Hi Earnie,
I'll reiterate some of Olly's and Morten's points.
On Mon, Feb 14, 2000 at 05:24:44PM -0800, Earnie Boyd wrote:
> --- Olly Betts <[EMAIL PROTECTED]> wrote:
> > * "CC" is an unwise choice for DOS/Windows (not a problem if we check
> > that the compiler groks C++ so we can ignore this)
> >
>
> What do you mean CC is an unwise choice for DOS/Windows. I use it all the time
> for DOS/Windows. Check out www.mingw.org.
Since DOS is case insensitive, "CC" could also mean "cc," which is
typically the C compiler. The current AC_PROG_CXX past would
incorrectly accept the C compiler as a working C++ compiler since the
test program used to check if the C++ compiler is working is also a
valid C program, thus allowing the C compiler "CC" in DOS to be
considered a valid C++ compiler, too. The simple C++ test I posted
for use in AC_PROG_CXX would hopefully prevent C compilers from being
considered working C++ compilers, unless they actually support C++.
> > * Prefer "aCC" to "CC" (appears to be safe everywhere)
> >
>
> Can you recapsulate why?
"aCC" and "CC" are both C++ compilers on HP-UX. "aCC" is the better
C++ compiler of the two, as Morten pointed out. It certainly seems
like a better idea to detect the better compiler before the bad one,
hence the preference of "aCC" to "CC." Since "aCC" is only available
on HP-UX, there should be no problem with placing it in the search
list before "CC."
HTH,
-Ossama
--
Ossama Othman <[EMAIL PROTECTED]>
Distributed Object Computing Laboratory, Univ. of California at Irvine
1024D/F7A394A8 - 84ED AA0B 1203 99E4 1068 70E6 5EB7 5E71 F7A3 94A8
