Aarni =?iso-8859-1?q?Ruuhim=E4ki?= <aarni@kymi.com> writes:
> So, the answer is that TIMESTAMP WITH TIMEZONE is good through 2037, and
> TIMESTAMP WITHOUT TIME ZONE is good through 10,000AD?
That was a reasonable answer at the time of the message you quote, but
PG 8.0 no longer relies on the OS for timezone support. So now, to
repeat the example:
regression=# select '1999-09-27'::timestamp with time zone;
timestamptz
------------------------
1999-09-27 00:00:00-04
(1 row)
regression=# select '2999-09-27'::timestamp with time zone;
timestamptz
------------------------
2999-09-27 00:00:00-05
(1 row)
or for that matter
regression=# select '0099-09-27 BC'::timestamp with time zone;
timestamptz
---------------------------
0099-09-27 00:00:00-05 BC
(1 row)
The points I made before about the dubiousness of this calculation are
as valid as ever, but it's no longer a matter of not being able to do
it, it's just a matter of how much space you want to expend in the
timezone data files. (The first and last dates shown above are outside
the range of the Americas/New_York timezone file that we supply, so PG
falls back to assuming local standard time.)
regards, tom lane