From: | Decibel! <decibel(at)decibel(dot)org> |
---|---|
To: | "Javier Fonseca V(dot)" <fonsecajavier(at)gmail(dot)com> |
Cc: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Trigger Procedure Error: NEW used in query that is not in a rule |
Date: | 2007-08-15 18:28:33 |
Message-ID: | 20070815182833.GP54135@nasby.net |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general pgsql-sql |
On Sat, Aug 11, 2007 at 02:43:30AM -0500, Javier Fonseca V. wrote:
> Hello.
>
> I'm doing a Trigger Procedure in pl/pgSQL. It makes some kind of auditing.
>
> I think that it's working alright except for the next line:
>
> EXECUTE 'INSERT INTO ' || quote_ident(somedynamictablename) || ' SELECT
> new.*';
>
> PostgreSQL keeps telling me: "ERROR: NEW used in query that is not in a
> rule". I think that this NEW problem is because of the scope of the EXECUTE
> statement (outside the scope of the trigger), so it doesn't recognize the
> NEW record.
Sort-of... the issue is that EXECUTE hands the string off to the
backend, which has no clue what "NEW" is; only the trigger procedure
understands NEW.
> Maybe I could fix it concatenating column names and the 'new' values but I
> want to do my trigger as flexible as possible (I have several tables to
> audit).
>
> Somebody has any suggestion?
You could theoretically make the trigger entirely dynamic by having it
pull the needed info out of the system catalogs... but I wouldn't want
to see the performance of that... If you care about performance *at
all*, I'd suggest writing some code that will generate the triggers for
a given table for you. I don't expect it'd be much harder than writing a
completely dynamic trigger.
--
Decibel!, aka Jim Nasby decibel(at)decibel(dot)org
EnterpriseDB http://enterprisedb.com 512.569.9461 (cell)
From | Date | Subject | |
---|---|---|---|
Next Message | Leif B. Kristensen | 2007-08-15 18:31:45 | Re: pg_dump on local Windows, pg_restore on Linux? |
Previous Message | Dmitry Koterov | 2007-08-15 18:27:53 | Re: Deadlocks caused by a foreign key constraint |
From | Date | Subject | |
---|---|---|---|
Next Message | Jon Collette | 2007-08-15 18:46:02 | Boolean without default declared |
Previous Message | Jon Horsman | 2007-08-15 13:16:23 | Re: When is a shared library unloaded? |