Re: timeofday() and CAST

From: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
To: Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl>
Cc: Vilson farias <vilson(dot)farias(at)digitro(dot)com(dot)br>, pgsql-general(at)postgresql(dot)org, SIMONE Carla MOSENA <simone(dot)mosena(at)digitro(dot)com(dot)br>, paulo(dot)pizarro(at)digitro(dot)com(dot)br
Subject: Re: timeofday() and CAST
Date: 2003-08-25 23:36:10
Message-ID: 21596.1061854570@sss.pgh.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl> writes:
> Hmm... I'm seeing Chile standard timezones, so it's not a matter of how
> big the country is. However, in the table at
> src/backend/utils/adt/datetime.c I'm seeing conflict between BST
> (British Summer Time), BST (Brazil Standard Time) and BST (Bering
> Standard Time), so they just picked one. However there's a timezone
> called "BRA" that's apparently "Brazil Time".

BST is not what he was worried about, though. I see the following table
in the zic sources:

# std dst
# -2:00 FNT FNST Fernando de Noronha
# -3:00 BRT BRST Brasilia
# -4:00 AMT AMST Amazon
# -5:00 ACT ACST Acre

I have added BRT, BRST, FNT, FNST to the datetime.c table, since they
don't conflict with any existing entries. ACT and ACST were there
already. We have a problem with AMT/AMST: they conflict with existing
entries for Armenia. Not sure what to do about it, but for the moment
the Armenians were there first.

I suspect in the long run we will have to back off from this fixed-table
approach and allow the set of recognized timezone names to be taken from
a configuration file. There are just too many potential conflicts.

regards, tom lane

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Martijn van Oosterhout 2003-08-26 00:17:48 Re: Database recovery after fs crash
Previous Message Dennis Gearon 2003-08-25 22:40:24 Re: move to usenet?