To be pedantic, TZ does not itself contain the rules for DST. That's the domain of the timezone database, tzinfo.
Now, what I'd go for is to use tzinfo - which is well understood and well maintained - to set both the CLOCKxx and TZ parameters. That would probably require sticking it in a SYS1 dataset somewhere. On Thu, Apr 7, 2022 at 8:28 AM Paul Gilmartin < [email protected]> wrote: > On Thu, 7 Apr 2022 12:05:20 +0000, Seymour J Metz wrote: > > >I think that it would be easier to enhance the processing of CLOCKxx for > DST. There is a lot that is in CLOCKxx and not in TZ. > > > TZ is a POSIX requirrement: > < > https://pubs.opengroup.org/onlinepubs/9699919799/basedefs/V1_chap08.html#tag_08 > >. > > What's in CLOCKxx that's absent from TZ, which already contains the rules > for DST? > > But the issue that spun this thread is inconsistency. A utility taking a > single input > and generating values for both CLOCKxx and TZ might suffice if used. > > -- gil > > ---------------------------------------------------------------------- > For IBM-MAIN subscribe / signoff / archive access instructions, > send email to [email protected] with the message: INFO IBM-MAIN > -- Jay Maynard ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
