Re: More Perl cleanups

From: Michael Paquier <michael(at)paquier(dot)xyz>
To: Dagfinn Ilmari Mannsåker <ilmari(at)ilmari(dot)org>
Cc: pgsql-hackers(at)lists(dot)postgresql(dot)org
Subject: Re: More Perl cleanups
Date: 2025-03-15 06:26:10
Message-ID: Z9UdgmhuhOLy5i0-@paquier.xyz
Views: Raw Message | Whole Thread | Download mbox | Resend email
Thread:
Lists: pgsql-hackers

On Fri, Mar 14, 2025 at 05:39:30PM +0000, Dagfinn Ilmari Mannsåker wrote:
> In some recent-ish commits (ce1b0f9d, fd4c4ede and cc2c9fa6) we
> converted a lot of the TAP tests to use long command line options and
> fat commas (=>) to separate command line options from their arguments,
> so that perltidy does't break lines between them.
>
> However, those patches were nowhere near complete, so here's a follow-up
> to fix all the cases I could find. While I was there I failed to resist
> the urge to do some other minor tidy-ups that I think make things
> neater, but perltidy has no opinion on. I also eliminated some variables
> that were only initialised and then used once.

Thanks for digging into all that. Agreed that it would be nice to
apply a more consistent style for this release. I'll look more into
what you have here.

> The second patch might be more controversial: it eliminates unnecessary
> quoting from hash keys, both inside curly braces and before fat commas
> (except where a hash has a mix of keys that need quoting and not).

Hmm. I can partially get behind this one, seeing things like that in
what you have sent:

@@ -24,8 +24,8 @@ my $node = PostgreSQL::Test::Cluster->new('primary');
# Make sure pg_hba.conf is set up to allow connections from backupuser.
# This is only needed on Windows machines that don't use UNIX sockets.
$node->init(
- 'allows_streaming' => 1,
- 'auth_extra' => [ '--create-role' => 'backupuser' ]);
+ allows_streaming => 1,
+ auth_extra => [ '--create-role' => 'backupuser' ]);

This option handling style is inconsistent in the tree. Most of the
time these keywords are not quoted when it comes to our internal test
modules..
--
Michael

In response to

Browse pgsql-hackers by date

  From Date Subject
Next Message Michael Paquier 2025-03-15 06:29:24 Re: Allow default \watch interval in psql to be configured
Previous Message Amit Kapila 2025-03-15 06:24:40 Re: long-standing data loss bug in initial sync of logical replication