Re: IPC::Run::time[r|out] vs our TAP tests

Daniel Gustafsson <daniel@yesql.se>

From: Daniel Gustafsson <daniel@yesql.se>
To: Andrew Dunstan <andrew@dunslane.net>
Cc: Heikki Linnakangas <hlinnaka@iki.fi>, PostgreSQL Hackers <pgsql-hackers@lists.postgresql.org>, Tom Lane <tgl@sss.pgh.pa.us>
Date: 2025-02-20T14:09:58Z
Lists: pgsql-hackers

Attachments

> On 20 Feb 2025, at 14:06, Andrew Dunstan <andrew@dunslane.net> wrote:

> Actually, since ok() and friends return true iff the test succeeds, instead of
> +    ok(! $self->{timeout}->is_expired, 'psql query_until did not time out');
> +    return undef if $self->{timeout}->is_expired;
> you can avoid doing the same test twice and say:
>      ok(! $self->{timeout}->is_expired, 'psql query_until did not time out') || return undef;
> although for slightly technical reasons perlcritic disapproves of "return undef" and prefers that you just write a bare "return" so we should also fix that.
> Sorry for taking a second bite at the cherry.

Not at all, I agree that this is an improvement so fixed in the attached along
with a fresh pgperltidy.

--
Daniel Gustafsson

Commits

  1. Restart BackgroundPsql's timer more nicely.

  2. Implement pg_wal_replay_wait() stored procedure

  3. psql: fix variable existence tab completion