August 2015 - macports-tickets - lists.macosforge.org

[MacPorts] #46483: installation stopped accidentally
by MacPorts 30 Sep '15

30 Sep '15

#46483: installation stopped accidentally -------------------------------+-------------------------------- Reporter: pritish.naik.04@… | Owner: macports-tickets@… Type: defect | Status: new Priority: High | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: -------------------------------+-------------------------------- pritishs-MacBook-Pro-3:~ pritish04$ sudo port install netcdf-fortran ---> Computing dependencies for netcdf-fortran ---> Dependencies to be installed: libgcc cctools llvm-3.5 cloog gmp isl gcc_select ld64 libmpc mpfr ---> Extracting llvm-3.5 Error: org.macports.extract for port llvm-3.5 returned: Failed to locate 'xz' in path: '/opt/local/bin:/opt/local/sbin:/bin:/sbin:/usr/bin:/usr/sbin' or at its MacPorts configuration time location, did you move it? Error: Failed to install llvm-3.5 Please see the log file for port llvm-3.5 for details: /opt/local/var/macports/logs/_opt_local_var_macports_sources_rsync.macports.org_release_tarballs_ports_lang_llvm-3.5/llvm-3.5/main.log Error: The following dependencies were not installed: libgcc cctools llvm-3.5 cloog gmp isl gcc_select ld64 libmpc mpfr To report a bug, follow the instructions in the guide: http://guide.macports.org/#project.tickets Error: Processing of port netcdf-fortran failed -- Ticket URL: <https://trac.macports.org/ticket/46483> MacPorts <https://www.macports.org/> Ports system for OS X

1 4

[MacPorts] #45737: dbus +no_root deleted my root account on MacOSX 10.5.8
by MacPorts 30 Sep '15

30 Sep '15

#45737: dbus +no_root deleted my root account on MacOSX 10.5.8 ------------------------+-------------------------------- Reporter: newchief@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.2 Keywords: | Port: ------------------------+-------------------------------- Although installation of dbus with +no_root use flag ended with error message that +no_root should not be used on my system, still actions were taken! my "System Administrator (root)" user with uid 0 was replaced by "Message Bus (root)" with uid 503. As a result, I couldn't do "sudo" anymore! Apart from shell commands that require root, all other system administrative operations seem to work. I deleted "Message Bus (root)" user from list of users being logged as my usual user (which is Administrator of this system). Now sudo says there's no root user entry in password file (well, root IS listed in /etc/passwd file, but I guess it is not where MacOSX keeps user info). System is doomed, wihout sudo, no macports anymore. -- Ticket URL: <https://trac.macports.org/ticket/45737> MacPorts <https://www.macports.org/> Ports system for OS X

1 13

[MacPorts] #46733: Multiple definitons of flock TCL command cause problems, e.g., with restore_ports.tcl
by MacPorts 30 Sep '15

30 Sep '15

#46733: Multiple definitons of flock TCL command cause problems, e.g., with restore_ports.tcl --------------------+-------------------------------- Reporter: mta@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: base | Version: 2.3.99 Keywords: | Port: --------------------+-------------------------------- When I ran restore_ports.tcl after upgrading to Yosemite it failed with the error {{{ expected integer but got "-exclusive" obtaining lock on /opt/local/var/macports/registry/.registry.lock while executing "registry::exclusive_lock" (procedure "mportexec" line 32) invoked from within "mportexec $workername $install_target" Unable to execute target 'install' for port 'fftw-3-single': expected integer but got "-exclusive" obtaining lock on /opt/local/var/macports/registry/.registry.lock while executing "install_ports $operationList" (file "restore_ports/restore_ports.tcl" line 300) }}} I tracked this down to the fact that there are two definitions of the flock TCL command installed as part of MacPorts. One is defined in TclX_FlockObjCmd which is in base/vendor/tclx8.4/generic/tclXflock.c. The other is defined in FlockCmd which is in base/src/pextlib1.0/flock.c. Macports seems to want, and get, the latter one. For some reason restore_ports sometimes, but not always, gets the other. Commenting out the call to TclX_FlockInit in Tclxcmd_Init which is in base/vendor/tclx8.4/generic/tclXinit.c seems to have fixed the problem. Unfortunately this isn't all that easy to do as part of the build since these files are installed unchanged from vendor archives. Perhaps a new archive with the appropriate change needs to be part of the MacPorts package. -- Ticket URL: <https://trac.macports.org/ticket/46733> MacPorts <https://www.macports.org/> Ports system for OS X

1 5

[MacPorts] #47283: Backslashes in file names in destroot cause activate to fail
by MacPorts 30 Sep '15

30 Sep '15

#47283: Backslashes in file names in destroot cause activate to fail ---------------------------------+-------------------------------- Reporter: jpo@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: base | Version: 2.3.3 Keywords: backslash, activate | Port: ---------------------------------+-------------------------------- Seems like a simple enough fix for anyone who is familiar with TCL and the base plumbing. I'm not unfortunately. Consider the following: {{{ PortSystem 1.0 name backslash-test version 1 categories devel platforms darwin description tests backslashes in filenames for the activate target fetch {} checksum {} use_configure no build {} destroot { touch "${destroot}${prefix}/share/back\\slash" } }}} Which produces the following results {{{ $ sudo port destroot backslash-test ---> Fetching distfiles for backslash-test ---> Verifying checksums for backslash-test ---> Extracting backslash-test ---> Configuring backslash-test ---> Building backslash-test ---> Staging backslash-test into destroot $ find backslash-test -type f backslash-test/work/.macports.backslash-test.state backslash-test/work/destroot/opt/local/share/back\slash $ sudo port activate backslash-test $ sudo port install backslash-test ---> Activating backslash-test @1_0 Error: org.macports.activate for port backslash-test returned: Image error: Source file /opt/local/var/macports/software/backslash- test/mpextractFapBvhDF/opt/local/share/back lash does not appear to exist (cannot lstat it). Unable to activate port backslash-test. Please see the log file for port backslash-test for details: /opt/local/var/macports/logs /_Volumes_case_src_macports_devel_backslash-test/backslash-test/main.log To report a bug, follow the instructions in the guide: http://guide.macports.org/#project.tickets Error: Processing of port backslash-test failed }}} I've attached a full log of an install of the above portfile from a clean state. -- Ticket URL: <https://trac.macports.org/ticket/47283> MacPorts <https://www.macports.org/> Ports system for OS X

1 11

[MacPorts] #46948: valgrind-devel suppressions file
by MacPorts 28 Sep '15

28 Sep '15

#46948: valgrind-devel suppressions file -------------------------+-------------------------------- Reporter: jhi@… | Owner: macports-tickets@… Type: enhancement | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: valgrind-devel -------------------------+-------------------------------- Hi, I just learned that while the main valgrind port doesn't yet support Yosemite, there is valgrind-devel, which does. Great, thanks for that! It seems that there are some purported leaks from system libraries (I'm running Yosemite 10.10.2, kernel 14.1.0). I used --gen-suppressions=all to create a suppressions file, which you might find useful. I left the suppressions as they were, didn't collapse any of the "stack traces", and named them unimaginatively just "leakN". -- Ticket URL: <https://trac.macports.org/ticket/46948> MacPorts <https://www.macports.org/> Ports system for OS X

1 7

[MacPorts] #47558: valgrind-devel build failure with Xcode 6.3.1 (apple clang 6.1)
by MacPorts 28 Sep '15

28 Sep '15

#47558: valgrind-devel build failure with Xcode 6.3.1 (apple clang 6.1) -----------------------+-------------------------------- Reporter: william@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: valgrind-devel -----------------------+-------------------------------- It looks as if the compiler version detection fails for current Xcode command-line tools … main.log attached. -- Ticket URL: <https://trac.macports.org/ticket/47558> MacPorts <https://www.macports.org/> Ports system for OS X

1 3

[MacPorts] #47649: [Port Abandoned] redis
by MacPorts 25 Sep '15

25 Sep '15

#47649: [Port Abandoned] redis --------------------+---------------------------- Reporter: mf2k@… | Owner: brianjlandau@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: Keywords: | Port: redis --------------------+---------------------------- The maintainer of redis has not responded to tickets in years. The port currently has [query:status=assigned&status=new&status=reopened&port=~redis&col=id&col=summary&col=status&col=type&col=priority&col=milestone&col=component&order=priority an open ticket]. -- Ticket URL: <https://trac.macports.org/ticket/47649> MacPorts <https://www.macports.org/> Ports system for OS X

1 1

[MacPorts] #46029: port submission: qtchooser
by MacPorts 25 Sep '15

25 Sep '15

#46029: port submission: qtchooser ------------------------------+-------------------------------- Reporter: rjvbertin@… | Owner: macports-tickets@… Type: submission | Status: new Priority: High | Milestone: Component: ports | Version: 2.3.3 Keywords: qt4-mac, qt5-mac | Port: ------------------------------+-------------------------------- Please find attached a portfile with an initial set-up for qtchooser, the utility that makes it possible/easy to have Qt4 and Qt5 (and soon enough probably Qt6) installed in parallel. It replaces certain of the standard Qt binaries (like qmake) with links to itself, and then uses configuration files and (optional) arguments to determine what version of the binary in question is called. Thus, with qmake a symlink to qtchooser > qmake --version # prints the version of the default Qt installation > qmake -qt=4 --version # prints the version of the Qt4 installation > qmake -qt=5 --version # prints the version of the Qt5 installation What's missing are the configuration files that will determine which Qt goes and is expected where, though I have given an upbeat by assuming that'd be ${prefix}/libexec/qtN . I hope this helps and brings the possibility to have concurrent qt4-mac and qt5-mac installs a step closer. -- Ticket URL: <https://trac.macports.org/ticket/46029> MacPorts <https://www.macports.org/> Ports system for OS X

1 18

[MacPorts] #47549: opencv @2.4.11_0 - fails to load tiff files
by MacPorts 23 Sep '15

23 Sep '15

#47549: opencv @2.4.11_0 - fails to load tiff files -----------------------------+-------------------------------- Reporter: jose.goicolea@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: opencv -----------------------------+-------------------------------- The latest version of opencv @2.4.11_0 fails when loading .tiff files. However, if I activate the previous version @2.4.10 _0 it runs ok. My operating system is mac OS X 10.7.5, xcode version 4.6.2 I noticed the problem when running auto-multiple-choice, which uses the opencv libraries. With the help of the auto-multiple-choice developer, we have spotted the problem and reproduced it in a basic example. This example includes a simple load.c file and a tiff file, included. The compilation and execution commands for the test case I provide are {{{ gcc -I/opt/local/include/opencv -I/opt/local/include -L/opt/local/lib -lopencv_core -lopencv_highgui -lopencv_imgproc load.c -o load ./load t6_0000.tif }}} Under opencv @2.4.10_0 it compiles and runs ok (no output expected). Under opencv @2.4.11_1 it fails with "Segmentation fault: 11". I have run it with gdb and backtrace, the output is: {{{ Program received signal EXC_BAD_ACCESS, Could not access memory. Reason: KERN_INVALID_ADDRESS at address: 0x000000010720e180 0x000000010060dadc in put1bitbwtile () (gdb) backtrace #0 0x000000010060dadc in put1bitbwtile () #1 0x000000010060b7b0 in gtStripContig () #2 0x000000010060953d in TIFFReadRGBAStrip () #3 0x000000010027f264 in cv::TiffDecoder::readData () #4 0x0000000100262a9a in cv::imread_ () #5 0x000000010026651f in cvLoadImage () #6 0x000000010000322b in main () (gdb) }}} Regards -- Ticket URL: <https://trac.macports.org/ticket/47549> MacPorts <https://www.macports.org/> Ports system for OS X

1 7

[MacPorts] #48390: auto-multiple-choice @1.2.1 prepare for Perl 5.22 migration
by MacPorts 22 Sep '15

22 Sep '15

#48390: auto-multiple-choice @1.2.1 prepare for Perl 5.22 migration ------------------------------+---------------------------------- Reporter: Joel.Brogniart@… | Owner: macports-tickets@… Type: enhancement | Status: new Priority: High | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: auto-multiple-choice ------------------------------+---------------------------------- A patch to ease migration of auto-multiple-choice to Perl 5.22. (For now the migration doesn't work see ticket #47015). -- Ticket URL: <https://trac.macports.org/ticket/48390> MacPorts <https://www.macports.org/> Ports system for OS X

1 13