Re: Segfault when creating partition with a primary key and sql_droptrigger exists - Mailing list pgsql-hackers

From Dmitry Dolgov
Subject Re: Segfault when creating partition with a primary key and sql_droptrigger exists
Date
Msg-id CA+q6zcVnxBachcKS+_xTEhL6sg6pO48+WQJLNnC6h=LKdVxt6g@mail.gmail.com
Whole thread Raw
In response to Re: Segfault when creating partition with a primary key and sql_droptrigger exists  (Michael Paquier <michael@paquier.xyz>)
Responses Re: Segfault when creating partition with a primary key and sql_droptrigger exists  (Michael Paquier <michael@paquier.xyz>)
List pgsql-hackers
> On Wed, 26 Sep 2018 at 05:33, Michael Paquier <michael@paquier.xyz> wrote:
>
> On Tue, Sep 25, 2018 at 01:39:59PM +0200, Dmitry Dolgov wrote:
> > Shouldn't it be fixed by adding EventTriggerAlterTableStart? Judging from the
> > following call of ATController, we can just pass NULL as parsetree.
>
> Hmm.  I don't think that this is correct as this data could always be
> used to fetch a command tag, right?  It seems to me instead that we
> should pass down IndexStmt and handle things like the attached.
> Thoughts?

Yes, you're right. Although probably it's not great that the sequence of
EventTriggerAlterTableStart, EventTriggerAlterTableRelid and
EventTriggerAlterTableEnd is distributed between different functions (which
most likely is worth a comment in AlterTableInternal), but it's a minor
concern.


pgsql-hackers by date:

Previous
From: Alvaro Herrera
Date:
Subject: Re: [PATCH] Improve geometric types
Next
From: David Rowley
Date:
Subject: Re: Performance of the partitioning in the large scale