| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> | 
|---|---|
| To: | "Robert Kruuus" <robert(dot)kruus(at)gov(dot)sk(dot)ca> | 
| Cc: | pgsql-bugs(at)postgresql(dot)org | 
| Subject: | Re: BUG #4789: ERROR 22008 on timestamp import | 
| Date: | 2009-05-01 17:49:35 | 
| Message-ID: | 15728.1241200175@sss.pgh.pa.us | 
| Views: | Whole Thread | Raw Message | Download mbox | Resend email | 
| Thread: | |
| Lists: | pgsql-bugs | 
"Robert Kruuus" <robert(dot)kruus(at)gov(dot)sk(dot)ca> writes:
> More generally, even 
> SELECT '1999-08-06 00:12:57.9999999Z'::timestamp with time zone; will throw
> the same error, so it seems to be something in the way Postgres
> rounds/truncates the field.
Hmph.  Is your installation built with --enable-integer-datetimes?
(Try "show integer_datetimes;" if you're not sure.)
On my devel machine, I get this in 8.3
regression=# select '1999-08-06 00:12:57.999999900Z'::timestamptz;
      timestamptz       
------------------------
 1999-08-05 20:12:58-04
(1 row)
and this in 8.4
regression=# select '1999-08-06 00:12:57.999999900Z'::timestamptz;
ERROR:  date/time field value out of range: "1999-08-06 00:12:57.999999900Z"
LINE 1: select '1999-08-06 00:12:57.999999900Z'::timestamptz;
               ^
but I bet it's the change in the default integer_datetimes setting
that is the relevant difference.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Kevin Grittner | 2009-05-01 17:54:36 | Re: BUG #4789: ERROR 22008 on timestamp import | 
| Previous Message | Robert Kruuus | 2009-05-01 16:57:02 | BUG #4789: ERROR 22008 on timestamp import |