Date: Sun, 26 Feb 2023 16:52:39 -0600
From: Justin Pryzby <pryzby@telsasoft.com>
To: pgsql-hackers@postgresql.org
Cc: Peter Eisentraut <peter@eisentraut.org>,
	Andres Freund <andres@anarazel.de>
Subject: meson vs make: missing/inconsistent ENV
Message-ID: <20230226225239.GL1653@telsasoft.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
User-Agent: Mutt/1.9.4 (2018-02-28)
Archived-At: 
 <https://www.postgresql.org/message-id/20230226225239.GL1653%40telsasoft.com>
Precedence: bulk

I noticed warnings:
Use of uninitialized value $ENV{"with_icu"} in string eq at /home/pryzbyj/src/postgres/src/bin/pg_dump/t/002_pg_dump.pl line 56.

and looked through: git grep ^export '*/Makefile'

and found that:
src/bin/pg_dump/meson.build is missing with_icu since 396d348b0

Also, e6927270c added ZSTD to src/bin/pg_basebackup/meson.build, but
it's not in ./Makefile ??  Maybe that was for consistency with other
places, or pre-emptive in case the tap tests want to do tests involving
the ZSTD tool.  But it'd be better if ./Makefile had it too.

The rest I think are not errors:

src/test/meson.build is missing PG_TEST_EXTRA
src/bin/pg_upgrade/meson.build and ../src/test/recovery/meson.build
are missing REGRESS_SHLIB

Is there any consideration of promoting these or other warnings to
fatal?

-- 
Justin