Re: CREATE CONSTRAINT TRIGGER in SQL pane (pgAdmin 1.14.1) - Mailing list pgadmin-support

On Wed, 2012-01-04 at 05:58 -0800, Miso Vanek wrote:
> On 2. Jan, 22:57 h., guilla...@lelarge.info (Guillaume Lelarge) wrote:
> > On Mon, 2011-12-19 at 10:50 +0100, Guillaume Lelarge wrote:
> > > On Thu, 2011-12-15 at 15:39 +0200, Julius Tuskenis wrote:
> > > > Hello,
> >
> > > > After installing pgAdmin 1.14.1 I have noticed that CREATE CONSTRAINT
> > > > TRIGGER statements are shown in SQL pane when selecting table from
> > > > treeview. Is this done on purpose?
> > > > Frankly, I find no real use of statements like:
> > > > CREATE CONSTRAINT TRIGGER "RI_ConstraintTrigger_73501293"
> > > >    AFTER DELETE
> > > >    ON b_dok
> > > >    FOR EACH ROW
> > > >    EXECUTE PROCEDURE "RI_FKey_cascade_del"();
> >
> > > > I think the  constraint triggers should not be shown (at least by
> > > > default). They might be a good thing for debugging but not for
> > > > administrating the DB. I support the opinion stated in
> > > >http://www.pgadmin.org/support/faq.php:
> >
> > > > "
> > > > <...>
> > > > pgAdmin III considers constraint triggers as an internal implementation
> > > > detail, not interesting for the common administrator. In fact, CREATE
> > > > CONSTRAINT TRIGGER is for backward compatibility only, and shouldn't be
> > > > used in newer scripts any more. Some tools (e.g. pgAdmin II) imply this,
> > > > by showing a ADD CONSTRAINT when reverse engineering, while actually the
> > > > constraint information in the database is missing.
> > > > Run the adddepend script, which can be found in the backend's sources
> > > > contrib/adddepend directory. [AP]"
> >
> > > > I'd be glad if they disappeared from the SQL pane.
> >
> > > > pgAdmin 1.14.1
> > > > WinXp SP3
> > > > PostgreSQL 8.3.4 on i686-pc-linux-gnu, compiled by GCC gcc (Gentoo
> > > > 4.3.2-r3 p1.6, pie-10.1.5) 4.3.2
> >
> > > System constraints triggers shouldn't appear at all if you didn't select
> > > the "Show system objects". They never should appear in the SQL pane of a
> > > table description for example.
> >
> > > User constraints triggers should always appear.
> >
> > > The fact that system constraint triggers appear wasn't done on purpose.
> > > As a matter of fact, this is a bug.
> >
> > I tried to make it appear many times, but failed. Do you have
> > self-contained test case that I could use? Thanks.
> >
> > --
> > Guillaume
> >  http://blog.guillaume.lelarge.info
> >  http://www.dalibo.com
> >   PostgreSQL Sessions #3:http://www.postgresql-sessions.org
> >
> > --
> > Sent via pgadmin-support mailing list (pgadmin-supp...@postgresql.org)
> > To make changes to your subscription:http://www.postgresql.org/mailpref/pgadmin-support
> 
> This is the problem:
> 
> http://git.postgresql.org/gitweb/?p=pgadmin3.git;a=commitdiff;h=b41f071c9642f41896693be4e29f55409d8cd9ce
> 
> -       trig_sql += wxT("NOT tgisconstraint\n");
> +       trig_sql += wxT("1=1 \n");
> 
> After changing it to
> trig_sql += wxT("NOT tgisconstraint\n");
> it seems to be ok.
> 

Yeah, I know what commit brings this issue because I did this commit.
This isn't my question.

The constraint trigger that appears in his pgadmin's window seems to be
the system trigger of a foreign key. This trigger shouldn't appear as it
is a system trigger. I created a trigger with a DELETE cascade, but I
don't see the system trigger, even with the "Show system objects"
checkbox checked. So, what I'm asking for is a test case showing the
issue.


-- 
Guillaume
http://blog.guillaume.lelarge.info
http://www.dalibo.com
PostgreSQL Sessions #3: http://www.postgresql-sessions.org