Re: pruning stale C language variants

David Holland <[email protected]> writes:

>  > > There is one fly in this ointment, which is that we might need to feed
>  > > c2x, c1x, or even c9x to older compilers. I'd expect that any needed
>  > > support for this and c1x/c9x would already be in place; it might be
>  > > necessary to extend that to provide c2x in place of c23 for some
>  > > versions.
>  > 
>  > If so that is a pre-existing problem in the code.
>
> No... what I'm saying is that if you set the pkgsrc-level language to
> c11, and are using an old compiler, we might need to transform that to
> --std=c1x before running the compiler, and likewise for c23/c2x.

Agreed completely - that's what I meant.   However, cleaning up aliases
is unrelated to the fact that in the code before I touched it, there was
no translation like this; it was necessary before and it's still necessary.

The code:

  .if !empty(FORCE_C_STD)
  .  for std in ${_GCC_C_DIALECTS}
  .    if !empty(FORCE_C_STD:M${std})
  CWRAPPERS_APPEND.cc+=   -std=${std}
  .    endif
  .  endfor
  .endif

So whatever is in FORCE_C_STD, if it's a legit std name, is just added.
This is where we need a translation table based on compiler type and
version.  Maybe a small program in rust to query a datbase in sqlite?