Re: CVS commit: pkgsrc/geography/gama

["David H. Gutteridge" <david%gutteridge.ca@localhost>]

On Tue, 11 Jul 2023 at 16:45:29 +0000, nia wrote:
> >   We should somehow make e.g. USE_LANGUAGES=c++17 lead to GCC_REQD=foo
> >   if the compiler is gcc, and have similar mechanisms architecturally
> >   for other compilers.  It is not reasonable to open-code the same
> >   information in tons of packages -- even if it is the status quo.
>
> I don't think this is a realistic solution for the simple reason
> that GCC tends to add support for new features incrementally.
>
> Examples where this does not work:
>
> https://freshbsd.org/netbsd/pkgsrc/commit/9cajoI8lhnNMBlwE
> Baseline C++11 support in GCC is not enough for this package.
>
> https://freshbsd.org/netbsd/pkgsrc/commit/3f1yUWsekhyn0bwE
> This is support for an extension which is arguably not tied
> to any particular standard.

Indeed. There are also cases where the GCC in NetBSD base doesn't have
the same feature set as the same release in pkgsrc, e.g., with the
whole webkit-gtk thing a while ago, where technically GCC 7 has
<filesystem> available in its "experimental" features, but it was
left out in NetBSD base, so saying "GCC 7 supports many C++17 features"
is even more nuanced. Thus we then hit GCC_REQD with a higher minimum
to work around that.

I don't have the energy to respond further to this thread tonight, but
would add whatever the ultimate decision, it would be nice to have
something that's either self-documenting or a standard about having a
comment that explicitly notes what is intended/expected. Having just a
USE_LANGUAGES= c++ and GCC_REQD with no comment isn't very helpful.
(That's not directed at you, Nia.)

(It might be helpful to retitle this thread, though I haven't done so.)

Dave