Re: BUG #4100: GMT offsets are wrong - Mailing list pgsql-bugs

From Tom Lane
Subject Re: BUG #4100: GMT offsets are wrong
Date
Msg-id 715.1207664419@sss.pgh.pa.us
Whole thread Raw
In response to Re: BUG #4100: GMT offsets are wrong  (Magnus Hagander <magnus@hagander.net>)
List pgsql-bugs
Magnus Hagander <magnus@hagander.net> writes:
> Peter Coxhead wrote:
>> I find the results to selecting the time in a GMT offset are
>> backwards, so

> I think you're being bitten by the stupid rules of POSIX timezones.
> They declare + as being west of GMT, and - as east. Which is not what
> you'd expect, but it's what the standard says.

Well, it's just as sane as the other way.  The problem is that POSIX
specified one sign convention and ISO-8601 chose the other one ...
so we're kinda stuck here.

The general rule is that PG follows the ISO convention for everything
*except* POSIX-style timezone names, where obviously we must do what
POSIX said.  I'm not entirely certain if we were consistent about this
as far back as 7.4, though.  Our timezone code got pretty heavily
revamped in 8.0.

            regards, tom lane

pgsql-bugs by date:

Previous
From: "Mikael Omma"
Date:
Subject: BUG #4091: ECPG $select incompability with Informix
Next
From: Bruce Momjian
Date:
Subject: Re: BUG #4091: ECPG $select incompability with Informix