The configure script for the local copy of libxslt that OO compiles has
a line in it that is invalide for FreeBSD's /bin/test that causes the
configure script to fail. Specifically, the script uses == for
equality and /bin/test expects =/ I'm not sure why OO is compiling the
local copy of libxslt, but this patch removes the configuration error.
This is resolved somewhat indirectly. OO patches the libxslt
configuration script, and this patch alters that patch.
How-To-Repeat: I suspect that there's an installed ports interaction that's causing
the compilation to go on, but casuing a local copy of libxslt to get
compiled by the port should fail on any FreeBSD.
Over to maintainer(s).
The supplied patch worked fine for me. The system is:
> uname -a
FreeBSD wep4017.physik.uni-wuerzburg.de 8.0-CURRENT FreeBSD 8.0-CURRENT #0: Thu Aug 28 13:17:09 CEST 2008 firstname.lastname@example.org:/usr/obj/usr/src/sys/GENERIC amd64
Thanks. This patch fixes the build process on -STABLE, too.
On Tue, Sep 09, 2008 at 04:42:04PM +0200, Martin wrote:
> Thanks. This patch fixes the build process on -STABLE, too.
Glad to help.
http://www.isi.edu/~faber PGP: http://www.isi.edu/~faber/pubkeys.asc
Unexpected attachment on this mail? See http://www.isi.edu/~faber/FAQ.html#SIG
I wonder why this bugfix does not get committed.
Perhaps the steps to reproduce the bug are missing, because some people
are telling me, they have no problems.
So this is how I build OpenOffice and get this error:
make install clean WITH_TTF_BYTECODE_ENABLED=yes WITH_CUPS=yes
Fixed by but we do not reference your patch to avoid
SCA issue. (all patch submitters shall sign to SCA
otherwise we cannot upstream)