From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
cc: Corey Huinker <corey.huinker@gmail.com>,
        jian he <jian.universality@gmail.com>, Tomas Vondra <tomas@vondra.me>,
        pgsql-hackers@lists.postgresql.org
Subject: Re: Extended Statistics set/restore/clear functions.
In-reply-to: <aTIzH-2Spb4H6BNP@paquier.xyz>
References: 
 <CACJufxGGzkb58BU+YyTa9cBAawhybwk2cPFZ1XupS-8xuAzN9A@mail.gmail.com>
 <aRwGeN7o22HJclg5@paquier.xyz>
 <CADkLM=czQ0UUyBn3AiHZfJJWRwXH5e2xn056yG0mpuXpW09RSw@mail.gmail.com>
 <CACJufxHVKiVEvCsc11oxOM=0GgjqChFwi-ii4RJuw95ZFB_fHQ@mail.gmail.com>
 <CADkLM=eRop0t==hg0TU3KuDH6DgnUxQ6vZxV9th=6vW1JG36MQ@mail.gmail.com>
 <CACJufxGgEnOzFq0HjmkKBTnEUgToGFC0g0A4wPWoJMzOuoWUhw@mail.gmail.com>
 <CADkLM=crdyK3w7-EDEdDfXHGidxm2ms+bQYNFOZUD4eOLuLXEQ@mail.gmail.com>
 <aSVas5CIr1mvR2VK@paquier.xyz> <aSZgkssl7NT1jI2h@paquier.xyz>
 <1341064.1764895052@sss.pgh.pa.us> <aTIzH-2Spb4H6BNP@paquier.xyz>
Comments: In-reply-to Michael Paquier <michael@paquier.xyz>
	message dated "Fri, 05 Dec 2025 10:19:27 +0900"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <1353178.1764901790.1@sss.pgh.pa.us>
Date: Thu, 04 Dec 2025 21:29:50 -0500
Message-ID: <1353179.1764901790@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/1353179.1764901790%40sss.pgh.pa.us>
Precedence: bulk

Michael Paquier <michael@paquier.xyz> writes:
> On Thu, Dec 04, 2025 at 07:37:32PM -0500, Tom Lane wrote:
>> This of course arises because the source code is passing the address
>> of a local ErrorSaveContext variable to that macro.  We don't have
>> any other instances of that coding pattern AFAICS, so I wonder if
>> that was really the most adapted way to do it.

> Thanks for the report.  Right.  There are three instances of that in
> pg_dependencies.c, two in pg_ndistinct.c.  I would not mind doing the
> attached to calm down these warnings, matching with the other areas of
> the code, that simply removes the macro and checks the state value
> directly.

Ah, right, now that you mention it, 56b1e88c8 did this same fix
before.  This way is fine with me.

I didn't review the error message changes closely.  I do wonder if
some of them are can't-happen cases that could just use an elog()
instead of a translatable message.

			regards, tom lane