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

Re: [MacPorts] #24956: Having trouble using mysql5 and php5
by MacPorts 05 Mar '15

05 Mar '15

#24956: Having trouble using mysql5 and php5 ----------------------+-------------------------------- Reporter: samii@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: wiki | Version: Resolution: | Keywords: Port: php5 | ----------------------+-------------------------------- Changes (by jmr@…): * cc: pixilla@… (added) Comment: Should this still be open? Does it even describe a bug? -- Ticket URL: <https://trac.macports.org/ticket/24956#comment:18> MacPorts <https://www.macports.org/> Ports system for OS X

1 0

[MacPorts] #47049: gfortran backtrace issue
by MacPorts 05 Mar '15

05 Mar '15

#47049: gfortran backtrace issue --------------------------------+-------------------------------- Reporter: barry.j.mcinnes@… | Owner: macports-tickets@… Type: request | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: gfortran backtrace | Port: gfortran --------------------------------+-------------------------------- We have gdb to work and have set the com.apple.taskgated.plist to add the -p as specified in the macports install of gdb. We are 10.10.2 and latest as of March 5 with macports and fortran 4.9.2. Just concerned that Linux fortran gives the extra lin information and macports version does not. Here is the user description of a small test program, that displays this issue. He has a much larger program he wants the line numbers displayed if at all possible on the macports fortran. Compare this with the Linux backtrace lower down. The key difference for this very short demo program is that the Linux backtrace shows the source label, source file name, and source line number in the backtrace. "MAIN__ at btest.f90:12". This is the symbol information from a non-stripped object file, a.out in this case. The backtrace on Mac is missing all of this. I am not concerned about the first three lines in the Linux call stack. These are probably low-level calls within a stripped runtime library. This is not a new problem. Backtrace symbols were missing in the last several versions of gfortran, Mac OS, and Xcode on PSD Macs. mac56:/home/dallured/btest 7> gfortran --version | grep gcc GNU Fortran (MacPorts gcc49 4.9.2_1) 4.9.2 mac56:/home/dallured/btest 8> gfortran -g -ffpe-trap=zero btest.f90 mac56:/home/dallured/btest 9> ./a.out Start program. Divide by zero. Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation. Backtrace for this error: #0 0x10b15bfbc #1 0x10b15c4df #2 0x7fff94646f19 #3 0x10b155cb7 #4 0x10b155e4e Floating exception ================================================== linux64:~/btest 1025> /usr/locale/bin/gfortran --version | grep GCC GNU Fortran (GCC) 4.9.2 linux64:~/btest 1026> /usr/locale/bin/gfortran -g -ffpe-trap=zero btest.f90 linux64:~/btest 1027> ./a.out Start program. Divide by zero. Program received signal SIGFPE: Floating-point exception - erroneous arithmetic operation. Backtrace for this error: #0 0x7F225DDA8347 #1 0x7F225DDA894E #2 0x3585A3269F #3 0x400978 in MAIN__ at btest.f90:12 Floating exception (core dumped) -- Ticket URL: <https://trac.macports.org/ticket/47049> MacPorts <https://www.macports.org/> Ports system for OS X

1 1

[MacPorts] #47027: mesa @10.4.4_2 build failure
by MacPorts 05 Mar '15

05 Mar '15

#47027: mesa @10.4.4_2 build failure ------------------------------+-------------------------------- Reporter: jane.r.allison@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: lion | Port: mesa ------------------------------+-------------------------------- mesa build fails in following circumstances: 1. As part of overall upgrade to outdated ports (see mesa-upgrade.log), i.e. {{{sudo -E port upgrade outdated}}} 2. When explicitly (re)installing mesa using default settings (see mesa- default.log), i.e. {{{port -E sudo install mesa}}} 3. When explicitly (re)installing mesa using llvm-gcc-4.2 as recommended on the Problem Hotline page (see mesa-llvm-gcc-4.2.log), i.e. {{{sudo -E port install mesa configure.compiler=llvm-gcc-4.2}}} 4. When attempting to install texlive, which in turn installs pymol, which requires mesa (see mesa-pymol.log), i.e. {{{sudo -E port install texlive +full -medium}}} To make matters more confusing, different errors are encountered each time. Machine specs are: OS 10.7.5, Xcode version 4.6.1 (but am using ports via the command line so not sure if this is relevant). -- Ticket URL: <https://trac.macports.org/ticket/47027> MacPorts <https://www.macports.org/> Ports system for OS X

1 9

[MacPorts] #45928: cvc4 @1.4 Adding Java variant - updated Portfile
by MacPorts 05 Mar '15

05 Mar '15

#45928: cvc4 @1.4 Adding Java variant - updated Portfile ----------------------------+-------------------------------- Reporter: corwin.amber@… | Owner: macports-tickets@… Type: enhancement | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.2 Keywords: java | Port: cvc4 ----------------------------+-------------------------------- Since CVC4 offers a Java library, which is not build by default but requires ./configure --enable-language-bindings=java, I added a variant (+java) so the user can request the jarfile and JNI code to be built. This requires an installation of a JDK in /System/Library/Frameworks. -- Ticket URL: <https://trac.macports.org/ticket/45928> MacPorts <https://www.macports.org/> Ports system for OS X

1 4

[MacPorts] #46931: vcs_blackbox @1.20150222 Update to latest upstream
by MacPorts 05 Mar '15

05 Mar '15

#46931: vcs_blackbox @1.20150222 Update to latest upstream --------------------+-------------------------------- Reporter: tal@… | Owner: macports-tickets@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: vcs_blackbox --------------------+-------------------------------- New upstream of vcs_blackbox. github.setup and checksums updated. -- Ticket URL: <https://trac.macports.org/ticket/46931> MacPorts <https://www.macports.org/> Ports system for OS X

1 7

[MacPorts] #47046: when will nginx@1.7.10 be available (1.7.9 has a bug on osx)
by MacPorts 05 Mar '15

05 Mar '15

#47046: when will nginx(a)1.7.10 be available (1.7.9 has a bug on osx) --------------------------+-------------------------------- Reporter: oskarstark@… | Owner: macports-tickets@… Type: request | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: --------------------------+-------------------------------- @see changelog here: http://nginx.org/en/CHANGES Fixxed in 1.7.10: {{{ *) Bugfix: alerts "sendfile() failed" if the "sendfile" directive was used on OS X; the bug had appeared in 1.7.8. }}} a downgrade to 1.7.7 is not possible, because 1.7.9 is the only available port! best Oskar -- Ticket URL: <https://trac.macports.org/ticket/47046> MacPorts <https://www.macports.org/> Ports system for OS X

1 4

[MacPorts] #47048: Update: t1utils 1.39
by MacPorts 05 Mar '15

05 Mar '15

#47048: Update: t1utils 1.39 ----------------------------+-------------------------------- Reporter: mschamschula@… | Owner: macports-tickets@… Type: update | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: haspatch | Port: t1utils ----------------------------+-------------------------------- t1utils has been updated to version 1.39. {{{ Version 1.37-1.39: Fix an infinite loop on some fonts and fix some security issues. }}} -- Ticket URL: <https://trac.macports.org/ticket/47048> MacPorts <https://www.macports.org/> Ports system for OS X

1 2

[MacPorts] #46356: [Port Abandoned] cvc4
by MacPorts 05 Mar '15

05 Mar '15

#46356: [Port Abandoned] cvc4 --------------------+----------------------- Reporter: mf2k@… | Owner: mdeters@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: Keywords: | Port: cvc4 --------------------+----------------------- The maintainer of cvc4 has no Trac activity and has not responded to ticket:45928 in a while. -- Ticket URL: <https://trac.macports.org/ticket/46356> MacPorts <https://www.macports.org/> Ports system for OS X

1 1

[MacPorts] #46926: libvisio-0.1 @0.1.1 build fails on Leopard/PPC
by MacPorts 05 Mar '15

05 Mar '15

#46926: libvisio-0.1 @0.1.1 build fails on Leopard/PPC -----------------------+-------------------------------- Reporter: khepler@… | Owner: macports-tickets@… Type: defect | Status: new Priority: Normal | Milestone: Component: ports | Version: 2.3.3 Keywords: | Port: libvisio-0.1 -----------------------+-------------------------------- Building libvisio-0.1 fails on Leopard/PPC (PowerPC G5) with the error: {{{ ...snip... /opt/local/include/boost/spirit/home/classic/symbols/symbols.hpp:105: warning: declaration of 'result' shadows a member of 'this' CXX VSDGeometryList.lo CXX VSDOutputElementList.lo CXX VSDPages.lo CXX VSDParagraphList.lo CXX VSDParser.lo CXX VSDMetaData.lo VSDMetaData.cpp:229: error: integer constant is too large for 'long' type make[3]: *** [VSDMetaData.lo] Error 1 make[3]: *** Waiting for unfinished jobs.... ...snip... }}} The offending line in VSDMetaData.cpp is: {{{ static const uint64_t epoch = 11644473600; }}} Is this considered an upstream bug or should it be patched in Macports because the compiler it chose assumes a non-suffixed literal integer is a long? Secondly, after applying the supplied patch (that appends ULL to the literal), the build fails if doxygen is not installed. -- Ticket URL: <https://trac.macports.org/ticket/46926> MacPorts <https://www.macports.org/> Ports system for OS X

1 6

Re: [MacPorts] #41600: bison: update to 3.0.2
by MacPorts 05 Mar '15

05 Mar '15

#41600: bison: update to 3.0.2 ------------------------------+---------------------- Reporter: akim.demaille@… | Owner: larryv@… Type: update | Status: assigned Priority: Normal | Milestone: Component: ports | Version: Resolution: | Keywords: haspatch Port: bison | ------------------------------+---------------------- Comment (by larryv@…): Replying to [comment:44 rjvbertin@…]: > BTW, the portfile claims that mww is the maintainer, is that no longer > the case? This ticket is well past the [https://guide.macports.org/chunked/project .update-policies.html#project.update-policies.nonmaintainer maintainer timeout period]. -- Ticket URL: <https://trac.macports.org/ticket/41600#comment:45> MacPorts <https://www.macports.org/> Ports system for OS X

1 0