Re: Technical Review

[Gary V . Vaughan]

On Friday 20 April 2001  7:12 pm, Akim Demaille wrote:
> >>>>> "Gary" == Gary V Vaughan <address@hidden> writes:
>
> Gary> Salut Akim!
>
> Coucou !

Qu'est-ce que c'est?

> The test suite includes this:
> > ## -------------------- ##
> > ## Autoconf & Libtool.  ##
> > ## -------------------- ##
> >
> > AT_SETUP([[Libtool]])
> >
> > # Skip this test if there is no libtoolize.
> > AT_CHECK([libtoolize --version || exit 77],
> >          ignore, ignore, ignore)
> > # CVS Autoconf probably does not work with too old Libtools.
> > AT_CHECK([[libtoolize --version | egrep '1\.3(\.[0-4])?$' && exit 77]],
> >          ignore, ignore, ignore)

There is a libtool-1.3.5, and Alexandre has threatened a 1.3.6 a couple of 
times... or are they know to work with autoconf-2.50?

> > # FIXME: Once Libtool `configure.ac' aware, change this.
> > # Using a configure.in, have libtoolize confess where libtool.m4 is.

libtool-1.3d and up are configure.ac aware -- at least I applied a patch from 
Edward Lee which I didn't test...

> > AT_DATA([configure.in],
> > [[AC_INIT
> > # The other tests are relying on Autoconf's own copy of install-sh
> > # etc. via AC_CONFIG_AUX_DIR(..).  Don't do that here, since we don't
> > # want libtoolize to pollute Autoconf's srcdir.
> > AC_CONFIG_AUX_DIR(.)
> > AC_PROG_LIBTOOL
> > ]])
> > AT_CHECK([libtoolize | sed -n ["s,^.*\`\(/[^']*\)'.*,\1,p"]],
> >          0, [stdout], ignore)
> >
> > # Make sure at-path contains something valid, and let the test suite
> > # display it when verbose.  And fail, skipping would too easily hide
> > # problems.
> > AT_CHECK([grep . stdout], 0, [ignore])
> > AT_CHECK([test -f "`cat stdout`"])
> >
> > # libtoolize installed everything but install-sh...
> > touch install-sh
> >
> > # Build the concatenation of libtool.m4 and configure.ac.
> > cp `cat stdout` configure.in
> > cat >>configure.in <<_EOF
> > AC_INIT
> > AC_CONFIG_AUX_DIR(.)
> > AC_CANONICAL_SYSTEM
> > AC_PROG_LIBTOOL
> > _EOF
> >
> > AT_CHECK_AUTOCONF
> >
> > # Ignore stderr, because ltconfig always loads the cache, which is
> > # /dev/null, and some shells choke on this.  For instance with Bash
> > # 2.05, you get:
> > #
> > #   loading cache /dev/null within ltconfig
> > #   ./ltconfig: .: /dev/null: not a regular file
> > #
> > # But the script executes properly.
> > AT_CHECK_CONFIGURE([], [], [], [ignore])

As an aside, this is only true for MLB snapshots of libtool.  1.4 has no 
ltconfig, and no need to slurp config.cache -- I will be making the same 
changes to MLB as it is merged back into the trunk after 1.4.

> > AT_CLEANUP(install-sh ltconfig libtool at-path ltmain.sh config.guess
> > config.sub)
>
> whichi basically just makes sure AC_PROG_LIBTOOL runs smoothly.  If
> you think some stronger tests are needed, please tell me!

Au contraire, all my suggestions are for weaker tests =)O|

Cheers,
        Gary
-- 
  ___              _   ___   __              _         mailto: address@hidden
 / __|__ _ _ ___ _| | / / | / /_ _ _  _ __ _| |_  __ _ ___       address@hidden
| (_ / _` | '_|// / |/ /| |/ / _` | || / _` | ' \/ _` | _ \
 \___\__,_|_|\_, /|___(_)___/\__,_|\_,_\__, |_||_\__,_|//_/
home page:  /___/                      /___/                  gpg public key:
http://www.oranda.demon.co.uk           http://www.oranda.demon.co.uk/key.asc