Re: Why do we test a build using g++?

On Mon, 29 Jun 2026 22:41:14 +0100
"Paul \"LeoNerd\" Evans" <[email protected]> wrote:

> I'd like to propose that (if there isn't another valid reason that is
> stronger than our desire to use C99 features) we find another way to
> test if the perl headers are compatible with C++. For example, could
> we ship a tiny "helloworld.cpp" program that also pulls in <perl.h>
> and <XSUB.h> or somesuch? Or maybe we write a tiny XS extension in
> C++ and compiled with g++, to use for such testing purposes?
> 
> Or failing all of those; would we be able to test with g++ in some
> sort of mode where we can ask it to accept things that C99 considers
> valid. I don't know if this one is possible - see above under "I'm
> not a C++ programmer", but I can't imagine we're the first and only
> people to ever run into this issue. Perhaps there's some way we can
> get around it?

These various chats haven't surfaced a good reason for keeping the g++
CI workflow around.

Tony points out some things we can try doing:

  > One option, rather than updating ccflags (config.sh, Config.pm,
  > inherited by CPAN builds), might be to update cflags.SH to add
  > -Wno-missing-field-initializers to the output of cflags, which is
  > only used to build the core object files.

  https://github.com/Perl/perl5/issues/24553#issuecomment-4953576892

so I'll have a go at that, but if that doesn't look too easy or doesn't
solve the problem, honestly I'm minded just to drop the `g++` target
entirely at this point.

Shout now if anyone has a different conclusion.

-- 
Paul "LeoNerd" Evans

[email protected]
http://www.leonerd.org.uk/  |  https://metacpan.org/author/PEVANS