Skip site navigation (1)Skip section navigation (2)
Date:      Sat, 27 Jan 2007 22:35:24 +0100
From:      Markus Brueffer <markus@freebsd.org>
To:        Mathieu Prevot <freebsd-ports@club-internet.fr>
Cc:        ports@freebsd.org
Subject:   Re: new port: libtool: link: cannot find the library `'
Message-ID:  <200701272235.29343.markus@freebsd.org>
In-Reply-To: <20070127181525.GA26664@scienceclue.ath.cx>
References:  <20070126090310.GA45018@scienceclue.ath.cx> <200701271723.54070.markus@FreeBSD.org> <20070127181525.GA26664@scienceclue.ath.cx>

next in thread | previous in thread | raw e-mail | index | archive | help
--nextPart1289631.0SDx3WbmCJ
Content-Type: text/plain;
  charset="iso-8859-15"
Content-Transfer-Encoding: quoted-printable
Content-Disposition: inline

On Saturday 27 January 2007 19:15, Mathieu Prevot wrote:
> On Sat, Jan 27, 2007 at 05:23:48PM +0100, Markus Brueffer wrote:
> > Hi Mathieu,
> >
> > The problem is that the buildsystem that kmess ships is hosed. Try the
> > following in the port's directory (regenerating the buildsystem):
> >
> > # make patch
> > # cd work/kmess-1.5pre1
> > # gmake -f admin/Makefile.common
> > # cd ../..
> > # make
> >
> > Now the build should succeed. I don't have much time at the moment, but
> > will come back to you tonight with details on how to fix this properly.
>
> It made it, indeed. Thanks for this.
> I will wait for details.

Here is a version of the port that should build and install flawlessly (onl=
y=20
compile tested, no runtime tests as I don't use MSN):

http://people.freebsd.org/~markus/stuff/kmess.tar.bz2

What I did was the following: As soon as you notice funny automake messages=
=20
after the configure stage, you know something is wrong with the buildsystem=
=2E=20
In most cases, by my own experience, these are only noop changes the author=
=20
did after generating the buildsystem that changed one or more timestamps.=20
Using gmake -d in the builddirectory you can see which files are affected a=
nd=20
correct the timestamps manually (look at the post-patch target). That's wha=
t=20
I did in this case.

In some rare cases, the author really forgot to regenerate the buildsystem=
=20
before releasing the distfile. If this is the case, you have to regenerate =
it=20
by yourself (as shown in the first mail). After that you can choose between=
=20
creating a mega patch for the port, hosted at an external site, or releasin=
g=20
a completly new tarball. The latter one is mostly more appropriate, as the=
=20
size of the patch often exceeds the size of a new tarball.

Please let me know if kmess works as expected. If so, I'd like to commit it=
=20
(judging from the information from the website, this version should be quit=
e=20
stable).

Btw: This port is currently unmaintained. Would you like to be the new=20
maintainer of it? (hint, hint) :)

Best regards,

Markus

=2D-=20
Markus Brueffer =A0 =A0| GPG-Key: http://people.FreeBSD.org/~markus/markus.=
asc
markus@brueffer.de | FP: 3F9B EBE8 F290 E5CC 1447 8760 D48D 1072 78F8 A8D4
markus@FreeBSD.org | FreeBSD: The Power to Serve!

--nextPart1289631.0SDx3WbmCJ
Content-Type: application/pgp-signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (FreeBSD)

iD8DBQBFu8Wh1I0Qcnj4qNQRAuVIAKCRDqeSdeApVXWioOGOrY7GRBLFTgCgglYq
6RYXYs6fMBzKMn3CtV0bleo=
=uu3W
-----END PGP SIGNATURE-----

--nextPart1289631.0SDx3WbmCJ--



Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200701272235.29343.markus>