Re: TAP test command_fails versus command_fails_like - Mailing list pgsql-hackers

From Dagfinn Ilmari Mannsåker
Subject Re: TAP test command_fails versus command_fails_like
Date
Msg-id 87bjv72iom.fsf@wibble.ilmari.org
Whole thread Raw
In response to Re: TAP test command_fails versus command_fails_like  (Dagfinn Ilmari Mannsåker <ilmari@ilmari.org>)
List pgsql-hackers
Dagfinn Ilmari Mannsåker <ilmari@ilmari.org> writes:

> Andrew Dunstan <andrew@dunslane.net> writes:
>
>> On 2025-02-12 We 8:58 AM, Dagfinn Ilmari Mannsåker wrote:
>>>
>>>> Another question is whether command_fails and command_fails_like is
>>>> the only pair or there are more which need stricter checks?
>>> If we do this, we should do it across the board for
>>> PostgreSQL::Test::Utils and ::Cluster at least.  Once we bump the
>>> minimum perl version to 5.20 or beyond we should switch to using
>>> function signatures (https://perldoc.perl.org/perlsub#Signatures), which
>>> gives us this checking for free.
>>
>> Is there any reason we can't move to 5.20? Are there any buildfarm
>> animals using such an old version? 5.20 is now almost 10 years old.
>
> Red Hat Enterprise Linux 7 has Perl 5.16 and is on Extended Lifecycle
> Support until 2028-06-30. I don't know how long other distros based on
> that (e.g. CentOS, Scientific Linux) are supported, but I can see that
> Amazon Linux 2 is almost out of support (2025-06-30).

Ah, I found the thread from when we bumped it from 5.8 to 5.14 in 2022,
which has a list of then-current perl versions in the buildfarm:

https://postgr.es/m/20220902190329.jxvdeehkbfnrfmtl%40awork3.anarazel.de

- ilmari



pgsql-hackers by date:

Previous
From: Andres Freund
Date:
Subject: Re: Small memory fixes for pg_createsubcriber
Next
From: Paul Jungwirth
Date:
Subject: Re: SQL:2011 application time