July 2013 - macports-tickets - lists.macosforge.org

[MacPorts] #39812: tests for scotch 6.0.0 are failing
by MacPorts 21 Jul '13

21 Jul '13

#39812: tests for scotch 6.0.0 are failing --------------------+-------------------------------- Reporter: mk@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: | Port: scotch --------------------+-------------------------------- Scotch 6.0.0 builds fine, but this is what I get on Snow Leopard if I enable the tests by commenting and uncommenting the corresponding lines with "build.target" in scotch's to-be-published Portfile-patch for version 6.0.0: {{{ :info:build make CFLAGS="-O3 -Drestrict=__restrict -DCOMMON_PTHREAD -DCOMMON_PTHREAD_BARRIER -DCOMMON_RANDOM_FIXED_SEED -DCOMMON_TIMING_OLD -DSCOTCH_PTHREAD -DSCOTCH_RENAME -DSCOTCH_PTSCOTCH" CC="/opt/local/bin/openmpicc" SCOTCHLIBS="-lptscotch -lscotch -lptscotcherr" check_strat_par check_scotch_dgraph_band check_scotch_dgraph_grow check_scotch_dgraph_redist :info:build make[2]: Entering directory `/opt/local/var/macports/build/_Users_marko_WC_SVN_MacPorts_science_scotch/scotch/work/scotch_6.0.0/src/check' :info:build /opt/local/bin/openmpicc -O3 -Drestrict=__restrict -DCOMMON_PTHREAD -DCOMMON_PTHREAD_BARRIER -DCOMMON_RANDOM_FIXED_SEED -DCOMMON_TIMING_OLD -DSCOTCH_PTHREAD -DSCOTCH_RENAME -DSCOTCH_PTSCOTCH -I../../include -L../../lib test_strat_par.c -o test_strat_par -lptscotch -lscotch -lptscotcherr -lm :info:build ./test_strat_par :info:build Parallel mapping strategy, SCOTCH_STRATDEFAULT :info:build Parallel mapping strategy, SCOTCH_STRATRECURSIVE :info:build Parallel ordering strategy, SCOTCH_STRATDEFAULT :info:build Parallel ordering strategy, SCOTCH_STRATLEVELMAX :info:build Parallel ordering strategy, SCOTCH_STRATLEVELMIN :info:build Parallel ordering strategy, SCOTCH_STRATLEVELMAX | SCOTCH_STRATLEVELMIN :info:build Parallel ordering strategy, SCOTCH_STRATLEAFSIMPLE :info:build Parallel ordering strategy, SCOTCH_STRATSEPASIMPLE :info:build /opt/local/bin/openmpicc -O3 -Drestrict=__restrict -DCOMMON_PTHREAD -DCOMMON_PTHREAD_BARRIER -DCOMMON_RANDOM_FIXED_SEED -DCOMMON_TIMING_OLD -DSCOTCH_PTHREAD -DSCOTCH_RENAME -DSCOTCH_PTSCOTCH -I../../include -L../../lib test_scotch_dgraph_band.c -o test_scotch_dgraph_band -lptscotch -lscotch -lptscotcherr -lm :info:build openmpirun -n 4 ./test_scotch_dgraph_band data/bump.grf :info:build ./test_scotch_dgraph_band(0): ERROR: main: Cannot initialize (2) :info:build ./test_scotch_dgraph_band(1): ERROR: main: Cannot initialize (2) :info:build ./test_scotch_dgraph_band(2): ERROR: main: Cannot initialize (2) :info:build ./test_scotch_dgraph_band(3): ERROR: main: Cannot initialize (2) :info:build -------------------------------------------------------------------------- :info:build openmpirun has exited due to process rank 3 with PID 23992 on :info:build node markos-imac exiting improperly. There are three reasons this could occur: :info:build :info:build 1. this process did not call "init" before exiting, but others in :info:build the job did. This can cause a job to hang indefinitely while it waits :info:build for all processes to call "init". By rule, if one process calls "init", :info:build then ALL processes must call "init" prior to termination. :info:build :info:build 2. this process called "init", but exited without calling "finalize". :info:build By rule, all processes that call "init" MUST call "finalize" prior to :info:build exiting or it will be considered an "abnormal termination" :info:build :info:build 3. this process called "MPI_Abort" or "orte_abort" and the mca parameter :info:build orte_create_session_dirs is set to false. In this case, the run-time cannot :info:build detect that the abort call was an abnormal termination. Hence, the only :info:build error message you will receive is this one. :info:build :info:build This may have caused other processes in the application to be :info:build terminated by signals sent by openmpirun (as reported here). :info:build :info:build You can avoid this message by specifying -quiet on the openmpirun command line. :info:build :info:build -------------------------------------------------------------------------- :info:build make[2]: *** [check_scotch_dgraph_band] Error 1 :info:build make[2]: Leaving directory `/opt/local/var/macports/build/_Users_marko_WC_SVN_MacPorts_science_scotch/scotch/work/scotch_6.0.0/src/check' :info:build make[1]: *** [ptcheck] Error 2 }}} -- Ticket URL: <https://trac.macports.org/ticket/39812> MacPorts <http://www.macports.org/> Ports system for OS X

1 2

[MacPorts] #39815: latexdiff checksum error in upgrade from 0.5.1 to 1.0.2_2
by MacPorts 21 Jul '13

21 Jul '13

#39815: latexdiff checksum error in upgrade from 0.5.1 to 1.0.2_2 -----------------------+-------------------------------- Reporter: trg818@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: latexdiff | Port: -----------------------+-------------------------------- While making an extensive upgrade, the following error occurs during installation of latexdiff and prevents the upgrade of texlive:[[BR]] ---> Computing dependencies for latexdiff ---> Fetching archive for latexdiff ---> Attempting to fetch latexdiff-1.0.2_2.darwin_9.noarch.tbz2 from http://lil.fr.packages.macports.org/latexdiff ---> Attempting to fetch latexdiff-1.0.2_2.darwin_9.noarch.tbz2 from http://mse.uk.packages.macports.org/sites/packages.macports.org/latexdiff ---> Attempting to fetch latexdiff-1.0.2_2.darwin_9.noarch.tbz2 from http://packages.macports.org/latexdiff ---> Fetching distfiles for latexdiff ---> Verifying checksum(s) for latexdiff Error: Checksum (md5) mismatch for latexdiff.zip Error: Checksum (sha1) mismatch for latexdiff.zip Error: Checksum (rmd160) mismatch for latexdiff.zip Error: org.macports.checksum for port latexdiff returned: Unable to verify file checksums The logfile gives the following additional information: :info:checksum Portfile checksum: latexdiff.zip md5 6353392165de3066aba8323ffe5e b9c3 :info:checksum Distfile checksum: latexdiff.zip md5 8cdf86a504513b27c79e1d52b2a9 567d :error:checksum Checksum (sha1) mismatch for latexdiff.zip :info:checksum Portfile checksum: latexdiff.zip sha1 6a13b516514360998c033d310a8 b114bc661cddb :info:checksum Distfile checksum: latexdiff.zip sha1 b56629c619465a3f7c2b12f17f0 5980b579e5a62 :error:checksum Checksum (rmd160) mismatch for latexdiff.zip :info:checksum Portfile checksum: latexdiff.zip rmd160 76bd16b964f37ad74da0591c5 1d6bea7f137278d :info:checksum Distfile checksum: latexdiff.zip rmd160 e7b3e085869b1173a4b764bfa aaf3f62b1e9c33f :info:checksum The correct checksum line may be: :info:checksum checksums md5 8cdf86a504513b27c79e1d52b2a9567d \ sha1 b56629c619465a3f7c2b12f17f05980b579e5a62 \ rmd160 e7b3e085869b1173a4b764bfaaaf3f62b1e9c33f -- Ticket URL: <https://trac.macports.org/ticket/39815> MacPorts <http://www.macports.org/> Ports system for OS X

1 7

[MacPorts] #39571: yelp 2.31.6 fails to build for +quartz
by MacPorts 20 Jul '13

20 Jul '13

#39571: yelp 2.31.6 fails to build for +quartz --------------------------------+-------------------------------- Reporter: c.herbig@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: yelp gnome +quartz | Port: yelp --------------------------------+-------------------------------- I know that it is not officially supported as a variant for yelp, but I suppose if no one points it out, that guarantees it will never work. In trying to compile yelp +quartz, everything seems to work up to the application.c file, when it generates errors due to specific gdk x11 function calls. I hope that something can be done to work around this. You might be interested to know that if you attempt to install yelp 3.x, the same problem is there. Here is the relevant (hopefully) part of the log: {{{ :info:build yelp-application.c: In function ‘application_uri_resolved’: :info:build yelp-application.c:546: warning: implicit declaration of function ‘gdk_x11_window_move_to_current_desktop’ :info:build yelp-application.c:546: warning: nested extern declaration of ‘gdk_x11_window_move_to_current_desktop’ :info:build yelp-application.c:554: warning: implicit declaration of function ‘gdk_x11_get_server_time’ :info:build yelp-application.c:554: warning: nested extern declaration of ‘gdk_x11_get_server_time’ :info:build make[2]: *** [yelp-yelp-application.o] Error 1 }}} P.S. I have never posted a bug ticket before, so I apologize in advance for any problems. I would also like to thank the maintainers for finally updating yelp to a version that will compile on 64 bit x86. -- Ticket URL: <https://trac.macports.org/ticket/39571> MacPorts <http://www.macports.org/> Ports system for OS X

1 4

[MacPorts] #39543: Checksum failure on sane-backends-1.0.22.tar.gz
by MacPorts 20 Jul '13

20 Jul '13

#39543: Checksum failure on sane-backends-1.0.22.tar.gz ---------------------------+-------------------------------- Reporter: drkem@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: sane-backends | Port: ---------------------------+-------------------------------- Mac OS 10.8.4; XCode 4.6.3 Following error running 'port upgrade outdated' command: Error: Checksum (md5) mismatch for sane-backends-1.0.22.tar.gz Error: Checksum (sha1) mismatch for sane-backends-1.0.22.tar.gz Error: Checksum (rmd160) mismatch for sane-backends-1.0.22.tar.gz Error: org.macports.checksum for port sane-backends returned: Unable to verify file checksums Please advise. Thanks. (Inexperienced non-IT professional user; keep answer simple and be gentle, please.) -- Ticket URL: <https://trac.macports.org/ticket/39543> MacPorts <http://www.macports.org/> Ports system for OS X

1 3

[MacPorts] #39333: scotch build error, mpicc not found
by MacPorts 20 Jul '13

20 Jul '13

#39333: scotch build error, mpicc not found -----------------------------+-------------------------------- Reporter: Peter.Danecek@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: Keywords: | Port: scotch -----------------------------+-------------------------------- I guess the problem is here: {{{ [...] mpicc -O3 -Drestrict=__restrict -DCOMMON_PTHREAD -DCOMMON_RANDOM_FIXED_SEED -DCOMMON_TIMING_OLD -DSCOTCH_PTHREAD -DSCOTCH_RENAME -DSCOTCH_RENAME_PARSER -DSCOTCH_PTSCOTCH -DSCOTCH_VERSION=5 -DSCOTCH_RELEASE=1 -DSCOTCH_PATCHLEVEL=10 dummysizes.c -o dummysizes -lm make[2]: mpicc: No such file or directory [...] }}} The path should be given explicitly. Usually, this probably isn't noted, because there are some link or similar around, but on my freshly installed system its not set yet. -- Ticket URL: <https://trac.macports.org/ticket/39333> MacPorts <http://www.macports.org/> Ports system for OS X

1 13

[MacPorts] #37231: Build script(s) expecting script tclIndex in the wrong place
by MacPorts 20 Jul '13

20 Jul '13

#37231: Build script(s) expecting script tclIndex in the wrong place ------------------------+-------------------------------- Reporter: macports@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.2 Keywords: tclIndex | Port: ------------------------+-------------------------------- Some builds (such as DSDP) fail because ''port'' cannot find the script {{{ /System/Library/Frameworks/Tcl.framework/Versions/8.5/Resources/tclIndex }}} On a clean install of Mountain Lion, XCode 4.5.2 this file is found instead at {{{ /System/Library/Frameworks/Tk.framework/Versions/8.5/Resources/Scripts/tclIndex }}} Workarounds include pre-building a package that knows the right script location, or installing a version of TCL that places the script in the former spot. -- Ticket URL: <https://trac.macports.org/ticket/37231> MacPorts <http://www.macports.org/> Ports system for Mac OS

1 3

[MacPorts] #39808: Updated Boost to 1.54.0
by MacPorts 20 Jul '13

20 Jul '13

#39808: Updated Boost to 1.54.0 ------------------------------+-------------------------------- Reporter: brad.clemetson@… | Owner: macports-tickets@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: | Port: ------------------------------+-------------------------------- Updated the boost library to compile and install 1.54.0 -- Ticket URL: <https://trac.macports.org/ticket/39808> MacPorts <http://www.macports.org/> Ports system for OS X

1 8

[MacPorts] #39805: libcerf: compile error on 10.6
by MacPorts 20 Jul '13

20 Jul '13

#39805: libcerf: compile error on 10.6 ---------------------+--------------------- Reporter: mojca@… | Owner: mojca@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: Keywords: | Port: libcerf ---------------------+--------------------- libcerf fails to compile on 10.6 buildbot with {{{ w_of_z.c:84:40: error: imaginary constants are a GCC extension w_of_z.c:176:43: error: imaginary constants are a GCC extension w_of_z.c:180:59: error: imaginary constants are a GCC extension w_of_z.c:225:45: error: imaginary constants are a GCC extension }}} https://build.macports.org/builders/buildports-snowleopard- x86_64/builds/18815 It seems that this is either gcc 4.2's fault: * http://gcc.gnu.org/gcc-4.2/c99status.html or a problem with combining {{{-pedantic}}} and {{{-Werror}}} There is no hurry to fix this (deadline would be the release of gnuplot months ahead) and simply declaring in the portfile that it doesn't support 10.6 (or that it needs a different compiler) should be ok as well. -- Ticket URL: <https://trac.macports.org/ticket/39805> MacPorts <http://www.macports.org/> Ports system for OS X

1 2

[MacPorts] #38706: Update of digikam port
by MacPorts 19 Jul '13

19 Jul '13

#38706: Update of digikam port ----------------------+-------------------------------- Reporter: jan@… | Owner: macports-tickets@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: haspatch | Port: digikam ----------------------+-------------------------------- I did a few updates to the digikam port: * Added kde4-baseapps and kde4-runtime dependency. They are not absolutely necessary. But some functionality like trash or Flickr authentication depends on these. This change should resolve #31926 and #37592 * Changed hard-coded `/opt/local` to `${prefix}` -- Ticket URL: <https://trac.macports.org/ticket/38706> MacPorts <http://www.macports.org/> Ports system for OS X

1 8

[MacPorts] #39785: netcdf fails to index
by MacPorts 19 Jul '13

19 Jul '13

#39785: netcdf fails to index --------------------+----------------------- Reporter: cal@… | Owner: takeshi@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.1.3 Keywords: | Port: netcdf --------------------+----------------------- When I try to generate PortIndex, I see: {{{ Failed to parse file science/netcdf/Portfile: grep: /opt/local/lib/libhdf5.settings: No such file or directory }}} Please ensure the Portfile does execute correctly without the file /opt/local/lib/libhdf5.settings present (e.g. by wrapping the call to grep with a `catch` statement). If you want to change the default variants and/or throw an error depending on whether a certain variant is present in a dependent port, you can use the `active_variants` PortGroup (see source:trunk/dports/_resources/port1.0/group/active_variants-1.1.tcl, which has some documentation). -- Ticket URL: <https://trac.macports.org/ticket/39785> MacPorts <http://www.macports.org/> Ports system for OS X

1 4