Re: AC_PROG_CC not working

[Akim Demaille]

>>>>> "Morten" == Morten Eriksen <address@hidden> writes:

Morten> One argument for using the output of uname in some form
Morten> instead of the current scheme is that the Cygwin test (at
Morten> least) will fail if not using GCC as the compiler. I'm using
Morten> MSVC++ cl.exe now, and the Cygwin check gives me a false
Morten> negative because it's checking for the current compiler to
Morten> define __CYGWIN__.

Arg!

Morten> (BTW, isn't it bad style of Autoconf to need to check for
Morten> Cygwin, Mingw and EMX OS/2 in the first place? I can see the
Morten> result is used in the AC_EXEEXT check -- but this macro is
Morten> fubar anyhow and in bad need of a rewrite.)

I don't know.  To be honest, I'm starting to despair: there are way
too many things that need to be completely redesigned in Autoconf, but
*we can't wait*.  2.50 is already significantly superior to 2.13, and
we need to stop somewhere.  We need to stop for a while, knowing it's
far from being perfect, and then, for 2.51, address one single big
issue, and so on for the other releases.

IMHO, the major tasks for 2.51 are:
- abstracting autoconf, adhoc, autotest into autoscript or something
- redesigning the compiler testing suite.

Later:
- specializing macros (AC_CHECK_FUNCS(fnmatch) uses AC_FUNC_FNMATCH
- autorescan (checks that a configure.in is `complete')
- autosystem.h
- etc.