Re: BUG #5852: Function date_trunc is not IMMUTABLE - Mailing list pgsql-bugs

From Tom Lane
Subject Re: BUG #5852: Function date_trunc is not IMMUTABLE
Date
Msg-id 21812.1296139204@sss.pgh.pa.us
Whole thread Raw
In response to BUG #5852: Function date_trunc is not IMMUTABLE  ("Oleg" <serovov@gmail.com>)
List pgsql-bugs
"Oleg" <serovov@gmail.com> writes:
> Please mark function date_trunc as IMMUTABLE for using in index.

The version for timestamp without time zone already is marked immutable.
The version for timestamp with time zone can't be marked immutable,
because it isn't: its behavior depends on the timezone setting.

Depending on what it is you hope to accomplish, you might be able to do
something like this:

regression=# create table t1 (f1 timestamptz);
CREATE TABLE
regression=# create index ii on t1 (date_trunc('day', f1 AT TIME ZONE 'UTC'));
CREATE INDEX

This is immutable because the truncation happens with respect to a fixed
timezone.  (Doesn't have to be UTC, any constant timezone name will do.)

            regards, tom lane

pgsql-bugs by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Re: BUG #5853: when the walsender cannot exit when reboot slave computer
Next
From: Heikki Linnakangas
Date:
Subject: Re: 回复:Re: [BUGS] BUG #5853: when the walsender cannot exit when reboot slave computer