From: "Regina Obe" <lr@pcorp.us>
To: "'David Rowley'" <david.rowley@2ndquadrant.com>
Cc: "'PostgreSQL-development'" <pgsql-hackers@postgresql.org>
References: <001501d40f88$75186950$5f493bf0$@pcorp.us>
 <CAKJS1f9r9-u16wEkQT2sKh5wRf3DJVVPSwoNAoJnnyqhH39dDg@mail.gmail.com>
In-Reply-To: 
 <CAKJS1f9r9-u16wEkQT2sKh5wRf3DJVVPSwoNAoJnnyqhH39dDg@mail.gmail.com>
Subject: RE: Regression on PostgreSQL 10 ORDER/GROUP BY expression not found
 in targetlist
Date: Fri, 29 Jun 2018 13:30:03 -0400
Message-ID: <000701d40fce$d1098a50$731c9ef0$@pcorp.us>
MIME-Version: 1.0
Content-Type: text/plain;
	charset="utf-8"
Content-Transfer-Encoding: quoted-printable
Thread-Index: AQJsRSJPxw50c9w80aArPo91dZMGXgJZOOQ+ozRDeMA=
Content-Language: en-us
Precedence: bulk

> On 29 June 2018 at 21:06, Regina Obe <lr@pcorp.us> wrote:
> > I think the exact query that is triggering it is this one though I
> > don't have PostgreSQL 10 head installed locally to confirm.
> >
> > SELECT 't3280', 'L1b' || topology.TopoGeo_AddLinestring('bug3280',
> > geom)  FROM bug3280.edge where edge_id =3D 1  ORDER BY 1;
>=20
> Can you try in REL_10_STABLE? There was a fix committed [1] for =
something
> very similar just a week ago.
>=20
> [1]
> =
https://git.postgresql.org/gitweb/?p=3Dpostgresql.git;a=3Dcommit;h=3Da4c9=
5b0b80
> c70677c09c0d5c82a6fba875160288
>=20
> --
>  David Rowley                   http://www.2ndQuadrant.com/
>  PostgreSQL Development, 24x7 Support, Training & Services

Thanks for this input.  I guess I was mistaken about the commit that =
caused this
and it's most likely this one since this one fails as well.

Here is a trivial example to exercise that doesn't involve PostGIS.

CREATE TABLE edge AS SELECT 1 AS edge_id;
SELECT 't3280', 'L1b' || generate_series(1,10)=20
FROM edge
ORDER BY 1;


Now that said, I think I should change our test since it's using =
questionable assumptions anyway.
My only concern with this issue is that it's behavior that was allowed =
before in 10 a stable branch
and it's changing in a micro release.

I doubt few others would be bitten by though, but probably still =
worthwhile to put a note about it in the upcoming release notes
if this new behavior is kept.

Thanks,
Regina Obe