Re: BUG #1757: timestamp 'epoch' is not absolute - Mailing list pgsql-bugs

From Tom Lane
Subject Re: BUG #1757: timestamp 'epoch' is not absolute
Date
Msg-id 5178.1120769539@sss.pgh.pa.us
Whole thread Raw
In response to BUG #1757: timestamp 'epoch' is not absolute  ("Steve Bennett" <S.Bennett@lancaster.ac.uk>)
List pgsql-bugs
"Bennett, Steve" <s.bennett@lancaster.ac.uk> writes:
>>> It appears that the timestamp 'epoch' is being interpreted as
>>> relative to the local timezone, rather than being an absolute time.
>>
>> This is correct for timestamp ... you want timestamp with time zone.

> Sorry, but that's dumb (IMHO). The unix epoch is not relative to the
> local timezone.

Sorry, but you're out of luck on that.  A timestamp without time zone
cannot represent any absolute time at all --- assuming that it does
amounts to ascribing a timezone to it, which we don't do in general.

Perhaps it would make more sense to refuse the "epoch" keyword in the
context of timestamp without timezone ...

>   create function epoch(integer)
>     returns timestamp with time zone
>     as 'select timestamp with time zone \'epoch\'
>         + interval \'1 second\' * $1;'
>     language sql immutable;

> Is there a better way?

You might want to call this to_timestamp(), since that's what it's going
to be called in 8.1 ;-)

            regards, tom lane

pgsql-bugs by date:

Previous
From: "Jim C. Nasby"
Date:
Subject: FreeBSD 4.x pthreads issues
Next
From: Volkan YAZICI
Date:
Subject: Re: BUG #1756: PQexec eats huge amounts of memory