On Tue, Feb 7, 2017 at 3:11 AM, Okano, Naoki <okano.na...@jp.fujitsu.com> wrote:
> On Wednesday, November 16, 2016 4:31 PM Okano Naoki wrote:
>> > But in any case it would be a serious mistake to do this without first
>> > implementing CREATE OR REPLACE TRIGGER.  I think that's an entirely 
>> > separate
>> > proposal and you would be well advised to treat it as such.
>> I see. There are more problems than I expected...
>> Let me start with 'OR REPLACE' clause.
> I tried to cretae a patch for CREATE OR REPLACE TRIGGER.
> An example of execution is shown below.
> example)
> 1.define a new trigger
>   CREATE TRIGGER regular_trigger
>        AFTER UPDATE ON table_name
>        REFERENCING OLD TABLE AS oldtable_1 NEW TABLE AS newtable_1
>        EXECUTE PROCEDURE func_1();
> 2.redinfe a trigger in single command
>  CREATE OR REPLACE TRIGGER regular_trigger
>        AFTER UPDATE OR DELETE ON table_name
>        REFERENCING OLD TABLE AS oldtable_2 NEW TABLE AS newtable_2
>        FOR EACH ROW
>        EXECUTE PROCEDURE func_2();
> If the named trigger does not exist.
> a new trigger is also created by using OR REPLACE clause.
> A regular trigger cannot be replaced by a constraint trigger and vice varsa.
> because a constraint trigger has a different role from regular triger.
> Please give me feedback.

You should add this to the next CommitFest so it doesn't get missed.


Robert Haas
EnterpriseDB: http://www.enterprisedb.com
The Enterprise PostgreSQL Company

Sent via pgsql-hackers mailing list (pgsql-hackers@postgresql.org)
To make changes to your subscription:

Reply via email to