Tom Lane schrieb:
> "Titus von Boxberg" <ut(at)bhi-hamburg(dot)de> writes:
> It works for me if we want to have an "NFIELDS" construct. Personally
> I'm still not convinced that we need one --- what's the use-case?
I have removed the NFIELDS construct
> I'd prefer arbitrary expression, but I suppose there's no harm in doing
> the simple case first and generalizing if there's demand.
I took the "no harm" way.
Attached please find the updated patch.
The patch is against HEAD of 050721.
I switched the syntax to your proposal, renamed the functions
in pl_comp.c and updated the sgml doc source and regression
test files accordingly.
In response to
pgsql-patches by date
|Next:||From: Petr Jelinek||Date: 2005-07-24 12:46:13|
|Subject: Re: per user/database connections limit again|
|Previous:||From: Rocco Altier||Date: 2005-07-24 05:49:26|
|Subject: Regression - GNUmakefile - pg_usleep|