December 2012 - macports-tickets - lists.macosforge.org

[MacPorts] #32664: php5-unit @3.5.15 (php, devel), Revision 2 + installation does not
by MacPorts 14 Feb '13

14 Feb '13

#32664: php5-unit @3.5.15 (php, devel), Revision 2 + installation does not ------------------------------------+--------------------------------------- Reporter: polkehn@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.0.3 Keywords: | Port: php5-unit ------------------------------------+--------------------------------------- Hi * When trying the shell command "phpunit" I get an error (see below). * tried to uninstall and reinstall the port. Same result :( * Mac OS X 10.7.2 * Macports 2.0.3 {{{ PHP Warning: require_once(PHP/CodeCoverage/Filter.php): failed to open stream: No such file or directory in /opt/local/bin/phpunit on line 37 PHP Stack trace: PHP 1. {main}() /opt/local/bin/phpunit:0 Warning: require_once(PHP/CodeCoverage/Filter.php): failed to open stream: No such file or directory in /opt/local/bin/phpunit on line 37 Call Stack: 0.0091 629144 1. {main}() /opt/local/bin/phpunit:0 PHP Fatal error: require_once(): Failed opening required 'PHP/CodeCoverage/Filter.php' (include_path='.::/opt/local/lib/php/pear') in /opt/local/bin/phpunit on line 37 PHP Stack trace: PHP 1. {main}() /opt/local/bin/phpunit:0 Fatal error: require_once(): Failed opening required 'PHP/CodeCoverage/Filter.php' (include_path='.::/opt/local/lib/php/pear') in /opt/local/bin/phpunit on line 37 Call Stack: 0.0091 629144 1. {main}() /opt/local/bin/phpunit:0 }}} Thanks a lot for your support! -- Ticket URL: <https://trac.macports.org/ticket/32664> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 3

[MacPorts] #30835: Scribus 1.4.0rc3 builds forever
by MacPorts 13 Feb '13

13 Feb '13

#30835: Scribus 1.4.0rc3 builds forever -----------------------------------------+---------------------------------- Reporter: greisberger@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.0.99 Keywords: | Port: scribus -----------------------------------------+---------------------------------- I'm trying to install scribus and it's stalled in the build phase. It's still compiling the same file since yesterday: $ ps auwx | grep cc1 macports 98903 97,4 2,7 2618160 171044 s002 R+ 6:37pm 611:55.18 /Developer/usr/llvm-gcc-4.2/libexec/gcc/i686-apple-darwin10/4.2.1/cc1plus -quiet -I/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_print_scribus/scribus/work/scribus-1.4.0.rc3/scribus -F/opt/local/lib -I/opt/local/lib/QtGui.framework/Headers -I/opt/local/lib/QtXml.framework/Headers -I/opt/local/lib/QtSql.framework/Headers -I/opt/local/lib/QtNetwork.framework/Headers -I/opt/local/lib/QtCore.framework/Headers -I/opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_print_scribus/scribus/work/scribus-1.4.0.rc3 -I/opt/local/lib/QtDesigner.framework/Headers -I/opt/local/lib/QtDeclarative.framework/Headers -I/opt/local/lib/QtScriptTools.framework/Headers -I/opt/local/lib/QtDBus.framework/Headers -I/opt/local/lib/QtOpenGL.framework/Headers -I/opt/local/lib/QtMultimedia.framework/Headers -I/opt/local/lib/QtXmlPatterns.framework/Headers -I/opt/local/lib/QtWebKit.framework/Headers -I/opt/local/lib/QtHelp.framework/Headers -I/opt/local/lib/QtTest.framework/Headers -I/opt/local/lib/QtScript.framework/Headers -I/opt/local/lib/QtSvg.framework/Headers -I/opt/local/lib/Qt3Support.framework/Headers -I/opt/local/share/qt4/mkspecs/default -I/opt/local/include -I/opt/local/include/libxml2 -I/opt/local/Library/Frameworks/Python.framework/Versions/2.7/include/python2.7 -I/opt/local/include/freetype2 -I/opt/local/include/cairo -imultilib x86_64 -D__DYNAMIC__ -DNDEBUG -DHAVE_CARBON=1 -DQT_GUI_LIB -DQT_XML_LIB -DQT_SQL_LIB -DQT_NETWORK_LIB -DQT_CORE_LIB -DQT_CORE_LIB -DQT_GUI_LIB -DQT_THREAD_LIB -DQT_XML_LIB -DHAVE_DLFCN_H -DHAVE_UNISTD_H -DHAVE_CONFIG_H -DQT_NO_DEBUG -DSHAREDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/scribus/" -DDOCDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/doc/scribus/" -DICONDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/scribus/icons/" -DSAMPLESDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/scribus/samples/" -DSCRIPTSDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/scribus/scripts/" -DTEMPLATEDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/scribus/templates/" -DDESKTOPDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/share/applications/" -DLIBDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/lib/scribus/" -DPLUGINDIR="/Applications/MacPorts/Scribus1.4.0.rc3.app/Contents/lib/scribus/plugins/" -isysroot / /opt/local/var/macports/build/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_print_scribus/scribus/work/scribus-1.4.0.rc3/scribus/util.cpp -fPIC -quiet -dumpbase util.cpp -mmacosx-version-min=10.6 -m64 -mtune=core2 -auxbase-strip CMakeFiles/Scribus.dir/util.cpp.o -O2 -O2 -Wall -fPIC -D__private_extern__=extern -o - cc1plus uses 97% CPU but that's all. I only noticed that the output of cc1plus is - not an object file, that's all. I really have no idea what happens. OS: Snow Leopard x86_64 XCode: 4.0.2 FS: case sensitive -- Ticket URL: <https://trac.macports.org/ticket/30835> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 12

[MacPorts] #31485: Asterisk fails to build on Lion
by MacPorts 13 Feb '13

13 Feb '13

#31485: Asterisk fails to build on Lion --------------------------------+------------------------------------------- Reporter: andrew@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.0.3 Keywords: | Port: asterisk --------------------------------+------------------------------------------- Asterisk fails to build on Lion with Xcode 4.2 DP7. Build error is :info:build error: invalid value '6' in '-O6' -- Ticket URL: <https://trac.macports.org/ticket/31485> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 8

[MacPorts] #25942: aacgain 1.9 update
by MacPorts 13 Feb '13

13 Feb '13

#25942: aacgain 1.9 update -------------------------------------+-------------------------------------- Reporter: ryandesign@… | Owner: tristan@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 1.9.1 Keywords: | Port: aacgain -------------------------------------+-------------------------------------- aacgain 1.9 is out so the port should probably be updated. {{{ Begin forwarded message: From: "David Lasker" <dave(a)altosdesign.com> Date: August 3, 2010 16:28:58 CDT To: "'Samuli Suominen'" <ssuominen(a)gentoo.org>, <marillat(a)free.fr>, <toni(a)links2linux.de>, "'Miguel Angel Alvarez'" <maacruz(a)gmail.com>, "'Prakash Punnoor'" <prakashp(a)arcor.de>, <binguri(a)yahoo.it>, "'Wolf Wilke'" <wwilke(a)genion.de>, "'Ben Allison'" <benski(a)winamp.com>, <hardloaf(a)gmail.com>, <mp3gain(a)hotmail.com>, <macports- mgr(a)lists.macosforge.org> Subject: [macports-mgr] aacgain 1.9 released I have completed the next release of aacgain. Updated files are in the mp3gain CVS repository on Sourceforge. This release incorporates the following changes: 1.Fixed a bug in mp4v2 which caused iTunes music video files to fail with "not a valid mp4/m4a file" error. 2.Gives a more meaningful error message when attempting to process Apple Lossless files (which are still not supported by aacgain). 3.Built with updated mp4v2 and faad2 code, which should build error-free on modern Windows, Linux, and Mac systems. 4.Built with latest mp3gain version 1.5.2 sources checked into CVS, which include id3v2 tag support. 5.MSVC++ solution and project files have been converted to Visual C++ 2010 Express Edition. 6.I now have access to both Linux and Mac systems, so I have modified Prakash Punoor's Unix build scripts to work correctly with the latest code changes. I have successfully built and tested aacgain on Ubuntu 10.04 and Mac OS X 10.6. In my previous email announcing the beta test of this release, I said: A major change was made in the way the global_gain adjustments are made. In previous releases, the global_gain fields in the track were edited in-place. This was done by subclassing some of the mp4v2 C++ implementation files. In this release, I create a new audio ('soun') track in the mp4 file, copy the samples from the old track to the new track with the global_gain fields modified, then delete the original track. This means that files with multiple tracks (i.e. video files) might have the order of the tracks changed. After much testing, I have removed that code and have gone back to subclassing. I found many changes were being made to the track atoms (e.g. removal of the mp4a.pinf atom tree). Although the resulting files played correctly on everything I tested them on, I didn't feel comfortable with that approach. So I have gone back to subclassing, which edits the existing track data in-place. See the aacgain home page http://altosdesign.com/aacgain for more info. Dave _______________________________________________ macports-mgr mailing list macports-mgr(a)lists.macosforge.org http://lists.macosforge.org/mailman/listinfo.cgi/macports-mgr }}} -- Ticket URL: <http://trac.macports.org/ticket/25942> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 6

[MacPorts] #36116: libstdcxx build error
by MacPorts 12 Feb '13

12 Feb '13

#36116: libstdcxx build error -------------------------------+------------------------ Reporter: macsforever2000@… | Owner: jeremyhu@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.2 Keywords: | Port: libstdcxx -------------------------------+------------------------ Mac OS X 10.8.1[BR] Xcode 4.4.1 libstdcxx is not building for me. main.log attached. -- Ticket URL: <https://trac.macports.org/ticket/36116> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 17

[MacPorts] #37420: fizmo: Could not find libncursesw
by MacPorts 12 Feb '13

12 Feb '13

#37420: fizmo: Could not find libncursesw --------------------------+------------------------------- Reporter: ryandesign@… | Owner: Christoph.Ender@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.2 Keywords: | Port: fizmo --------------------------+------------------------------- fizmo doesn't build on the Snow Leopard buildbot. Here's the log from 0.7.6 but it's been happening for several versions: https://build.macports.org/builders/buildports-snowleopard- x86_64/builds/13274/steps/compile/logs/stdio During the configure phase it says: {{{ checking for ncursesw... no checking for /usr/include/ncurses.h... yes checking for libncuresw in /usr/lib... no checking for libncuresw in /usr/local/lib... no checking for libncuresw in /opt/local/lib... no Could not find libncursesw. Try setting the location using --with-ncursesw-libdir. }}} For one thing, there is a typo in the message ("libncuresw" should be "libncursesw"). But even ignoring that, this is peculiar since libncursesw should be there; there should be no difference in ncursesw in MacPorts between versions of OS X, and it builds fine on the Lion and Mountain Lion buildbots, which simply say: {{{ checking for ncursesw... yes }}} It also builds fine for me on my Snow Leopard, Lion and Mountain Lion test machines. The suggestion the configure script gives—to set --with-ncursesw-libdir (and presumably --with-ncursesw-includedir)—shouldn't be necessary because my understanding is that "checking for ncursesw" means "checking for ncursesw ''using pkgconfig''" and pkgconfig should be giving the correct answer. But for some reason on the Snow Leopard buildbot it's not. It makes me think there's a problem with the pre-built pkgconfig or ncurses binary for Snow Leopard. (I build from source on my test machine instead of using the binaries.) I wish we had access to the work directory on the buildbot so that we could read the config.log; unfortunately we don't. Oddly neither of those ncursesw configure options are listed in `./configure --help`; if they are valid options they should be listed there. -------- Another problem is that the fizmo configure script still exits with the code 0 even though this problem occurred, so MacPorts tries to proceed to the build phase, which fails with: {{{ make: *** No rule to make target `all'. Stop. }}} -------- I'm also not sure why the code in config-libs.m4 that checks for ncursesw is trying to include jpeglib.h; that seems like a copy-pasto. {{{ for dir in $with_ncursesw_libdir /usr/lib /usr/local/lib /opt/local/lib ; do AC_MSG_CHECKING(for libncuresw in $dir) LDFLAGS="-L$dir" AC_TRY_LINK( [#include <stdio.h> #include <ncurses.h> #include "$ncursesw_h_dir/jpeglib.h"], [initscr();], [AC_MSG_RESULT(yes) ncursesw_l_dir=$dir break], [AC_MSG_RESULT(no)]) done }}} -- Ticket URL: <https://trac.macports.org/ticket/37420> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 10

[MacPorts] #22870: muniversal group: merge .la dependency_libs value
by MacPorts 12 Feb '13

12 Feb '13

#22870: muniversal group: merge .la dependency_libs value ---------------------------------+------------------------------------------ Reporter: pguyot@… | Owner: macports-tickets@… Type: enhancement | Status: new Priority: Normal | Milestone: Component: base | Version: 1.8.1 Keywords: | Port: ---------------------------------+------------------------------------------ .la files are automatically generated text files that contain linking information. muniversal group fails if two .la files differ. However, it seems we could apply some heuristics, for example with the dependency_libs key. i386 and x86_64 libgettextlib.la files are identical except for the following line: {{{ # Libraries that this one depends upon. dependency_libs=' /opt/local/lib/libintl.la -L/opt/local/lib -lc -ltermcap' }}} {{{ # Libraries that this one depends upon. dependency_libs=' -R/opt/local/lib /opt/local/lib/libintl.la -L/opt/local/lib -lc /opt/local/lib/libiconv.la -lncurses' }}} That these lines could probably be merged by putting every item in each, and that wouldn't create a problem. The attached patch does just this and allows gettext +universal to be built on 10.6. The patch could be extended to process inherited_linker_flags as well (see ticket #22801). What do you think? -- Ticket URL: <http://trac.macports.org/ticket/22870> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 4

[MacPorts] #26183: cssh 3.28 request to update port
by MacPorts 11 Feb '13

11 Feb '13

#26183: cssh 3.28 request to update port --------------------------------------+------------------------------------- Reporter: nonstop.server@… | Owner: macports-tickets@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 1.9.1 Keywords: sysutils | Port: cssh --------------------------------------+------------------------------------- Request to update cssh to the latest version.[[BR]] * [http://sourceforge.net/projects/clusterssh/files/ Cluster SSH Download Page] * [http://sourceforge.net/scm/?type=git&group_id=89139 Information on Git Repository for Cluster SSH] * [http://github.com/duncs/clusterssh Git Hub Mirror for Cluster SSH] -- Ticket URL: <http://trac.macports.org/ticket/26183> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 6

[MacPorts] #18376: Remedy socket 35 errors on all Python builds
by MacPorts 11 Feb '13

11 Feb '13

#18376: Remedy socket 35 errors on all Python builds -----------------------------+---------------------------------------------- Reporter: ben@… | Owner: macports-tickets@… Type: defect | Status: new Priority: High | Milestone: Port Bugs Component: ports | Version: 1.7.0 Keywords: python, socket | Port: -----------------------------+---------------------------------------------- OSX's poll is broken. Apple knows this, and removes the HAVE_POLL option as can be seen in the first line of the pyconfig patch here: http://www.opensource.apple.com/darwinsource/10.5.6/python-30.1.2/fix/pycon… This patch should be applied to all Python builds that MacPorts supports, as it fixes a nasty bug that is being mis-reported all over the place on OSX. Here's one such example of a traceback caused by it: {{{ File "../uploader.py", line 94, in <module> resp, data = http.request(post_uri, 'POST', body=content, headers=headers) File "build/bdist.macosx-10.5-i386/egg/httplib2/__init__.py", line 1050, in request File "build/bdist.macosx-10.5-i386/egg/httplib2/__init__.py", line 854, in _request File "build/bdist.macosx-10.5-i386/egg/httplib2/__init__.py", line 823, in _conn_request File "/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/httplib.py", line 874, in request self._send_request(method, url, body, headers) File "/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/httplib.py", line 914, in _send_request self.send(body) File "/opt/local/Library/Frameworks/Python.framework/Versions/2.6/lib/python2.6/httplib.py", line 719, in send self.sock.sendall(str) File "<string>", line 1, in sendall socket.error: [Errno 35] Resource temporarily unavailable }}} And another report caused by the same problem: http://www.cherrypy.org/ticket/598 Applying the same regexp that Apple does, fixes the problem in my tests. Credit for finding this bug and its solution should be attributed to Phil Jenvey, I merely bugged him until he found the problem for me. :) -- Ticket URL: <http://trac.macports.org/ticket/18376> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 7

[MacPorts] #35975: qwt-60 fails to build against qt4-mac +framework
by MacPorts 11 Feb '13

11 Feb '13

#35975: qwt-60 fails to build against qt4-mac +framework -------------------------------+-------------------------------------------- Reporter: mathew@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.2 Keywords: | Port: qt4-mac -------------------------------+-------------------------------------------- qwt-60 build fails with an error about not finding -lqwt. If I change -lqwt to -framework qwt in the Makefile(s) after the build fails, then next attempt completes successfully. This may possibly be related to #34864 and/or #35506, specifically the changes to qt4-mac referenced here: https://trac.macports.org/ticket/35506#comment:19 It seems that the macro qtAddLibrary referenced in designer.pro and textengines.pro is not properly detecting the qt framework build, telling the Makefile to look for a shared library instead of a framework. The macro qtAddLibrary is defined in share/qt4/mkspecs/features/qt_functions.prf as part of the qt4-mac install, which is why I filed the bug against qt4-mac instead of qwt-60. -- Ticket URL: <https://trac.macports.org/ticket/35975> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 4