From: Tom Lane <tgl@sss.pgh.pa.us>
To: sander.evers@topicus.nl
cc: pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #18662: ORDER BY after GROUPING SETS does not order correctly
 for certain WHERE condition
In-reply-to: <18662-515b2a51728c237c@postgresql.org>
References: <18662-515b2a51728c237c@postgresql.org>
Comments: In-reply-to PG Bug reporting form <noreply@postgresql.org>
	message dated "Thu, 17 Oct 2024 12:56:39 -0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <657101.1729177393.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Thu, 17 Oct 2024 11:03:13 -0400
Message-ID: <657102.1729177393@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/657102.1729177393%40sss.pgh.pa.us>
Precedence: bulk

PG Bug reporting form <noreply@postgresql.org> writes:
> I'm seeing inconsistent results for an ORDER BY in combination with GROU=
PING
> SETS.

Yup.  Your example actually works correctly as of HEAD (v18-to-be):

regression=3D# select * from (values (1, 1), (1, 2), (2,1), (2,2)) as t (a=
,b)
where a=3D1                         =

group by grouping sets ((b),(a,b))
order by a, b;
 a | b =

---+---
 1 | 1
 1 | 2
   | 1
   | 2
(4 rows)

but I'm afraid there's zero chance of back-patching the fix,
as it's far too invasive (and under-tested, for now) for that.

> This might be the same bug as
> https://www.postgresql.org/message-id/CAL48EtKDHCKnOkLdSgOmgBZBcahU2zpBq=
yzeET_ZM74uNZBFHg@mail.gmail.com

Indeed.  Richard Guo recently completed the fix I alluded to in that
thread:

https://git.postgresql.org/gitweb/?p=3Dpostgresql.git&a=3Dcommitdiff&h=3D2=
47dea89f
https://git.postgresql.org/gitweb/?p=3Dpostgresql.git&a=3Dcommitdiff&h=3Df=
5050f795

			regards, tom lane