From: | Jean-Luc Lachance <jllachan(at)nsd(dot)ca> |
---|---|
To: | Alvaro Herrera <alvherre(at)dcc(dot)uchile(dot)cl> |
Cc: | SZUCS Gábor <surrano(at)mailbox(dot)hu>, pgsql-general(at)postgresql(dot)org |
Subject: | Re: ORDER BY random() LIMIT 1 slowness |
Date: | 2002-12-18 19:36:25 |
Message-ID: | 3E00CE39.3CFF7C51@nsd.ca |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Alvara,
But instead of returning an error, currval() should return last_value if
nextval() was not called (with all the caveat of couse). I think it
would be more usefull that way.
JLL
Alvaro Herrera wrote:
>
> On Wed, Dec 18, 2002 at 02:09:42PM -0500, Jean-Luc Lachance wrote:
> > OK Gabor,
> >
> > I'm the one who misunderstood.
> >
> > To me, it seem to be a bug (or at least a mis-feature) that one cannot
> > call currval() before calling nextval().
> >
> > Does anyone know why it should be like this?
>
> It doesn't make sense to call currval() if you haven't called nextval()
> before. The meaning of currval() is "the value that was last assigned
> to you". If you haven't called nextval(), there isn't a value assigned
> to you.
>
> If you want to know what was the last value the sequence gave to anyway,
> SELECT last_value FROM sequence. But be aware that this is
> non-transaction safe, non-isolatable, non-anything.
>
> --
> Alvaro Herrera (<alvherre[a]dcc.uchile.cl>)
> "Entristecido, Wutra
> echa a Freyr a rodar
> y a nosotros al mar" (cancion de Las Barreras)
From | Date | Subject | |
---|---|---|---|
Next Message | scott.marlowe | 2002-12-18 19:45:16 | Re: Measuring CPU time use? (Another stupid question) |
Previous Message | Joe Conway | 2002-12-18 19:35:11 | Re: Measuring CPU time use? (Another stupid question) |