Re: PL/pgSQL : notion of deferred execution - Mailing list pgsql-general

From Tom Lane
Subject Re: PL/pgSQL : notion of deferred execution
Date
Msg-id 2099.1134401368@sss.pgh.pa.us
Whole thread Raw
In response to Re: PL/pgSQL : notion of deferred execution  (Frank van Vugt <ftm.van.vugt@foxi.nl>)
Responses Re: PL/pgSQL : notion of deferred execution
List pgsql-general
Frank van Vugt <ftm.van.vugt@foxi.nl> writes:
>> It might also work for you to do this with access rights.

> Well, actually that's kinda how it's working at the moment, however I would
> rather avoid this additional dependancy on user/role since it's really
> field/situation-dependent (see below).

After reading the explanation, this seems way too specialized a
situation to make a good argument that we need a general feature
of the sort.

You can get to where you need to be by passing an indicator in
the trigger arguments, eg,

CREATE TRIGGER ... BEFORE UPDATE ... EXECUTE PROCEDURE mytrigger(0);
CREATE TRIGGER ... AFTER  UPDATE ... EXECUTE PROCEDURE mytrigger(1);

            regards, tom lane

pgsql-general by date:

Previous
From: Tom Lane
Date:
Subject: Re: PostGreSQL 8.1.0 : out of memory during vacuum full analyze
Next
From: Frank van Vugt
Date:
Subject: Re: PL/pgSQL : notion of deferred execution