBS,LRECL=X wasn't
the modal format, because it's the most compatible with all possible data.
Or that access methods didn't convert from the format on physical media to
that requested in JCL or DCB. IBM belatedly provided that facility for UNIX
files but never extended that support to Classic data sets.
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
