Message-ID: <ad4f7823df09d4a7428dd400482eca1856a2fe5a.camel@j-davis.com>
Subject: Re: pgsql: Trial fix for old cross-version upgrades.
From: Jeff Davis <pgsql@j-davis.com>
To: Tom Lane <tgl@sss.pgh.pa.us>
Cc: Andrew Dunstan <andrew@dunslane.net>, Jeff Davis
 <jdavis@postgresql.org>,  pgsql-committers@lists.postgresql.org
Date: Mon, 24 Feb 2025 08:07:54 -0800
In-Reply-To: <816167.1740278884@sss.pgh.pa.us>
References: <E1tlBJt-000LUO-16@gemulon.postgresql.org>
	 <003dc9936317ab987faa0242f8e33e1cd2fcaf57.camel@j-davis.com>
	 <40c1e76334f2baa747334c8fc513d12b8cb297bf.camel@j-davis.com>
	 <3815127.1740189601@sss.pgh.pa.us>
	 <5ebc188e731ac2b98d68459ce1a9ef3066981774.camel@j-davis.com>
	 <3892121.1740193866@sss.pgh.pa.us>
	 <976dcc37-b629-490e-a052-a057477d062f@dunslane.net>
	 <301492.1740249291@sss.pgh.pa.us> <684274.1740273302@sss.pgh.pa.us>
	 <7fe52cfdc373df817e303050f1f10f25dcdf4390.camel@j-davis.com>
	 <816167.1740278884@sss.pgh.pa.us>
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable
User-Agent: Evolution 3.44.4-0ubuntu2 
MIME-Version: 1.0
Archived-At: 
 <https://www.postgresql.org/message-id/ad4f7823df09d4a7428dd400482eca1856a2fe5a.camel%40j-davis.com>
Precedence: bulk

On Sat, 2025-02-22 at 21:48 -0500, Tom Lane wrote:
> CREATE INDEX vcharidx ON vchartmp USING GIST ( text(a) );
>=20
> The index column will be given the name "text".=C2=A0 However, it
> dumps as
>=20
> CREATE INDEX vcharidx ON public.vchartmp USING gist (((a)::text));

Thank you! I dismissed the naming issue too early, going down the path
of something related to GiST.

> and when *that* gets loaded, the index column is given the name
> "a", because FigureColname treats function-like constructs
> differently from cast-like constructs.

Ugh.

> We have dealt with some similar issues in the past, and the
> solution was to allow index columns to be referenced by
> column number not name.=C2=A0 (ALTER INDEX ... ALTER COLUMN ...
> SET STATISTICS does that, not sure if there are other places.)
> Recommend adopting the same solution here.

I'll submit a patch to the -hackers thread. There are a few minor
choices here that might get some discussion.

Regards,
	Jeff Davis