Re: [PATCHES] Datetime patch

From: Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us>
To: Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us>
Cc: greg(at)turnstep(dot)com, PostgreSQL-development <pgsql-hackers(at)postgreSQL(dot)org>
Subject: Re: [PATCHES] Datetime patch
Date: 2003-07-25 20:12:24
Message-ID: 200307252012.h6PKCO918514@candle.pha.pa.us
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers pgsql-patches


[ Moved to hackers.]

(The discussion is whether we should support dates of the format
yy-mm-dd. We already support yyyy-mm-dd, but we have code that would
see 97-01-01 and detect the first part was a year.)

Tom Lane wrote:
> Bruce Momjian <pgman(at)candle(dot)pha(dot)pa(dot)us> writes:
> > I have never seen YY/MM/DD, only YYYY-MM-DD.
>
> You have apparently forgotten what was standard practice just a few
> years ago.

Well, seeing as it hasn't worked for a date in three years, I don't see
how anyone could be using it unless they are only entering dates
per-2000, which seems unlikely. It will be come useful again in 2032.

> > The huge problem is
> > deciding out how to decode 03-02-01. I think we have to require the
> > century for those.
>
> No, the entire point is to drive it off datestyle, *not* off the input
> value ranges.

If they supply a four-digit year, we assume yyyy-mm-dd, if not we follow
datestyle. I can see someone wanting yy-mm-dd, but then we need a _new_
setting to control that, because the detection code for a year being >
31 just doesn't work in 2003.

I see what you are saying, that using the four-digit leading as
specifying a year is arbitrary, but it does allow us to accept both ISO
and US/European dates cleanly. I guess the question is whether it is
worth allowing yy-mm-dd using a _new_ setting. I still think we will
need the 4-digit rule for ordinary users.

The driving thing here was consistency, so the same session didn't
accept 19-8-03 and 8-19-03 while other dates like 01-01-03 were
following datestyle. I don't see how the 4-digit rule actually is
inconsistent in that way.

> > If that is the only issue, I can ask on general, but I doubt someone is
> > going to pipe up.
>
> I really dislike the idea that we are going to legislate this behavior
> in a three-person discussion on -patches. The people who will be
> screaming about it don't read -patches.

I would be shocked to find someone screaming. I have asked on general,
and if someone come up with a valid use for it, we can adjust it, even
during beta. We can't tighten during beta, but we can loosen.

--
Bruce Momjian | http://candle.pha.pa.us
pgman(at)candle(dot)pha(dot)pa(dot)us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message The Hermit Hacker 2003-07-25 20:47:44 Re: v7.3.4 bundled ...
Previous Message Tom Lane 2003-07-25 19:51:40 Re: Datetime patch

Browse pgsql-patches by date

  From Date Subject
Next Message Bruce Momjian 2003-07-25 21:48:42 Re: Minor pager corrections in print.c and help.c (psql)
Previous Message Tom Lane 2003-07-25 19:51:40 Re: Datetime patch