From: | "Stephan Szabo" <sszabo(at)kick(dot)com> |
---|---|
To: | <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Fw: I remember why I suggested CREATE FUNCTION...AS NULL |
Date: | 2000-09-09 02:02:42 |
Message-ID: | 004501c01a02004501c01a02$0c19ad40$0c64010a@kick.comc19ad40004501c01a02$0c19ad40$0c64010a@kick.comc64010a@kick.com |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
> >> ALTER FUNCTION numpeople() AS
> >> 'SELECT COUNT(*) FROM managers' LANGUAGE 'plpgsql';
>
> > Actually, I think this would still work, because it
> > doesn't check the table name existance until it's
> > used for the first time, not when it's created.
>
> But SQL-language function bodies are checked at entry, not only
> at first use. (I consider it a deficiency of plpgsql that it
> doesn't do likewise.)
Yeah (didn't think about SQL ones). Although the deferred
creation with like a nulled out prosrc would probably solve
these problems as long as you never wanted to actually
call the function until the restore was done, so we wouldn't
want any default values, rules or triggers that might use them
to be activated.
From | Date | Subject | |
---|---|---|---|
Next Message | Alexey Raschepkin | 2000-09-09 03:22:56 | Sources from the CVS cannot be compiled |
Previous Message | Tom Lane | 2000-09-09 01:48:30 | Re: I remember why I suggested CREATE FUNCTION...AS NULL |