From: | Andres Freund <andres(at)anarazel(dot)de> |
---|---|
To: | Michael Paquier <michael(at)paquier(dot)xyz> |
Cc: | Peter Eisentraut <peter(dot)eisentraut(at)2ndquadrant(dot)com>, pgsql-hackers <pgsql-hackers(at)postgresql(dot)org> |
Subject: | Re: ALTER TABLE on system catalogs |
Date: | 2018-08-20 10:48:25 |
Message-ID: | 20180820104825.a7yryvlycwimgttf@alap3.anarazel.de |
Views: | Raw Message | Whole Thread | Download mbox | Resend email |
Thread: | |
Lists: | pgsql-hackers |
On 2018-08-20 11:37:49 +0900, Michael Paquier wrote:
> On Fri, Jul 13, 2018 at 11:05:10AM +0200, Peter Eisentraut wrote:
> > After reviewing that thread, I think my patch would still be relevant.
> > Because the pending proposal is to not add TOAST tables to some catalogs
> > such as pg_attribute, so my original use case of allowing ALTER TABLE /
> > SET on system catalogs would still be broken for those tables.
>
> Something has happened in this area in the shape of 96cdeae, and the
> following catalogs do not have toast tables: pg_class, pg_index,
> pg_attribute and pg_largeobject*. While 0001 proposed upthread is not
> really relevant anymore because there is already a test to list which
> catalogs do not have a toast table. For 0002, indeed the patch is still
> seems relevant. The comment block at the beginning of
> create_toast_table should be updated.
I still object to the approach in 0002.
Greetings,
Andres Freund
From | Date | Subject | |
---|---|---|---|
Next Message | Amit Khandekar | 2018-08-20 11:10:17 | Re: partitioning - changing a slot's descriptor is expensive |
Previous Message | Andres Freund | 2018-08-20 10:46:22 | Re: remove ATTRIBUTE_FIXED_PART_SIZE |