Date: Wed, 7 Dec 2022 20:39:30 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: Pavel Luzanov <p.luzanov@postgrespro.ru>
Cc: Nathan Bossart <nathandbossart@gmail.com>, pgsql-hackers@postgresql.org
Subject: Re: add \dpS to psql
Message-ID: <20221208023929.GB27893@telsasoft.com>
References: <20221206193606.GB3078082@nathanxps13>
 <661148f4-c7f1-dec1-2bc8-29f3bd58e242@postgrespro.ru>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <661148f4-c7f1-dec1-2bc8-29f3bd58e242@postgrespro.ru>
User-Agent: Mutt/1.9.4 (2018-02-28)
Archived-At: 
 <https://www.postgresql.org/message-id/20221208023929.GB27893%40telsasoft.com>
Precedence: bulk

On Wed, Dec 07, 2022 at 10:48:49AM +0300, Pavel Luzanov wrote:
> Furthermore. The VACUUM privilege allows you to also execute VACUUM FULL.
> VACUUM and VACUUM FULL are commands with similar names, but work completely
> differently.
> It may be worth clarifying on this page:
> https://www.postgresql.org/docs/devel/ddl-priv.html
> 
> Something like: Allows VACUUM on a relation, including VACUUM FULL.

Since (as you said) they work completely differently, I think it'd be
more useful if vacuum_full were a separate privilege, rather than being
included in vacuum.  And cluster could be allowed whenever vacuum_full
is allowed.

> There is a very similar command to VACUUM FULL with a different name -
> CLUSTER.  The VACUUM privilege does not apply to the CLUSTER command.
> This is probably correct.

I think if vacuum privilege allows vacuum full, then it ought to also
allow cluster.  But I suggest that it'd be even better if it doesn't
allow either, and there was a separate privilege for those.

Disclaimer: I have not been following these threads.

-- 
Justin