Re: Inputting relative datetimes - Mailing list pgsql-hackers

From Tom Lane
Subject Re: Inputting relative datetimes
Date
Msg-id 23049.1314728712@sss.pgh.pa.us
Whole thread Raw
In response to Re: Inputting relative datetimes  (Robert Haas <robertmhaas@gmail.com>)
List pgsql-hackers
Robert Haas <robertmhaas@gmail.com> writes:
> Well, I'm fine with not back-patching it, but think the existing
> behavior is flat wrong.

I'm not arguing that this way isn't better, just that it's different.
There have been zero user complaints about this behavior since Tom
Lockhart put it in, more than ten years ago.  That sort of militates
against a hard-line "it's flat wrong" stance.

But more to the point, since there wasn't an error before and there
isn't an error now, this is just a silent behavioral change, and we
avoid doing those in released branches.  People don't want to have to
retest their applications against minor releases.
        regards, tom lane


pgsql-hackers by date:

Previous
From: Heikki Linnakangas
Date:
Subject: Re: WIP: Fast GiST index build
Next
From: Peter Eisentraut
Date:
Subject: Re: pg_upgrade automatic testing