From: | Tom Lane <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Subject: | Re: Question on trigger data visibility |
Date: | 2010-08-30 16:08:38 |
Message-ID: | 8485.1283184518@sss.pgh.pa.us |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Terry Lee Tucker <terry(at)chosen-ones(dot)org> writes:
> On Monday, August 30, 2010 07:20:14 Maurice Gittens wrote:
>> So, it seems that even though my trigger is defined as AFTER INSERT
>> FOR EACH STATEMENT, the inserted row
>> does not appear to be included in the join.
>>
>> So, now to my question: Should, as a matter of principle, statement
>> level triggers not "see" rows recently inserted into the tablex?
> They do "see" those rows. Are you sure that the inner join with tab_Y is not
> causing the problem? Just a guess...
It also seems worth pointing out that this trigger would hardly ensure
referential integrity. As quoted, it would succeed so long as there is
at least one tablex row that is properly referencing some tablez row.
Surely you want to require that they *all* do.
regards, tom lane
From | Date | Subject | |
---|---|---|---|
Next Message | Alvaro Herrera | 2010-08-30 16:45:41 | Re: Database corruption |
Previous Message | wei725 | 2010-08-30 15:27:06 | Re: How to convert a binary filed to an integer field? |