From: Tom Lane <tgl@sss.pgh.pa.us>
To: exclusion@gmail.com
cc: pgsql-bugs@lists.postgresql.org
Subject: Re: BUG #17351: Altering a composite type created for a partitioned
 table can lead to a crash
In-reply-to: <17351-6db1870f3f4f612a@postgresql.org>
References: <17351-6db1870f3f4f612a@postgresql.org>
Comments: In-reply-to PG Bug reporting form <noreply@postgresql.org>
	message dated "Sat, 01 Jan 2022 18:00:01 +0000"
MIME-Version: 1.0
Content-Type: text/plain; charset="us-ascii"
Content-ID: <3057470.1641072425.1@sss.pgh.pa.us>
Date: Sat, 01 Jan 2022 16:27:05 -0500
Message-ID: <3057471.1641072425@sss.pgh.pa.us>
Archived-At: 
 <https://www.postgresql.org/message-id/3057471.1641072425%40sss.pgh.pa.us>
Precedence: bulk

PG Bug reporting form <noreply@postgresql.org> writes:
> When executing the following queries:
> create table pt (a int, b int) partition by list (b);
> create table t(a pt, check (a = '(1, 2)'::pt));
> alter table pt alter column a type char(4);
> \d+ t
> The server crashes with the following stack:

Hmm.  We really ought to reject the ALTER TABLE.  We do if "pt"
is a plain table:

regression=# create table pt (a int, b int);
CREATE TABLE
regression=# create table t(a pt);
CREATE TABLE
regression=# alter table pt alter column a type char(4);
ERROR:  cannot alter table "pt" because column "t.a" uses its row type

So something is mistakenly skipping that check for partitioned
tables.

I think we're also failing to worry about the rowtype of the
constant in t's check constraint; it seems like that has to
be complained of as well, even if the underyling columns
aren't of type "pt".

			regards, tom lane