Horton

Re: BUG #17409: Unable to alter data type of clustered column which is referenced by foreign key

Tom Lane <tgl@sss.pgh.pa.us>

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Japin Li <japinli@hotmail.com>
Cc: Michael Paquier <michael@paquier.xyz>, holly.roberts@starlingbank.com, peter@eisentraut.org, pgsql-bugs@lists.postgresql.org
Date: 2022-03-11T00:37:45Z
Lists: pgsql-bugs

Attachments

Japin Li <japinli@hotmail.com> writes:
> Attach a new patch to fix the replica identify case, also add test cases.

Now that we realize we need to de-duplicate, it seems to me we should
postpone the get_rel_name() calls so that we don't have to do that
work repeatedly; as attached.

Also, while I've not done anything about it here, the proposed test
cases seem remarkably cavalier about their choices of test table
names.  If you want to use names as generic as "parent" and "child",
they'd better be temp tables to avoid risk of conflict against other
concurrent regression tests.  But most of alter_table.sql prefers
to use names starting with "at".

			regards, tom lane

Commits

  1. Restore the previous semantics of get_constraint_index().

  2. Change get_constraint_index() to use pg_constraint.conindid