[Cosmo-dev] Timezones and the Cosmo UI

[Matthew Eernisse]

Here's what I've got:

http://www.fleegix.org/timezone/parse.html

Right now all it actually displays is the calcuated offset, but the 
parsing process identifies the correct rule/period, so the logic for 
figuring, for example, 'CDT' vs. 'CST' would be trivial.

Rough/buggy code, caveat emptor, blah blah. :)

I don't know if this will prove useful for Cosmo or not. If we just 
stick to American timezones by default, it's a 100K download -- 
presumably we could get that down to about 50K if we strip the copious 
commenting pre-download. The calculation seems pretty speedy, but Bobby 
rightly mentions the issue of duplication of code, and potential 
incompatibilities with non-Olson stuff.

However, Olson does seem to be pretty much the defacto standard, so we 
might consider using this along with aliases for the common timezones 
like US/Pacific as BCM mentions. It would reduce requests back to the 
mothership, and potentially reduce the amount of data we have to stuff 
into each calendar event we're sending back and forth.


Matthew


Brian Moseley wrote:
> On 10/19/06, Bobby Rullo <br at osafoundation.org> wrote:
> 
>> Parsing of the Olson files in JavaScript is cool, but I think we need
>> a solution that works with icalendar VTIMEZONES. Most of the
>> timezones out there are not Olson ones. They don't even "quack like a
>> duck" Example - ical calls a timezone "US/Pacific" but olson calls it
>> "America/Los Angeles".
> 
> i believe that US/Pacific is a commonly used alias for America/Los
> Angeles. i can't find a definitive source for this, but the w3c note
> on time zones (<http://www.w3.org/TR/2005/NOTE-timezone-20051013/>)
> mentions tz aliases.
> 
> reading over the wikipedia page on time zones
> (<http://en.wikipedia.org/wiki/Time_zone>) and several linked pages,
> it looks like most computer systems do in fact use olson, but some
> augment or replace the "canonical" olson tzids with more familiar
> aliases like US/Pacific.
> _______________________________________________
> cosmo-dev mailing list
> cosmo-dev at lists.osafoundation.org
> http://lists.osafoundation.org/mailman/listinfo/cosmo-dev
>