Re: [ADMIN] Messed up time zones

From: Laszlo Nagy <gandalf(at)shopzeus(dot)com>
To: Steve Crawford <scrawford(at)pinpointresearch(dot)com>
Cc: pgsql-performance(at)postgresql(dot)org
Subject: Re: [ADMIN] Messed up time zones
Date: 2012-08-03 18:37:45
Message-ID: 501C1A79.8080707@shopzeus.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-admin pgsql-performance


> So you took two distinct points in time, threw away some critical
> information, and are surprised why they are now equal?
Well, I did not want to throw away any information. The actual
representation could be something like:

"2012-11-04 01:30:00-08 in Europe/Budapest, Winter time"

and

"2012-11-04 01:30:00-08 in Europe/Budapest, Summer time".

It would be unambiguous, everybody would know the time zone, the UTC
offset and the time value, and conversion back to UTC would be
unambiguous too.

I presumed that the representation is like that. But I was wrong. I have
checked other programming languages. As it turns out, nobody wants to
change the representation just because there can be an ambiguous hour in
every year. Now I think that most systems treat ambiguous time stamps as
if they were in standard time. And who am I to go against the main flow?
I'm sorry, I admit that the problem was in my head.

In response to

Browse pgsql-admin by date

  From Date Subject
Next Message Kevin Grittner 2012-08-03 19:03:19 Re: PostgreSQL oom_adj postmaster process to -17
Previous Message Tom Lane 2012-08-03 18:32:52 Re: PostgreSQL oom_adj postmaster process to -17

Browse pgsql-performance by date

  From Date Subject
Next Message Josh Berkus 2012-08-03 22:30:22 Re: Linux memory zone reclaim
Previous Message Steve Crawford 2012-08-03 18:25:53 Re: [ADMIN] Messed up time zones