From: Tom Lane <tgl@sss.pgh.pa.us>
To: pgsql-hackers@lists.postgresql.org
Subject: ATTACH PARTITION seems to ignore column generation status
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <2793382.1672944799.1@sss.pgh.pa.us>
Content-Transfer-Encoding: quoted-printable
Date: Thu, 05 Jan 2023 13:53:19 -0500
Message-ID: <2793383.1672944799@sss.pgh.pa.us>
Archived-At: <https://www.postgresql.org/message-id/2793383.1672944799%40sss.pgh.pa.us>
Precedence: bulk

This does not seem good:

regression=3D# create table pp (a int, b int) partition by range(a);
CREATE TABLE
regression=3D# create table cc (a int generated always as (b+1) stored, b =
int);
CREATE TABLE
regression=3D# alter table pp attach partition cc for values from ('1') to=
 ('10'); =

ALTER TABLE
regression=3D# insert into pp values(1,100);
INSERT 0 1
regression=3D# table pp;
  a  |  b  =

-----+-----
 101 | 100
(1 row)

I'm not sure to what extent it's sensible for partitions to have
GENERATED columns that don't match their parent; but even if that's
okay for payload columns I doubt we want to allow partitioning
columns to be GENERATED.

			regards, tom lane