On Mon, 18 Aug 2008, Ton Voon wrote:
I can understand the arguments about EST, but if there's no ambiguity about
CET, would it be correct to add that in? I note that EST, MST and HST are
supported timezones: http://search.cpan.org/dist/DateTime-TimeZone/.
If it is just a case of adding CET in, I'll happily spend the time to do it,
provide a patch and document the steps required. But I'm just trying to
understand the arguments against this first.
The EST, MST, and HST aliases all come from the Olson database. I have
added a few aliases myself in the DT::TZ distro, but in this case I'd be a
little leary.
The problem is that CET could map to many, many different time zones
including Europe/Paris, Europe/Vienna, Europe/Tirane, and many more.
At the present time, all of these time zones follow the same rules, but
they did not do so in the past, and there's no guarantee that they will in
the future.
What's weird is that there is a CET zone in the Olson database.
Unfortunately, it maps to the C-Eur rules, not the EU rules, and I think
they're just slightly different for modern time zones. Apparently I'd been
excluding them for some reason. I wish I remembered why.
I guess I could add them back. Whatever this Solaris box returns for CET
comes from the Olson database, so the sysadmin shouldn't be surprised by
what DT::TZ does either.
-dave
/*==========================
VegGuide.Org
Your guide to all that's veg
==========================*/
