Postgres Pro
Downloads
Home   >   mailing lists

wierd AND condition evaluation for plpgsql - Mailing list pgsql-hackers

From Louis-David Mitterrand
Subject wierd AND condition evaluation for plpgsql
Date
Msg-id 20020528072042.GA7165@apartia.org
Whole thread Raw
Responses Re: wierd AND condition evaluation for plpgsql  ("Joel Burton" <joel@joelburton.com>)
Re: wierd AND condition evaluation for plpgsql  (Tom Lane <tgl@sss.pgh.pa.us>)
Re: wierd AND condition evaluation for plpgsql  (Peter Eisentraut <peter_e@gmx.net>)
List pgsql-hackers
Tree view 
Hi,

I just noticed plpgsql evaluates all AND'ed conditions even if the first
one fails. Example:
elsif TG_OP = ''UPDATE'' and old.type_reponse = ''abandon''

This will break stuff if the trigger is used on INSERT as
"old.type_reponse" will be substituted and return an error.

Shouldn't plpgsql shortcut AND conditions when a previous one fails, as
perl does?

--    OENONE: Quoi ?   PHEDRE: Je te l'ai prédit, mais tu n'as pas voulu.
(Phèdre,J-B Racine, acte 3, scène 3)

pgsql-hackers by date:

Previous
From: Joe Conway
Date:
Subject: Re: revised sample SRF C function; proposed SRF API
Next
From: Michael Meskes
Date:
Subject: Re: Replication status