Horton

Re: Allowing REINDEX to have an optional name

Bernd Helmle <mailings@oopsware.de>

View thread View on postgr.es
From: Bernd Helmle <mailings@oopsware.de>
To: Michael Paquier <michael@paquier.xyz>, Ashutosh Bapat <ashutosh.bapat@enterprisedb.com>
Cc: Simon Riggs <simon.riggs@enterprisedb.com>, Ashutosh Bapat <ashutosh.bapat.oss@gmail.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2022-05-11T08:00:56Z
Lists: pgsql-hackers 
Am Mittwoch, dem 11.05.2022 um 14:42 +0900 schrieb Michael Paquier:
> Agreed.  Nobody is going to remember the difference.  REINDEX's
> parsing grammar is designed to be extensible because we have the
> parenthesized flavor.  Why don't you add an option there to skip the
> catalogs, like a SKIP_CATALOG?

+1

Having an option is probably the best idea. Though we have REINDEX
SYSTEM, so i throw SKIP_SYSTEM into the ring as an alternative. This
would be consistent with the meaning of both commands/options.

	Thanks,
		Bernd

Commits

  1. Tweak a bit the new TAP tests of REINDEX DATABASE/SYSTEM

  2. Rework logic and simplify syntax of REINDEX DATABASE/SYSTEM

  3. Add more tests for REINDEX DATABASE/SYSTEM with relfilenode changes