Re: ON DELETE triggers don't work as documented

From: Joe Conway <mail(at)joeconway(dot)com>
To: Mike Glover <mpg4(at)mdi(dot)cc>
Cc: pgsql-bugs(at)postgresql(dot)org
Subject: Re: ON DELETE triggers don't work as documented
Date: 2002-12-11 18:18:08
Message-ID: 3DF78160.9040606@joeconway.com
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-bugs

Mike Glover wrote:
> Please enter a FULL description of your problem:
> ------------------------------------------------ The docs state (section
> 23.9):
>
> If a non-NULL value is returned then the operation proceeds with that row
> value. Note that returning a row value different from the original value of
> NEW alters the row that will be inserted or updated.
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

>
> the above suggests that returning NEW for a delete should cause the delete
> to proceed. In fact, I've found it necessary to return a record with the
> row format of the table and all empty fields.

No, actually it suggests that returning NEW should cause an *update* or
*insert* to proceed. In the case of a delete, NEW is not set. See a few lines
above:

NEW
Data type RECORD; variable holding the new database row for INSERT/UPDATE
operations in ROW level triggers.

OLD
Data type RECORD; variable holding the old database row for UPDATE/DELETE
operations in ROW level triggers.

It is perhaps confusing, but probably necessary so that a single function can
handle inserts, updates, and deletes (see TG_OP).

Joe

In response to

Browse pgsql-bugs by date

  From Date Subject
Next Message Rudy Lippan 2002-12-12 03:21:54 initdb segfaulting.
Previous Message Stephan Szabo 2002-12-11 16:37:28 Re: bug or feature?