Re: Make all Perl warnings fatal - Mailing list pgsql-hackers

From Andrew Dunstan
Subject Re: Make all Perl warnings fatal
Date
Msg-id 57346790-53cb-67ac-b62f-42813110f9fc@dunslane.net
Whole thread Raw
In response to Re: Make all Perl warnings fatal  (Peter Eisentraut <peter@eisentraut.org>)
Responses Re: Make all Perl warnings fatal
Re: Make all Perl warnings fatal
List pgsql-hackers


On 2023-08-21 Mo 02:20, Peter Eisentraut wrote:
To avoid a complete bloodbath on cfbot, here is an updated patch set that includes a workaround for the getprotobyname() issue mentioned below.


On 10.08.23 07:58, Peter Eisentraut wrote:
We have a lot of Perl scripts in the tree, mostly code generation and TAP tests.  Occasionally, these scripts produce warnings.  These are AFAICT always mistakes on the developer side (true positives).  Typical examples are warnings from genbki.pl or related when you make a mess in the catalog files during development, or warnings from tests when they massage a config file that looks different on different hosts, or mistakes during merges (e.g., duplicate subroutine definitions), or just mistakes that weren't noticed, because, you know, there is a lot of output in a verbose build.

I wanted to figure put if we can catch these more reliably, in the style of -Werror.  AFAICT, there is no way to automatically turn all warnings into fatal errors.  But there is a way to do it per script, by replacing

     use warnings;

by

     use warnings FATAL => 'all';

See attached patch to try it out.

The documentation at <https://perldoc.perl.org/warnings#Fatal-Warnings> appears to sort of hand-wave against doing that.  Their argument appears to be something like, the modules you use might in the future produce additional warnings, thus breaking your scripts.  On balance, I'd take that risk, if it means I would notice the warnings in a more timely and robust way.  But that's just me at a moment in time.

Thoughts?


It's not really the same as -Werror, because many warnings can be generated at runtime rather than compile-time.

Still, I guess that might not matter too much since apart from plperl we only use perl for building / testing.

Regarding the dangers mentioned, I guess we can undo it if it proves a nuisance.

+1 to getting rid if the unnecessary call to getprotobyname().


cheers


andrew

--
Andrew Dunstan
EDB: https://www.enterprisedb.com

pgsql-hackers by date:

Previous
From: "Tristan Partin"
Date:
Subject: Re: meson: pgxs Makefile.global differences
Next
From: Jeff Davis
Date:
Subject: Re: [17] Special search_path names "!pg_temp" and "!pg_catalog"