Horton

Re: on placeholder entries in view rule action query's range table

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Justin Pryzby <pryzby@telsasoft.com>
Cc: Andrew Dunstan <andrew@dunslane.net>, Amit Langote <amitlangote09@gmail.com>, vignesh C <vignesh21@gmail.com>, Alvaro Herrera <alvherre@alvh.no-ip.org>, pgsql-hackers@lists.postgresql.org
Date: 2023-01-12T16:01:42Z
Lists: pgsql-hackers 
Justin Pryzby <pryzby@telsasoft.com> writes:
> What about also including a dump from an old version, too ?
> Then the upgrade test can test actual upgrades.

The BF clients already do that (if enabled), but they work from
up-to-date installations of the respective branch tips.  I'd not
want to have some branches including hypothetical output of
other branches, because it'd be too easy for those files to get
out of sync and deliver misleading answers.

			regards, tom lane

Commits

  1. Acquire locks on views in AcquirePlannerLocks, too.

  2. Get rid of the "new" and "old" entries in a view's rangetable.

  3. vacuumlazy.c: Save get_database_name() in vacrel.