Re: [patch] config.status exit status on Ultra 10 w/Solaris 8

2001-02-21 Thread Akim Demaille
"Lars J. Aas" <[EMAIL PROTECTED]> writes: > This patch fixes the exit code of config.status for Ultra 10s running > Solaris 8. > > BTW, I noticed many un-wrapped "exit 0"s in acgeneral.m4, but I only did > something with this one. Thanks Lars! Sorry I still didn't look at your other patches (s

DOS Support

2001-02-21 Thread John Poltorak
I tried the latest autoconf snapshot yesterday with the hope that it included support for DOS style pathnames. Was I wrong to expect this? I found that configure would not find INSTALL or PERL even though they are on the path, although setting up $INSTALL and $PERL worked OK. Configure eventual

autoconf-2.13: AC_C_{CONST,INLINE} broken under g++ (gcc-2.95.2)?

2001-02-21 Thread Adam J. Richter
Under autoconf-2.13 and gcc-2.95.2, AC_C_CONST and AC_C_INLINE will incorrectly return a negative result if called after AC_PROG_CXX. The reason for this is that AC_PROG_CXX turns the compiler by AC_TRY_COMPILE into the C++ compiler, which has stricter standards than the C compile

Re: autoconf-2.13: AC_C_{CONST,INLINE} broken under g++ (gcc-2.95.2)?

2001-02-21 Thread Akim Demaille
Thanks, fixed long ago in CVS Autoconf.

Re: DOS Support

2001-02-21 Thread Lars J. Aas
On Wed, Feb 21, 2001 at 10:11:12AM +, John Poltorak wrote: : : I tried the latest autoconf snapshot yesterday with the hope that it : included support for DOS style pathnames. Was I wrong to expect this? : : I found that configure would not find INSTALL or PERL even though they : are on the

RE: Latest snapshot of AUTOCONF

2001-02-21 Thread Tim Van Holder
> Also, should it search the PATH when:- > > checking for a BSD compatible install... ? > Install.exe is on the path, but configure selects ./install-sh -c > instead. Yes it should - but the current CVS autoconf does not do this for AC_PROG_INSTALL. And even if it does, it always seems to selec

2.13 issue with AC_PROG_CXX

2001-02-21 Thread Gray Watson
Thanks much for autoconf. Continues to be extremely useful. I'm a little surprised that the test for C++ compiler support is fatal if no. The documentation does not imply this and I wish it did not have that behavior. I have a package which tests for C++ support and if not there it will no bui