Re: C99 support

[Roger Leigh]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

"Kevin P. Fleming" <address@hidden> writes:

> Roger Leigh wrote:
>
>> and these work well.  What it doesn't do is let me use features such
>> as mixed declarations and code.  These require you to use "gcc
>> -std=c99" or "c99" or similar, and I can't enable this portably.  If
>> autoconf could find out how to put a given compiler into C99 mode,
>> that would be great (in the same way as AC_PROG_GCC_TRADITIONAL
>> works for K&R C).
>
> This sort of test is also needed to test C99 support for anonymous
> unions and structures in declarators, which I have run into problems
> with before.
>
> If there's a way to create AC_PROG_CC_C99 or something similar, that
> would be wonderful.

Here we go (based on _AC_PROG_CC_STDC).  This currently only works
with GCC; it's trivial to support others.  The code is pretty
brain-dead, so I'm sure there are more interesting ways of testing
C99 support (possibly including the current ANSI C tests as well).

# AC_PROG_CC_C99
# ----------------
# If the C compiler in not in ISO C99 C mode by default, try to add an
# option to output variable @code{CC} to make it so.  This macro tries
# various options that select ISO C99 C on some system or another.  It
# considers the compiler to be in ISO C99 C mode if it handles mixed
# code and declarations, _Bool, inline and restrict.
AC_DEFUN([AC_PROG_CC_C99],
[AC_MSG_CHECKING([for $CC option to accept ISO C99 C])
AC_CACHE_VAL(ac_cv_prog_cc_c99,
[ac_cv_prog_cc_c99=no
ac_save_CC=$CC
AC_LANG_CONFTEST([AC_LANG_PROGRAM(
[[static inline int
test(const char *restrict text)
{
  int i = 0;
  while (*(text+i) != '\0')
    ++i;
  return 0;
}]],
[[_Bool success = 1;
  if (test("String literal") != 0)
    success = 0;
  char *restrict newvar = "Another string";]])])
# Try
# GCC                   -std=c99 -std=iso9899:1999
for ac_arg in "" -std=c99 -std=iso9899:1999
do
  CC="$ac_save_CC $ac_arg"
  _AC_COMPILE_IFELSE([],
                     [ac_cv_prog_cc_c99=$ac_arg
break])
done
rm -f conftest.$ac_ext conftest.$ac_objext
CC=$ac_save_CC
])
case "x$ac_cv_prog_cc_c99" in
  x|xno)
    AC_MSG_RESULT([none needed]) ;;
  *)
    AC_MSG_RESULT([$ac_cv_prog_cc_c99])
    CC="$CC $ac_cv_prog_cc_c99" ;;
esac
])# _AC_PROG_CC_C99


And this is a little patch for c.m4:

- --- /usr/share/autoconf/autoconf/c.m4   2004-11-23 23:19:42.000000000 +0000
+++ c99.m4      2004-11-29 18:19:33.000000000 +0000
@@ -461,6 +461,9 @@
   AC_CHECK_TOOL(CC, cc)
 fi
 if test -z "$CC"; then
+  AC_CHECK_PROGS(CC, [c99 c89])
+fi
+if test -z "$CC"; then
   AC_CHECK_PROG(CC, cc, cc, , , /usr/ucb/cc)
 fi
 if test -z "$CC"; then


It would be great if in the next year, AC_PROG_CC could default to
using a C99 compiler.  Since it should be backward-compatible with C89
(??), this shouldn't be required, but I would really like to see C99
as the default, with some means of choosing an older standard if so
required.


Regards,
Roger

- -- 
Roger Leigh
                Printing on GNU/Linux?  http://gimp-print.sourceforge.net/
                Debian GNU/Linux        http://www.debian.org/
                GPG Public Key: 0x25BFB848.  Please sign and encrypt your mail.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.2.5 (GNU/Linux)
Comment: Processed by Mailcrypt 3.5.8 <http://mailcrypt.sourceforge.net/>

iD8DBQFBq2m1VcFcaSW/uEgRAs6sAKDYDAqCORidgrHCJVz6pjo5Cu6ktQCeKAh6
9rRaZXOzwgI++jw3WfAGF38=
=bStj
-----END PGP SIGNATURE-----