Re: Half-applied UPDATE rule on view

From: Dean Rasheed <dean(dot)a(dot)rasheed(at)gmail(dot)com>
To: "A(dot)M(dot)" <agentm(at)themactionfaction(dot)com>
Cc: pgsql-general(at)postgresql(dot)org
Subject: Re: Half-applied UPDATE rule on view
Date: 2010-07-03 07:54:10
Message-ID: AANLkTikwvwm-Oisv0PczQI00cuiL5l9W_oxHNRBb0N4N@mail.gmail.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-general

On 2 July 2010 23:27, A.M. <agentm(at)themactionfaction(dot)com> wrote:
> Hello,
>
> I have encountered an odd behavior involving rules which the following script demonstrates (in postgresql 8.4.3). Note that at the end of the run, the "dud" table contains one row "spam1" when the update rule clearly contains two inserts to the "dud" table. It seems that the update rule on "test.job" cuts off execution after the first update executes (and succeeds)- why?

The problem is that after the first update, "deprecated" is non-NULL
and so no longer matches the view definition. All subsequent actions
in the rule are combined with the view definition, and so find no
matching rows.

Rules are total pain, full of gotchas like this. You're almost
certainly better off using triggers on your tables. You could patch
your rule by moving the update that marks the row as deprecated to the
end and changing its WHERE clause, but really you're just inviting
further pain by continuing to use rules IMO.

Regards,
Dean

In response to

Browse pgsql-general by date

  From Date Subject
Next Message Martijn van Oosterhout 2010-07-03 08:43:31 Re: C-Functions using SPI - Missing Magic Block
Previous Message Craig Ringer 2010-07-03 01:51:49 Re: New DB-design - help and documentation pointers appreciated