On Fri, 1 Jul 2005, Satoshi Nagayasu wrote:

> Hi all,
>
> Here is a first patch to allow these commands.
>
> > ALTER TABLE <table> ENABLE TRIGGER <trigname>
> > ALTER TABLE <table> DISABLE TRIGGER <trigname>
>

Hmmm.. just thinking about it for a second. I wonder if we should also
support:

ALTER TABLE DISABLE TRIGGERS

which would disable all triggers on the table. We would have a
complimentary ENABLE TRIGGERS as well, obviously. The reason I say this is
that the common case will be that people are doing a bulk load and want to
disable all triggers. However, this will be very useful for debugging
interactions between triggers on a table so a user might want to disable
only one of many triggers -- as your current grammar does.

Perhaps a way of making the grammar a little less ambiguous would be to
have the following to disable all triggers:

ALTER TABLE <table> DISABLE TRIGGERS

and the following to disable one:

ALTER TRIGGER <trigger> DISABLE


Just an idea.

Thanks,

Gavin

---------------------------(end of broadcast)---------------------------
TIP 1: subscribe and unsubscribe commands go to majordomo@postgresql.org

Hi,

Gavin Sherry wrote:
> Hmmm.. just thinking about it for a second. I wonder if we should also
> support:
>
> ALTER TABLE DISABLE TRIGGERS

I found some RDBMSes are supporting 'DISABLE TRIGGER ALL TRIGGERS' command.

Does anyone know about the SQL99 spec?

--
NAGAYASU Satoshi <nagayasus@nttdata.co.jp>

---------------------------(end of broadcast)---------------------------
TIP 7: don't forget to increase your free space map settings

On Fri, 1 Jul 2005, Satoshi Nagayasu wrote:

> Hi,
>
> Gavin Sherry wrote:
> > Hmmm.. just thinking about it for a second. I wonder if we should also
> > support:
> >
> > ALTER TABLE DISABLE TRIGGERS
>
> I found some RDBMSes are supporting 'DISABLE TRIGGER ALL TRIGGERS' command.
>
> Does anyone know about the SQL99 spec?

The spec says nothing about disabling triggers.

Gavin

---------------------------(end of broadcast)---------------------------
TIP 9: In versions below 8.0, the planner will ignore your desire to
choose an index scan if your joining column's datatypes do not
match

Gavin Sherry wrote:
> On Fri, 1 Jul 2005, Satoshi Nagayasu wrote:
>
> > Hi all,
> >
> > Here is a first patch to allow these commands.
> >
> > > ALTER TABLE <table> ENABLE TRIGGER <trigname>
> > > ALTER TABLE <table> DISABLE TRIGGER <trigname>
> >
>
> Hmmm.. just thinking about it for a second. I wonder if we should also
> support:
>
> ALTER TABLE DISABLE TRIGGERS
>
> which would disable all triggers on the table. We would have a
> complimentary ENABLE TRIGGERS as well, obviously. The reason I say this is
> that the common case will be that people are doing a bulk load and want to
> disable all triggers. However, this will be very useful for debugging
> interactions between triggers on a table so a user might want to disable
> only one of many triggers -- as your current grammar does.
>
> Perhaps a way of making the grammar a little less ambiguous would be to
> have the following to disable all triggers:
>
> ALTER TABLE <table> DISABLE TRIGGERS
>
> and the following to disable one:
>
> ALTER TRIGGER <trigger> DISABLE

The proposed syntax in TODO.detail/trigger is:

ALTER TABLE foo {DISABLE|ENABLE} TRIGGER { ALL | trigger_name [ ,... ] };

which would cleanly allow all triggers of a table to be disabled.

--
Bruce Momjian | http://candle.pha.pa.us
pgman@candle.pha.pa.us | (610) 359-1001
+ If your life is a hard drive, | 13 Roberts Road
+ Christ can be your backup. | Newtown Square, Pennsylvania 19073

---------------------------(end of broadcast)---------------------------
TIP 4: Don't 'kill -9' the postmaster