| From: | Greg Stark <gsstark(at)mit(dot)edu> |
|---|---|
| To: | pgsql-general(at)postgresql(dot)org |
| Subject: | Re: TIMESTAMP WITH( OUT)? TIME ZONE indexing/type choice... |
| Date: | 2003-02-18 15:56:54 |
| Message-ID: | 87k7fx4m61.fsf@stark.dyndns.tv |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-general |
Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> writes:
> Greg Stark <gsstark(at)mit(dot)edu> writes:
> > The documentation I've read makes it sound like these two data types are
> > equivalent in every way except for the default timezone assumed interpretation
> > when converting to and from textual representations. Is that not true?
>
> I wouldn't think so. For example, you get dissimilar results near
> daylight-savings-time boundaries:
Well how is that different from just saying the timestamp is always in GMT?
The confusing part is what happens when you cast from a timestamptz to a
timestamp. It doesn't seem to adjust for the current time zone of the
timestamptz, it just drops it.
> timestamp just stores the nominal HH:MM:SS value you give it, with no sense
> that it knows what time that really is, and no attempt to correct for
> different local timezones nor for daylight-savings changes.
Ok, I guess I understand now the difference between timestamp and timestamptz,
I just don't see what use a timestamp that doesn't represent a particular time
would ever be. It seems to serve only as a gotcha for unwary programmers who
take the default.
--
greg
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tim Stump | 2003-02-18 16:19:30 | Re: nodeRead: did not find '}' at end of plan node |
| Previous Message | Mike Mascari | 2003-02-18 15:55:50 | Re: Index not used with IS NULL |