Horton

Re: Make name optional in CREATE STATISTICS

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

View thread View on postgr.es
From: Tom Lane <tgl@sss.pgh.pa.us>
To: Michael Paquier <michael@paquier.xyz>
Cc: Alvaro Herrera <alvherre@alvh.no-ip.org>, Dean Rasheed <dean.a.rasheed@gmail.com>, Matthias van de Meent <boekewurm+postgres@gmail.com>, Simon Riggs <simon.riggs@enterprisedb.com>, PostgreSQL Hackers <pgsql-hackers@postgresql.org>
Date: 2022-07-23T03:54:27Z
Lists: pgsql-hackers 
Michael Paquier <michael@paquier.xyz> writes:
> I have just looked at 83011ce, and got what you've done here.  You
> have thrown away reindex_target_multitable and added three parts for
> SCHEMA, DATABASE and SYSTEM instead with their own options, enforcing
> the restriction on CONCURRENTLY at the end of REINDEX SYSTEM in the
> parser rather than indexcmds.c.

That does not seem like an improvement.  In v15:

regression=# REINDEX SYSTEM CONCURRENTLY db;
ERROR:  cannot reindex system catalogs concurrently

As of HEAD:

regression=# REINDEX SYSTEM CONCURRENTLY db;
ERROR:  syntax error at or near "CONCURRENTLY"
LINE 1: REINDEX SYSTEM CONCURRENTLY db;
                       ^

That is not a very helpful error, not even if the man page
doesn't show the syntax as legal.

			regards, tom lane

Commits

  1. Fix a few issues with REINDEX grammar

  2. Rework grammar for REINDEX

  3. parser: centralize common auxiliary productions

  4. Make the name optional in CREATE STATISTICS.