Re: Collect ObjectAddress for ATTACH DETACH PARTITION to use in event trigger

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Amit Kapila <amit(dot)kapila16(at)gmail(dot)com>
Cc: "houzj(dot)fnst(at)fujitsu(dot)com" <houzj(dot)fnst(at)fujitsu(dot)com>, "kuroda(dot)hayato(at)fujitsu(dot)com" <kuroda(dot)hayato(at)fujitsu(dot)com>, PostgreSQL Hackers <pgsql-hackers(at)lists(dot)postgresql(dot)org>
Subject: Re: Collect ObjectAddress for ATTACH DETACH PARTITION to use in event trigger
Date: 2022-07-23 08:44:28
Views: Raw Message | Whole Thread | Download mbox | Resend email
Lists: pgsql-hackers

On Fri, Jul 22, 2022 at 02:26:02PM +0530, Amit Kapila wrote:
> Yeah, that would be a good idea but I think instead of changing
> get_altertable_subcmdtypes(), can we have a new function say
> get_altertable_subcmdinfo() that returns additional information from
> address. The other alternative could be that instead of returning the
> address as a string, we can return some fields as a set of records
> (one row for each subcommand) as we do in
> pg_event_trigger_ddl_commands().

Changing get_altertable_subcmdtypes() to return a set of rows made of
(subcommand, object description) is what I actually meant upthread as
it feels natural given a CollectedCommand in input, and as
pg_event_trigger_ddl_commands() only gives access to a set of
CollectedCommands. This is also a test module so
there is no issue in changing the existing function definitions.

But your point would be to have a new function that takes in input a
CollectedATSubcmd, returning back the object address or its
description? How would you make sure that a subcommand maps to a
correct object address?

In response to


Browse pgsql-hackers by date

  From Date Subject
Next Message Thomas Munro 2022-07-23 08:47:48 Re: MultiXact\SLRU buffers configuration
Previous Message Andrey Borodin 2022-07-23 08:39:50 Re: MultiXact\SLRU buffers configuration