Re: More tests with USING INDEX replident and dropped indexes

Michael Paquier <michael@paquier.xyz>

From: Michael Paquier <michael@paquier.xyz>
To: Rahila <rahila.syed@2ndquadrant.com>
Cc: Euler Taveira <euler.taveira@2ndquadrant.com>, Masahiko Sawada <masahiko.sawada@2ndquadrant.com>, Postgres hackers <pgsql-hackers@lists.postgresql.org>
Date: 2020-08-26T12:08:51Z
Lists: pgsql-hackers

Attachments

On Tue, Aug 25, 2020 at 08:59:37PM +0530, Rahila wrote:
> Sorry, I didn't apply correctly.  The tests pass for me. In addition, I
> tested with partitioned tables.  It works as expected and makes the REPLICA
> IDENTITY 'n' for the partitions as well when an index on a partitioned
> table is dropped.

Indeed.  I have added a test for that.

While looking at this patch again, I noticed that the new tests for
contrib/test_decoding/ and the improvements for relreplident are
really two different bullet points, and that the new tests should not
be impacted as long as we switch to NOTHING the replica identity once
its index is dropped.  So, for now, I have applied the new decoding
tests with a first commit, and attached is an updated patch which
includes tests in the main regression test suite for
replica_identity.sql, which is more consistent with the rest as that's
the main place where we look at the state of pg_class.relreplident.
I'll go through this part again tomorrow, it is late here.
--
Michael

Commits

  1. Add regression tests for REPLICA IDENTITY with dropped indexes