Nicolas Goaziou <mail@nicolasgoaziou.fr> writes:

> Hello,
>
> Eric Abrahamsen <eric@ericabrahamsen.net> writes:
>
>> Out of curiosity, what's your stance on supporting time zones in Org's
>> timestamps? It would be an enormous amount of work, obviously, but in
>> principle?
>
> In principle I agree this would be a very good thing, if possible at
> all.
>
>> That's pretty much what it was before I messed with it, and I was a
>> little confused because no callers of `org-icalendar-convert-timestamp'
>> ever used the optional fourth argument at all. My thinking was that
>> users might want to explicitly set the timezone to "UTC". To be honest
>> I'm not sure why they would, but I also don't know why we'd accept t
>> when nothing uses it. I found the whole handling of utc-or-not a bit
>> confusing. What do you think?
>
> I have no strong opinion about this. I was merely concerned about
> consistency. E.g., when using `format-time-string', I never think about
> "UTC" for the last optional argument.

As it is now, a program can call `org-icalendar-convert-timestamp' and
pass in t for the "tz" argument, and that will go to
`format-time-string' correctly. If a user sets the property to "UTC",
that will also get converted to t. Again, I'm not sure this is actually
useful, but the call to `format-time-string' should be consistent.

> BTW, as long as Org doesn't support time zones, the "ox-icalendar"
> property may be prefixed with "ICALENDAR_" (e.g., ICALENDAR_TZ).

Done! How does the attached patch look?

Eric