Message-ID: <47AAB8B6.70807@archonet.com>
Date: Thu, 07 Feb 2008 07:52:22 +0000
From: Richard Huxton <dev@archonet.com>
User-Agent: Thunderbird 2.0.0.9 (X11/20071031)
MIME-Version: 1.0
To: Tiziano Slack <slack83@hotmail.it>
CC: pgsql-sql@postgresql.org
Subject: Re: TG_TABLE_NAME as identifier
References: <BAY109-W22937ED37F3081CAEC9F26CE2E0@phx.gbl>
In-Reply-To: <BAY109-W22937ED37F3081CAEC9F26CE2E0@phx.gbl>
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit

Tiziano Slack wrote:
> Thanks for the precious tips, but unfortunately the syntax:
> EXECUTE 'INSERT INTO '|| TG_TABLE_NAME ||' SELECT '|| NEW.*;is parsed correctly, but when the trigger is fired the following error is returned
> 
> ERROR:  column "*" not found in data type mytable
> CONTEXT:  SQL statement "SELECT  'INSERT INTO '||  $1  ||' SELECT '||  $2 .*"
> 
> Using 
> EXECUTE 'INSERT INTO '|| TG_TABLE_NAME ||' SELECT NEW.*';
> instead, returns:
> 
> ERROR:  NEW used in query that is not in a rule
> CONTEXT:  SQL statement "INSERT INTO anag_referenti SELECT NEW.*"
> 
> Hope someone can show me another way to do that, if exists ;)

You'll have to build the INSERT in full - column by column.

Have you considered plperl / pltcl etc. - that makes it easier to do.

-- 
   Richard Huxton
   Archonet Ltd