Date: Fri, 16 Jul 2021 17:57:07 -0400
From: Alvaro Herrera <alvherre@alvh.no-ip.org>
To: Justin Pryzby <pryzby@telsasoft.com>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@postgresql.org,
	Amit Langote <amitlangote09@gmail.com>
Subject: Re: CREATE TABLE .. PARTITION OF fails to preserve tgenabled for
 inherited row triggers
Message-ID: <202107162157.53swi4o7wo2q@alvherre.pgsql>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: <202107142345.2mdefvxfzhjy@alvherre.pgsql>
Archived-At: 
 <https://www.postgresql.org/message-id/202107162157.53swi4o7wo2q%40alvherre.pgsql>
Precedence: bulk

I noticed that in pg_dump --clean, we were still emitting DROP lines for
the triggers in the partitions, which raises errors; so I emptied that
too.

On 2021-Jul-14, Alvaro Herrera wrote:

> Looking over the tests added by your (Justin's) patch, there was a
> problem checking for non-existance of the CREATE TRIGGER line: it
> doesn't work to use "like => {}" (empty), you need to use
> "unlike => { everything }".  So your test was not catching the fact that
> we were, in fact, emitting the undesirable line.

Actually, that is wrong; unlike is there just to be able to subtract
dumps from the set that is going to be searched for the regexp, for when
you want to use one of the hashes (%full_runs) but ignore a few of
those.

> I have fixed that here, and verified that the tests are doing what we
> wanted them to do.

... and I added some more tests, for pg_upgrade.

-- 
Álvaro Herrera         PostgreSQL Developer  —  https://www.EnterpriseDB.com/
"La espina, desde que nace, ya pincha" (Proverbio africano)