Re: 84-gary-diagnose-version-mismatch.patch

[Scott James Remnant]

On Fri, 2004-02-20 at 19:08, Gary V. Vaughan wrote:

Something I didn't catch last time...

>       Sweeping changes to the user interface to libtool from
>       `configure.ac' to be more like AC_INIT and accept a space
>       delimited list of options.  Instead of calling `AC_LIBTOOL_DLOPEN;
>       AC_PROG_LIBTOOL', we now recommend `LT_INIT([dlopen])':
> 
>       * m4/libtool.m4 (AC_PROG_LIBTOOL, _AC_PROG_LIBTOOL)
>       (AC_LIBTOOL_SETUP): Removed.  Added AU_DEFUNs.
>       (LT_INIT): Replace with an Autoconf like interface which accepts a
>       version number as a minimum required libtool release at configure
>       time.
*snip*
>       * m4/lt~obsolete.m4 (AC_LIBTOOL_SETUP, _AC_PROG_LIBTOOL)
>       (_LT_PROG_LTMAIN):  More AC_DEFUNs that have been retracted.
> 
There's no need to add _AC_PROG_LIBTOOL to lt~obsolete.m4.

We only need to add to this file where a macro that was AC_DEFUN has
been moved to m4_define (or some other equivalent) and is *still*
referenced in the m4 files.

descent m4% grep _AC_PROG_LIBTOOL *.m4
lt~obsolete.m4:m4_ifdef([_AC_PROG_LIBTOOL],,            
[AU_DEFUN([_AC_PROG_LIBTOOL])])
descent m4%

There aren't any references left, so aclocal won't stupidly pull in an
old libtool.m4, which is what lt~obsolete.m4 is designed to prevent.

All the other macros in there are still referenced, so need to be there.

Scott
-- 
Have you ever, ever felt like this?
Had strange things happen?  Are you going round the twist?

signature.asc
Description: This is a digitally signed message part