Re[4]: AC_PROG_CXX does not work correctly after AC_PROG_CC

[Tim Van Holder]

> > My suggestion would probably be to
> >   a) have the EXEEXT tests fail graciously (i.e. have them
> >      give a 'warning: cannot determine executable extension
> >      using the <LANG> compiler', and not set any findings (so
> >      possible following AC_PROG_<COMPILER> calls can find it
> >      out properly).
> 
> Why? EXEEXT works like a sanity check for a compiler.  As I understand
> it, it bombs only if it cannot find an executable made by compiler.

Well, it's not really a sanity check, as it's only run for the first
compiler requested.  My point would be that on a system that has only
a C compiler, checking for CXX, then CC should behave the same as
checking for CC, then CXX.  Currently it doesn't, as the EXEEXT test
bombs out.  If it had deferred a decision, the check for CC afterwards
would have determined the correct EXEEXT.

> >   b) have the AC_PROG_<compiler> either bomb out if the
> >      requested compiler is not available, or set the respective
> >      variable to 'none', so scripts can decide what to do.
> 
> I've seen in mailing list archives that this feature (making checks
> optionals by user) was already considered couple of years ago,
> however, there are still no signs of actual implementation.

Well, in the case of AC_PROG_CXX, you can sort of detect the
no-compiler-found case by checking if CXX is g++ and GXX is not 'yes'.