| From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
|---|---|
| To: | "Dennis Vshivkov" <walrus(at)amur(dot)ru> |
| Cc: | pgsql-bugs(at)postgresql(dot)org |
| Subject: | Re: BUG #1617: TIME WITH TIME ZONE comparison is partially broken |
| Date: | 2005-04-23 22:57:22 |
| Message-ID: | 23584.1114297042@sss.pgh.pa.us |
| Views: | Whole Thread | Raw Message | Download mbox | Resend email |
| Thread: | |
| Lists: | pgsql-bugs |
"Dennis Vshivkov" <walrus(at)amur(dot)ru> writes:
> [8.0.1|src/backend/utils/adt/date.c] timetz_cmp_internal():
> /* Primary sort is by true (GMT-equivalent) time */
> t1 = time1->time + time1->zone;
> t2 = time2->time + time2->zone;
> This code is only valid when HAVE_INT64_TIMESTAMP is NOT the case.
Fixed, thanks for the report.
Note that this fix will invalidate any existing indexes on TIMETZ
columns in --enable-integer-datetimes installations; they'll have
to be REINDEXed after application of the patch. I suspect such
indexes are pretty few and far between, so I went ahead and back-patched
the fix into back branches despite the risk of incompatibility.
regards, tom lane
| From | Date | Subject | |
|---|---|---|---|
| Next Message | Tom Lane | 2005-04-23 23:04:51 | Re: BUG #1618: EXTRACTing EPOCH from TIME WITH TIME ZONE is broken |
| Previous Message | Magnus Hagander | 2005-04-23 21:59:03 | Re: BUG #1613: Installation flaw |