Re: Using AC_C_INLINE blows up C++ tests

[Paul Eggert]

Albert Chin <address@hidden> writes:

> The correct solution is checking for 'inline' for whatever the current
> language is. Of course, this means programmers cannot use 'inline' for
> C and C++ which is counterintuitive so the patch might be good enough.

OK, you guys talked me into it.  I installed the following patch.
Using AH_VERBATIM is a bit of a hack, but I couldn't think of a
cleaner hack.

2003-07-07  Paul Eggert  <address@hidden>

        * lib/autoconf/c.m4 (AC_C_INLINE): Wrap the '#define inline ...'
        inside '#ifndef __cplusplus'.  Problem reported by
        Bob Friesenhahn.

Index: c.m4
===================================================================
RCS file: /cvsroot/autoconf/autoconf/lib/autoconf/c.m4,v
retrieving revision 1.180
retrieving revision 1.181
diff -p -u -r1.180 -r1.181
--- c.m4        21 Jun 2003 00:01:04 -0000      1.180
+++ c.m4        7 Jul 2003 07:41:07 -0000       1.181
@@ -1020,12 +1020,25 @@ $ac_kw foo_t foo () {return 0; }
                    [ac_cv_c_inline=$ac_kw; break])
 done
 ])
+AH_VERBATIM([inline],
+[/* Define to `__inline__' or `__inline' if that's what the C compiler
+   calls it, or to nothing if 'inline' is not supported under any name.  */
+#ifndef __cplusplus
+#undef inline
+#endif])
 case $ac_cv_c_inline in
   inline | yes) ;;
-  no) AC_DEFINE(inline,,
-               [Define as `__inline' if that's what the C compiler calls it,
-                or to nothing if it is not supported.]) ;;
-  *)  AC_DEFINE_UNQUOTED(inline, $ac_cv_c_inline) ;;
+  *)
+    case $ac_cv_c_inline in
+      no) ac_val=;;
+      *) ac_val=$ac_cv_c_inline;;
+    esac
+    cat >>confdefs.h <<_ACEOF
+#ifndef __cplusplus
+#define inline $ac_val
+#endif
+_ACEOF
+    ;;
 esac
 ])# AC_C_INLINE