Gcc-5 fix forwarding to upstream (Was: Bug#797526: mapsembler2: FTBFS with GCC-5 - undefined reference to `prefix_trashable[abi:cxx11]')
Hi Olivier,
since debian/copyright of mapsembler2 is lacking Upstream-Contact
information I'm forwarding this to you since as far as I know you
have contact to upstream.
I'm not a user of mapsembler2 and thus have no idea how to trigger
the remove statement below for sensible testing. Please forward
the patch
https://anonscm.debian.org/viewvc/debian-med/trunk/packages/mapsembler2/trunk/debian/patches/gcc-5.patch?view=markup
to upstream for thorough revision. If you get an OK (or test it
yourself) feel free to upload (or ping me to do so).
Hope this helps and thanks to Gianfranco
Andreas.
On Tue, Sep 01, 2015 at 08:34:33AM +0000, Gianfranco Costamagna wrote:
> Hi Andreas
>
>
> /tmp/mapsembler2-2.2.3+dfsg $ grep prefix_trashable . -R |grep -v debian
> ./kissreads_graph/commons.cpp:char * prefix_trashable=(char *)"trashmeplease";
> ./mapsembler2_extend/mapsembler_extend.cpp:extern string prefix_trashable;
> ./mapsembler2_extend/mapsembler_extend.cpp: command = string("rm -f ")+prefix_trashable.c_str()+string("*"); system((char *)command.c_str());
> ./mapsembler2_extend/commons.cpp:char * prefix_trashable=(char *)"trashmeplease";
> ./mapsembler2_extend/Fragment.cpp:extern char * prefix_trashable;
> ./mapsembler2_extend/Fragment.cpp: string linear_seq_name = prefix_trashable+(*it);
> ./mapsembler2_extend/Fragment.cpp: string linear_seq_name = prefix_trashable+(*it);
> ./mapsembler2_extend/Fragment.cpp: graph.load_nodes_extremities((char *)(prefix_trashable+left_extrem_kmers[i]).c_str());
> ./mapsembler2_extend/Fragment.cpp: id_els first_id_els=graph.construct_graph(prefix_trashable+left_extrem_kmers[i], "LEFT");
> ./mapsembler2_extend/Fragment.cpp: graph.load_nodes_extremities((char *)(prefix_trashable+right_extrem_kmers[i]).c_str());
> ./mapsembler2_extend/Fragment.cpp: id_els first_id_els=graph.construct_graph(prefix_trashable+right_extrem_kmers[i], "RIGHT");
> ./mapsembler2_extend/Fragment.cpp: Bank * Left = new Bank((char*)(prefix_trashable+left_kmer).c_str());
> ./mapsembler2_extend/Fragment.cpp: Bank * Right = new Bank((char*)(prefix_trashable+right_kmer).c_str());
>
>
>
> specially lines:
> ./kissreads_graph/commons.cpp:char * prefix_trashable=(char *)"trashmeplease";
> ./mapsembler2_extend/mapsembler_extend.cpp:extern string prefix_trashable;
>
> ./mapsembler2_extend/commons.cpp:char * prefix_trashable=(char *)"trashmeplease";
> ./mapsembler2_extend/Fragment.cpp:extern char * prefix_trashable;
>
>
> you define it as extern string, and you define it as char * in the real definition.
>
>
>
>
> NOTE: I'm *not* a c++ savvy person, and I didn't test the patch below.
> But you asked for an hint, so here it is:
>
>
> file: mapsembler2_extend/mapsembler_extend.cpp
>
> -extern string prefix_trashable;
> +extern char* prefix_trashable;
>
>
> - command = string("rm -f ")+prefix_trashable.c_str()+string("*"); system((char *)command.c_str());
> + command = string("rm -f ")+prefix_trashable+string("*"); system((char *)command.c_str());
>
> or maybe even better
> - command = string("rm -f ")+prefix_trashable.c_str()+string("*"); system((char *)command.c_str());
> + command = string("rm -f ")+string(prefix_trashable)+string("*"); system((char *)command.c_str());
>
>
> (note, since this is called in an rm -f function, I don't want to give an untested patch).
>
> Please check if that "command" is correct during run, by commenting out the system call, and
>
> printing a "cout << command" or similar somewhere.
>
> If you are comfortable that the prefix_trashable is good, and the rm is not deleting files somewhere else,
> I guess your patch is good :)
>
> cheers,
>
> G.
>
>
>
> Il Martedì 1 Settembre 2015 9:56, Andreas Tille <tille@debian.org> ha scritto:
> Hi,
>
> any idea what librarie might need a transition to enable mapsembler2
> building?
>
> Thanks for any hint
>
> Andreas.
>
> On Mon, Aug 31, 2015 at 12:54:43PM +0200, Gilles Filippini wrote:
> > Source: mapsembler2
> > Version: 2.2.3+dfsg-2
> > Severity: serious
> > Justification: FTBFS
> >
> > -----BEGIN PGP SIGNED MESSAGE-----
> > Hash: SHA256
> >
> > Hi,
> >
> > During a rebuild against GCC-5 mapsembler2 FTBFS with:
> >
> > /tmp/ccXWIZkS.o: In function `main':
> > /«BUILDDIR»/mapsembler2-2.2.3+dfsg/mapsembler2_extend/mapsembler_extend.cpp:320: undefined reference to `prefix_trashable[abi:cxx11]'
> > collect2: error: ld returned 1 exit status
> > make[2]: *** [mapsembler_extend] Error 1
> > makefile:113: recipe for target 'mapsembler_extend' failed
> > make[2]: Leaving directory '/«BUILDDIR»/mapsembler2-2.2.3+dfsg/mapsembler2_extend'
> > make[1]: *** [override_dh_auto_build] Error 2
> > debian/rules:27: recipe for target 'override_dh_auto_build' failed
> > make[1]: Leaving directory '/«BUILDDIR»/mapsembler2-2.2.3+dfsg'
> > make: *** [build] Error 2
> > debian/rules:21: recipe for target 'build' failed
> > dpkg-buildpackage: error: debian/rules build gave error exit status 2
> >
> > Full build log on request.
> >
> > Thanks,
> >
> > _g.
> >
> > - -- System Information:
> > Debian Release: stretch/sid
> > APT prefers testing-updates
> > APT policy: (500, 'testing-updates'), (500, 'testing')
> > Architecture: amd64 (x86_64)
> > Foreign Architectures: i386
> >
> > Kernel: Linux 4.1.0-1-amd64 (SMP w/2 CPU cores)
> > Locale: LANG=fr_FR.UTF-8, LC_CTYPE=fr_FR.UTF-8 (charmap=UTF-8)
> > Shell: /bin/sh linked to /bin/dash
> > Init: systemd (via /run/systemd/system)
> >
> > -----BEGIN PGP SIGNATURE-----
> > Version: GnuPG v1
> >
> > iQEcBAEBCAAGBQJV5DJpAAoJEO/obGx//s+DCHUIAIDi4dIxSFsK96zlMbDjXjvQ
> > vSUOz/wkMt6PS/+NFZ6lAsjUQnuviZVl/le9fgnGP99CHYT+UAxshE2af46w5MZq
> > R+SMfdGCbGLMO2NhXtiaSMk5ooWKpn+Rk2WKLZQ2rT5B++G0MiXlc3u/i0YcKBsJ
> > fmNgUELLQKInZt2OvKTO/qO5CrQOhBaZvAWD94sxv5w988MEhrk0DVHbERrdfZ1C
> > K1nHkl9APNM4P+EXCfHxgfR1HKbCwIUhDntnzQduPOD3b/vBNPFUzUcFyH9C+ghh
> > ZCs8nQETxEr9Yt9wr2z/90Lgvtubbi8CxowAQ2dbJALzll8QFp1OB63r2vwqM4U=
> > =1W/8
> > -----END PGP SIGNATURE-----
> >
> > _______________________________________________
> > Debian-med-packaging mailing list
> > Debian-med-packaging@lists.alioth.debian.org
> > http://lists.alioth.debian.org/cgi-bin/mailman/listinfo/debian-med-packaging
>
> --
> http://fam-tille.de
>
--
http://fam-tille.de
Reply to: