On Tue, 5 Apr 2022 11:09:55 +0800, David Crayford wrote:
>I won't be losing any sleep over it! I have seen many IBM installation
>scripts that rely on cat supporting data sets.
>
IBM may do things internally, even in customer-facing code that
are not documented; non-GUPI. Many years ago, I complained
here that
//NAME IF FALSE THEN
seems to work as expected, but is documented as "unpredictable".
Why doesn't IBM:
o make it predictable
o document it as it predictable operates
o or prohibit it by making it a JCL error.
"Unpredictable" is treacherous.
Peter Relson (IIRC) riposted that many such constructs are either
reserved for IBM internal use or for future development which may
contradict our empirical inferences about their implementation.
> ... Just for fun, here is a
>list of z/OS UNIX core-utils that use fopen().
>
[ lots of them. ] IBM is free to rescind their unsupported operation and
correspondingly modify those installation scripts. Would Peter disagree?
--
gil
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN
