Re: Trigger and Function Question

From: Shay Moreno <moreno(at)tdk(dot)dk>
To: "'Tom Lane'" <tgl(at)sss(dot)pgh(dot)pa(dot)us>, jumstead(at)centurytel(dot)net
Cc: pgsql-novice(at)postgresql(dot)org
Subject: Re: Trigger and Function Question
Date: 2001-04-19 10:38:06
Message-ID: 5D38890DAE46D41186A800508B6F7187026FFD9A@oestmb010-b0.tdk.dk
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-novice

> Looks like you deleted and recreated the function, without also deleting
> and recreating the trigger that refers to it. Now the trigger is
> referring to a no-longer-existent function OID ... remake the trigger
> and try again.

Excuse other preposterous question of mine, but the reliance of OID
don't go against the Information Principle for relational databases? Even
if PosgreSQL isn't strictly relational, and if triggers and functions aren't
really in the scope of the relational model, what is the rationale for such
exposition of a fairly internal construct as an OID? Perhaps the use of the
name of the function in the internal reference from the trigger would be too
expensive performace-wise, but then perhaps the function could be at least
marked invalid, like the ORA-04063 error in
http://technet.oracle.com./doc/server.805/a58312/newcha25.htm#23933, in
order to make troubleshooting simpler.

Just my 2c.

--
_
/ \ Leandro Guimarães Faria Corcete Dutra +55 (11) 3040 8913
\ / Amdocs Brasil Ltda at Tele Danmark +45 3387 5214
X http://terravista.pt./Enseada/1989/ mailto:leandrod(at)amdocs(dot)com
/ \ Campanha fita ASCII mailto:moreno(at)tdk(dot)dk

Browse pgsql-novice by date

  From Date Subject
Next Message Peter Eisentraut 2001-04-19 15:47:58 Re: < Int4 - limit >
Previous Message Tom Lane 2001-04-19 04:58:04 Re: Trigger and Function Question