From: | Janning Vygen <vygen(at)gmx(dot)de> |
---|---|
To: | pgsql-general(at)postgresql(dot)org |
Cc: | "Philippe Lang" <philippe(dot)lang(at)attiksystem(dot)ch>, "Tom Lane" <tgl(at)sss(dot)pgh(dot)pa(dot)us> |
Subject: | Re: Trigger disactivation and SELECT WAITING |
Date: | 2005-07-27 08:01:23 |
Message-ID: | 200507271001.24012.vygen@gmx.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-general |
Am Mittwoch, 27. Juli 2005 09:47 schrieb Philippe Lang:
> Thanks Tom, thanks Janning,
>
> I found triggers very convenient to do different tasks in the database, and
> these tasks go far beyond what we can do in rules, Janning.
Right. There are some things that can't be done with rules.
> When a line is being inserted in an order, the insert trigger automatically
> inserts data in a subtable of the order line, for example. In this
> subtable, there are informations regarding the "planning" of the order.
> People can use the GUI to populate the order, but things won't break if the
> user opens the database table directly, which can happen sometimes. Without
> the trigger, an "insert function click" would have to be used each time an
> order line is being added, and this is not that effective from a
> user-experience point of view, I think. Or would require a lot a
> client-coding.
As far as i understand your example it can be done with rules, too.
> Now the use of a trigger has a drawback: when you want to duplicate an
> order, for example. During the duplication function, I would like to
> disable the trigger, in order to make a copy of the order, order lines, and
> order lines subtable data. This is much easier than keeping the trigger,
> and having to delete default data it inserts in the new order.
just a thought: maybe you can insert a column "copyof_id" in your tabel and
mark it if you copy orders. The trigger can check NEW.copyof and quit his
work if it is marked.
With rules there is a nice advantage when copying: rules are not invoked by
COPY command (but sometimes it is a disadvantage if you need the rule
invocation)
kind regards
janning
From | Date | Subject | |
---|---|---|---|
Next Message | denis | 2005-07-27 08:22:48 | Re: Postgresql with max_connections=4096 |
Previous Message | Richard Huxton | 2005-07-27 07:52:52 | Re: Postgresql with max_connections=4096 |