[Fink-users] autogen-5.10-2
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 Feedback: couldn't download. I tried a goodly number of the Sourceforge mirrors. - -- Package manager version: 0.29.99.cvs Distribution version: selfupdate-cvs Thu Aug 19 09:31:19 2010, 10.5, i386 Mac OS X version: 10.5.8 Xcode version: 3.1.4 gcc version: 4.0.1 (Apple Inc. build 5493) make version: 3.81 Feedback Courtesy of FinkCommander - -- Alexander Hansen Fink User Liaison -BEGIN PGP SIGNATURE- Version: GnuPG v1.4.10 (Darwin) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAkxtM38ACgkQB8UpO3rKjQ+REQCeKWGv4BN3X5oEWvfnNSIbgh5C mHYAnihBAKa2ZWkC4uYep8KZsZNLAtbp =rFnh -END PGP SIGNATURE- -- This SF.net email is sponsored by Make an app they can't live without Enter the BlackBerry Developer Challenge http://p.sf.net/sfu/RIM-dev2dev ___ Fink-users mailing list Fink-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-users
Re: [Fink-users] Upgrade of kdeedu4-mac-base fails
Jean-Fran?ois Mertens said: > > On 17 Aug 2010, at 17:45, Daniel E. Macks wrote: > >> Built fine for me on 10.6/i386 case-insensitive. Those libs were not >> listed in the original packaging, which caused problems when they >> later appeared during the build. So now they aren't appearing? Sounds >> like there's an unspecified dependency that causes it to only get >> built if [something] is detected. > > bash-4.0# otool -L /sw/opt/kde4/x11/lib/libavogadro-kalzium.0.dylib > /sw/opt/kde4/x11/lib/libavogadro-kalzium.0.dylib: > /sw/opt/kde4/x11/lib/libavogadro-kalzium.0.dylib (compatibility > version 0.0.0, current version 0.9.3) > /sw/lib/libopenbabel.3.dylib (compatibility version 4.0.0, current > version 4.3.0) > /sw/lib/qt4-x11/lib/libQtOpenGL.4.dylib (compatibility version 4.6.0, > current version 4.6.2) > /sw/lib/qt4-x11/lib/libQtGui.4.dylib (compatibility version 4.6.0, > current version 4.6.2) > /sw/lib/qt4-x11/lib/libQtCore.4.dylib (compatibility version 4.6.0, > current version 4.6.2) > /usr/lib/libSystem.B.dylib (compatibility version 1.0.0, current > version 111.1.5) > /usr/X11/lib/libGLU.1.dylib (compatibility version 1.3.0, current > version 1.3.0) > /usr/X11/lib/libGL.1.dylib (compatibility version 1.2.0, current > version 1.2.0) > /usr/lib/libstdc++.6.dylib (compatibility version 7.0.0, current > version 7.4.0) > /usr/lib/libgcc_s.1.dylib (compatibility version 1.0.0, current > version 1.0.0) > yielding > > bash-4.0# otool_deps kalzium4-x11 > kdeedu4-x11-base-shlibs, kdelibs4-x11-shlibs, libopenbabel3-shlibs, > qt4-x11-core-shlibs, qt4-x11-gui-shlibs > bash-4.0# fink dumpinfo -fdepends kalzium4-x11 > Information about 10836 packages read in 1 seconds. > depends: ocaml, kdeedu4-x11-base (>= 4.4.1-2) > > so libopenbabel3-shlibs is not even among the recursive deps. > > And also not among the builddeps Fix committed. dan -- Daniel Macks dma...@netspace.org -- This SF.net email is sponsored by Make an app they can't live without Enter the BlackBerry Developer Challenge http://p.sf.net/sfu/RIM-dev2dev ___ Fink-users mailing list Fink-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-users
[Fink-users] autogen-5.10-2 self-test failure (gperf problem)
But fails self-tests (10.6/i386):
* * * * LOG ENTRY 2 * * * *
cd /sw/build.build/autogen-5.10-2/autogen-5.10/build/agen5/test/testdir
gperf --version > /dev/null 2>&1 || die "no gperf program"
test -z ${gpdir} && {
gpdir=`mktemp -d ./.gperf.XX` 2>/dev/null
test -z "${gpdir}" && gpdir=.gperf.$$
test -d ${gpdir} || mkdir ${gpdir} || die "cannot mkdir ${gpdir}"
}
cd ${gpdir} || die cannot cd into ${gpdir}
gpdir=`pwd`
gperf_gperf=${gpdir}/gperf
( cat <<- '_EOF_'
%{
#include
%}
struct gperf_index { char const * name; int const idx; };
%%
_EOF_
idx=1
while read f
do echo "${f}, ${idx}"
idx=`expr ${idx} + 1`
done <<- _EOLIST_
[...]
exec 2> gperf.log
gperf --language=ANSI-C -H gperf_hash -N gperf_find --null-strings \
-C -E -I -t gperf.gperf > gperf-temp.c || \
die "gperf failed on ${gpdir}/gperf.gperf
`cat gperf.log`"
egrep -v '^_*inline$' gperf-temp.c > gperf.c
export CFLAGS=-g
${MAKE-make} gperf 1>&2
test $? -eq 0 -a -x ${gperf_gperf} || \
die "could not build gperf program
`cat gperf.log`"
exec 2>&8
echo
echo ShElL-OuTpUt-HaS-bEeN-cOmPlEtEd - 2
Killing AutoGen: gperf failed on
/sw/build.build/autogen-5.10-2/autogen-5.10/build/agen5/test/testdir/.gperf.lrb2PE/gperf.gperf
gperf: unrecognized option `--null-strings'
Usage: gperf
[-cCdDef[num]FGhHiIjkKlLnNorsStTvWZ7] [input-file]
Try `gperf --help' for more information.
AutoGen aborting on signal 15 (Terminated) in state EMITTING
processing template gperf.tpl
on line 4
for function EXPR (12)
../../../agen5/test/gperf.test: line 81: 56368 Abort trap ${AGexe}
--trace=server-shell ${testname}.def
+ failure AutoGen failed
+ local 'PS4=>fail>'
>fail>:
>fail>cd /sw/build.build/autogen-5.10-2/autogen-5.10/build/agen5/test/testdir
>fail>test -d ../FAILURES
>fail>mkdir ../FAILURES
>fail>set +x
FAILURE: AutoGen failed
FAIL: gperf.test
I have fink's gperf-2.7.2-1012 installed, which does not seem to know
the --null-strings option. Apple's 10.6 gperf ("GNU gperf 3.0."3) does
and removing fink's allows self-test to succeed. So the self-test
suite needs to be patched to use /usr/bin/gperf instead of relying on
PATH.
dan
--
Daniel Macks
dma...@netspace.org
--
This SF.net email is sponsored by
Make an app they can't live without
Enter the BlackBerry Developer Challenge
http://p.sf.net/sfu/RIM-dev2dev
___
Fink-users mailing list
Fink-users@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/fink-users
Re: [Fink-users] EMBOSS update - new dependencies?
Alexander Hansen said: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > On 8/10/10 5:37 PM, Daniel E. Macks wrote: >> Alexander Hansen said: > > Question: when (or more importantly, why) did EMBOSS suddenly require the installation of (all of) 263 dependencies, including what appears to be all of GNOME? >>> >>> The all-of-gnome thing probably came from gnuplot. You can use >>> gnuplot-nogtk to avoid that. >> >> Also the latest (as of -101) emboss package relies on external >> (separate packages) of several things that used to be internal. In >> particular, plplot is a huge library that is no longer build within >> emboss itself. The plplot packaging is itself using lots of gnome >> things that emboss doesn't need. To quote a note in the emboss >> packaging file: >> >> # plplot's -nognome varianting is a disaster (adding/removing can >> # change who if anyone supplies various headers and libs). >> >> It's fixable (tease apart the non-gnome stuff into a clean independent >> package set) but not yet fixed. The plplot that comes with emboss is >> its own packaging disaster that's *not* fixable and had led to all >> sorts of other breakage in the previous packaging of the emboss suite. >> So until I finish cleaning up plplot, "users of plplot" (i.e., you, if >> you update emboss) are stuck getting gnome stuff. >> >> If the previous emboss you had installed (I assume you had it, because >> update-all saw it was present-and-old) was working fine, there's no >> urgency to updating it. The changes are *just* to the packaging >> layout: shifting lots of things to external and solving lots of binary >> compatibility problems in its included libraries. Updates to plplot, >> which should trim out most of the gnome stuff, will probably take >> another week or so. >> > plplot builds against octave, so it brings that into the mix, too. Not any more. New slim plplot (no gnome, octave, tcltk, gcc4.x), and emboss updated to use it. Of course, I don't actually *use* any of this--all I know is that it compiled and the programs ran without crashing up until the point of requiring me to give them any data or commands. dan -- Daniel Macks dma...@netspace.org -- This SF.net email is sponsored by Make an app they can't live without Enter the BlackBerry Developer Challenge http://p.sf.net/sfu/RIM-dev2dev ___ Fink-users mailing list Fink-users@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/fink-users
