PATCH Re: libc `make' wants to do a CVS commit ???!!!

[Martin Buchholz]

>>>>> "Uli" == Ulrich Drepper <address@hidden> writes:

Uli> Martin Buchholz <address@hidden> writes:
>> Perhaps this is a little too optimized for the maintainers.  Isn't
>> there an --enable-maintainer-mode flag or something for this kind of purpose?

Uli> How about `configure --help' which tells you about

Uli>    --without-cvs   if CVS should not be used

I think it's too surprising for the generated files to be
automatically checked in.  In particular, the generated files won't
match the rest of the files in the repository.  I'm surprised you've
never had these sorts of conflicts.  I think --with-cvs is a bad idea,
and at the very least should be off by default.  The documentation for
--without-cvs is not very clear, either - what does `used' mean?

ObPedantry:

The output from ./configure --help is not lined up correctly.

  --with-fp               if using floating-point hardware [default=yes]
  --with-binutils=PATH    specify location of binutils (as and ld)
  --with-elf              if using the ELF object format
  --without-cvs   if CVS should not be used
  --with-headers=PATH     location of system headers to use
                          [e.g. /usr/src/linux/include]
                          [default=compiler default]

Index: configure.in
===================================================================
RCS file: /cvs/libc/configure.in,v
retrieving revision 1.289
diff -u -r1.289 configure.in
--- configure.in        2000/09/05 18:54:34     1.289
+++ configure.in        2000/11/01 09:33:38
@@ -66,7 +66,7 @@
   --with-elf             if using the ELF object format,
            elf=$withval, elf=no)
 AC_ARG_WITH(cvs, dnl
-[  --without-cvs         if CVS should not be used],
+[  --without-cvs                 if CVS should not be used],
            with_cvs=$withval, with_cvs=yes)
 AC_SUBST(with_cvs)