Date: Fri, 17 Oct 2025 14:39:15 -0500
From: Nathan Bossart <nathandbossart@gmail.com>
To: Peter Geoghegan <pg@bowt.ie>
Cc: Tom Lane <tgl@sss.pgh.pa.us>, pgsql-hackers@postgresql.org,
	david@justatheory.com, Andrew Dunstan <andrew@dunslane.net>
Subject: Re: abi-compliance-check failure due to recent changes to
 pg_{clear,restore}_{attribute,relation}_stats()
Message-ID: <aPKbY4vzdNWoKloK@nathan>
References: <aPJ03E2itovDBcKX@nathan>
 <1713509.1760721320@sss.pgh.pa.us>
 <aPKFmL8VjOKNVLcw@nathan>
 <1723302.1760726712@sss.pgh.pa.us>
 <aPKUzk2xFhZBbWmz@nathan>
 <CAH2-WznQY7UaGGdAgg0crnJ8j1R3MW8rMqEynDBMTio+7WAYXA@mail.gmail.com>
 <aPKY4fJVgKrX76p-@nathan>
 <CAH2-Wzne4pkk+jEg0vQC8qSdqHHqt64HHRgkzud3hvkGWKNnjA@mail.gmail.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=utf-8
Content-Disposition: inline
Content-Transfer-Encoding: 8bit
In-Reply-To: 
 <CAH2-Wzne4pkk+jEg0vQC8qSdqHHqt64HHRgkzud3hvkGWKNnjA@mail.gmail.com>
Archived-At: <https://www.postgresql.org/message-id/aPKbY4vzdNWoKloK%40nathan>
Precedence: bulk

On Fri, Oct 17, 2025 at 03:33:28PM -0400, Peter Geoghegan wrote:
> On Fri, Oct 17, 2025 at 3:28 PM Nathan Bossart <nathandbossart@gmail.com> wrote:
>> I was imagining this working more like what Tom suggested.  IOW we'd use
>> the latest commit listed in the file (perhaps always the first one) as the
>> baseline.
> 
> You said "I suppose this idea is entirely dependent on the maintainers
> of the abi-compliance-check code to adapt to it", which I understood
> to mean that you thought that the upstream tool would somehow be made
> to accept these kinds of ignore files. Obviously I misunderstood.

Sorry, I wasn't clear there.

>> Of course, this doesn't work too well if we have a bunch of ABI
>> breaks between buildfarm checks.  But my guess is that we could deal with
>> that pretty easily (e.g., make sure the buildfarm member in question runs
>> for every commit on the stable branch).
> 
> In practice I think that it would be up to the person writing the next
> suppression to verify that there were no unrelated changes in the
> interim between their new blessed/suppression commit and the prior
> one. That doesn't seem super onerous to me, given that even false
> positives don't seem to be all that common with
> abi-compliance-checker.

Agreed.  Even if someone forgets to do that validation, the chances of
missing something seem low.

-- 
nathan