On Sat, 6 Sep 2025 13:53:47 -0500 Tim Chase via Remind-fans <[email protected]> wrote:
> Based on my reading of the way you documented it, I *think* it > shouldn't be an issue because they're $T and $U in the context of the > non-local time-zone. Uhh... no, they're $T and $U in the local time zone. I really should document how this works internally. Maybe I'll put this in the man page, but here's how it works: 1 - When Remind starts parsing the REM command, it's in the local time zone. It stays in the local time zone right up until it hits the reminder type (MSG/CAL/SPECIAL/etc) or the SATISFY keyword (which is why SATISFY must come last.) This includes staying in the local time zone as it evaluates [expressions] on the way to the MSG/SPECIAL/SATISFY etc. 2 - Now that it has collected all of the trigger information, if there is a TZ parameter, Remind forcibly relocates itself into that time zone, including adjusting its notion of now() and today() to match the new time zone. 3 - All trigger calculations are done in the new time zone, including evaluation of any OMITFUNC functions, testing the OMIT context for omitted dates, and evaluation of the SATISFY expression. 4 - Once a trigger date and time have been found, Remind forcibly relocates itself back into the local time zone, adjusting the trigger date and time as needed, and restoring now() and today() to their local values. Remind then tells itself that there's nothing to see here and to please move along. 5 - Any further processing, such as expressions after the MSG/SPECIAL/whatever, the substitution filter, SCHED function evaluation, etc. take place in the local time zone and see the trigger date/time after its adjustment to the local time zone. Regards, Dianne. _______________________________________________ Remind-fans mailing list [email protected] https://dianne.skoll.ca/mailman/listinfo/remind-fans Remind is at https://dianne.skoll.ca/projects/remind/
