Re: cache problem (v2) - Mailing list pgsql-admin

From Andrew Sullivan
Subject Re: cache problem (v2)
Date
Msg-id 20070717152848.GD21688@phlogiston.dyndns.org
Whole thread Raw
In response to Re: cache problem (v2)  (De Leeuw Guy <G.De_Leeuw@eurofer.be>)
Responses Re: cache problem (v2)
List pgsql-admin
On Tue, Jul 17, 2007 at 07:59:32AM +0200, De Leeuw Guy wrote:
> the COPY start a transaction

Yes.  _Everything_ is in a transaction in PostgreSQL.  If you don't
explicitly do BEGIN. . .COMMIT, then the system does it implicitly
for each statement.

> if yes each time my trigger select a data
> and update then the value returned by the read are the value before the
> transaction. It's true ? if yes my question are : also in the same
> transaction (like my trigger) ?

Your trigger function runs inside the transaction of the calling
statement, unless you have explicitly started a transaction.

I'm starting to get the impression you have a misunderstanding of how
PostgreSQL works.  I think you maybe need to spend some more time
with the manual.  If you find the English version puzzling, you can
also work in French (at http://docs.postgresqlfr.org/) or German (I
see a translation at
http://www.postgresql.org/docs/books/pghandbuch.html.de; I dunno if
there's a more recent one).  Sorry, AFAIK there isn't one in Dutch.

A

--
Andrew Sullivan  | ajs@crankycanuck.ca
The whole tendency of modern prose is away from concreteness.
        --George Orwell

pgsql-admin by date:

Previous
From: Tom Lane
Date:
Subject: Re: compiling C Funktion for PostgreSQL
Next
From: "Abraham, Danny"
Date:
Subject: plpgsql debugger