Re: Boost updated to 1.58.0 in rawhide and f23
Rex Dieter wrote: > Rex Dieter wrote: > >> I'm helping to sort out the tagging mess, and submit things in smaller >> batches. > > In 6 batches of ~50 packages each, I have it all submitted for -testing in > bodhi. > > I'll followup with update id's once they get pushed. https://admin.fedoraproject.org/updates/FEDORA-2015-13504 https://admin.fedoraproject.org/updates/FEDORA-2015-13449 https://admin.fedoraproject.org/updates/FEDORA-2015-13531 https://admin.fedoraproject.org/updates/FEDORA-2015-13490 https://admin.fedoraproject.org/updates/FEDORA-2015-13454 https://admin.fedoraproject.org/updates/FEDORA-2015-13428 -- rex -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 13/08/15 10:49 -0500, Rex Dieter wrote: Rex Dieter wrote: I'm helping to sort out the tagging mess, and submit things in smaller batches. In 6 batches of ~50 packages each, I have it all submitted for -testing in bodhi. I'll followup with update id's once they get pushed. Thanks, Rex, I really appreciate your help getting this sorted. I'll install them from updates-testing and test and leave karma ASAP. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Rex Dieter wrote: > I'm helping to sort out the tagging mess, and submit things in smaller > batches. In 6 batches of ~50 packages each, I have it all submitted for -testing in bodhi. I'll followup with update id's once they get pushed. -- Rex -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Jonathan Wakely wrote: > On 11/08/15 15:43 +0100, Jonathan Wakely wrote: >>On 10/08/15 12:24 -0600, Kevin Fenzi wrote: >>>On Mon, 10 Aug 2015 17:51:30 +0100 >>>José Matos wrote: >>> On Saturday 18 July 2015 12:46:51 Jonathan Wakely wrote: > Hi, > > I've pushed a new version of Boost, 1.58.0, to rawhide and f23, > which will require all packages that depend on Boost to be rebuilt. > The plan was to update to 1.59.0 but that isn't going to be > released in time for the F23 schedule, so I'll update rawhide to > 1.59.0 at a later date. > > For the F23 rebuilds please use the tag f23-boost e.g. > > fedpkg build --target f23-boost Hi Jonathan, what is the status of this update for Fedora 23? >>> >>>Last I heard there were bodhi issues. It was unable to handle an update >>>with that many packages. ;( >> >>Right, it has proved impossible to create the update. >> >>>So, I think it needs to just be split up into several updates, or >>>perhaps Luke can come up with some way to get it processed. ;( >> >>If splitting it up is acceptable I can do that, but I think there may >>be another problem: when I was trying to add builds to the update they >>were not added correctly, but did get tagged in koji as if added to >>the update. >> >>So although I deleted the original (unsuccessful) update, now I can't >>add lots of the builds to a new update, because bodhi says there is >>already an update for that build. >> >>It looks like bodhi doesn't work for large updates, and corrupts the >>DB so that updates and koji tags get out of sync :-( > > > e.g. I'm getting an error like this for dozens of builds: > > ncmpcpp-0.6.4-5.fc23 update already exists! > > Is that error given because in koji that build has the tag > 'f23-updates-testing-pending' and if so, can I just delete the tag? > > How do I verify if there really is an update (submitted by someone > else) for that build, before removing the tag? (I'm pretty sure there > isn't an update, and it's just caused by the bodhi failures I was > getting while trying to create a huge update). > > Alternatively, should I just not include those builds in my new > upadte, since they are already tagged (even if not part of any > update)? I'm helping to sort out the tagging mess, and submit things in smaller batches. -- Rex -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 11/08/15 15:43 +0100, Jonathan Wakely wrote: On 10/08/15 12:24 -0600, Kevin Fenzi wrote: On Mon, 10 Aug 2015 17:51:30 +0100 José Matos wrote: On Saturday 18 July 2015 12:46:51 Jonathan Wakely wrote: Hi, I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which will require all packages that depend on Boost to be rebuilt. The plan was to update to 1.59.0 but that isn't going to be released in time for the F23 schedule, so I'll update rawhide to 1.59.0 at a later date. For the F23 rebuilds please use the tag f23-boost e.g. fedpkg build --target f23-boost Hi Jonathan, what is the status of this update for Fedora 23? Last I heard there were bodhi issues. It was unable to handle an update with that many packages. ;( Right, it has proved impossible to create the update. So, I think it needs to just be split up into several updates, or perhaps Luke can come up with some way to get it processed. ;( If splitting it up is acceptable I can do that, but I think there may be another problem: when I was trying to add builds to the update they were not added correctly, but did get tagged in koji as if added to the update. So although I deleted the original (unsuccessful) update, now I can't add lots of the builds to a new update, because bodhi says there is already an update for that build. It looks like bodhi doesn't work for large updates, and corrupts the DB so that updates and koji tags get out of sync :-( e.g. I'm getting an error like this for dozens of builds: ncmpcpp-0.6.4-5.fc23 update already exists! Is that error given because in koji that build has the tag 'f23-updates-testing-pending' and if so, can I just delete the tag? How do I verify if there really is an update (submitted by someone else) for that build, before removing the tag? (I'm pretty sure there isn't an update, and it's just caused by the bodhi failures I was getting while trying to create a huge update). Alternatively, should I just not include those builds in my new upadte, since they are already tagged (even if not part of any update)? -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 10/08/15 12:24 -0600, Kevin Fenzi wrote: On Mon, 10 Aug 2015 17:51:30 +0100 José Matos wrote: On Saturday 18 July 2015 12:46:51 Jonathan Wakely wrote: > Hi, > > I've pushed a new version of Boost, 1.58.0, to rawhide and f23, > which will require all packages that depend on Boost to be rebuilt. > The plan was to update to 1.59.0 but that isn't going to be > released in time for the F23 schedule, so I'll update rawhide to > 1.59.0 at a later date. > > For the F23 rebuilds please use the tag f23-boost e.g. > > fedpkg build --target f23-boost Hi Jonathan, what is the status of this update for Fedora 23? Last I heard there were bodhi issues. It was unable to handle an update with that many packages. ;( Right, it has proved impossible to create the update. So, I think it needs to just be split up into several updates, or perhaps Luke can come up with some way to get it processed. ;( If splitting it up is acceptable I can do that, but I think there may be another problem: when I was trying to add builds to the update they were not added correctly, but did get tagged in koji as if added to the update. So although I deleted the original (unsuccessful) update, now I can't add lots of the builds to a new update, because bodhi says there is already an update for that build. It looks like bodhi doesn't work for large updates, and corrupts the DB so that updates and koji tags get out of sync :-( -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Mon, 10 Aug 2015 17:51:30 +0100 José Matos wrote: > On Saturday 18 July 2015 12:46:51 Jonathan Wakely wrote: > > Hi, > > > > I've pushed a new version of Boost, 1.58.0, to rawhide and f23, > > which will require all packages that depend on Boost to be rebuilt. > > The plan was to update to 1.59.0 but that isn't going to be > > released in time for the F23 schedule, so I'll update rawhide to > > 1.59.0 at a later date. > > > > For the F23 rebuilds please use the tag f23-boost e.g. > > > > fedpkg build --target f23-boost > > Hi Jonathan, > what is the status of this update for Fedora 23? Last I heard there were bodhi issues. It was unable to handle an update with that many packages. ;( So, I think it needs to just be split up into several updates, or perhaps Luke can come up with some way to get it processed. ;( kevin pgpRqUK1dDsK9.pgp Description: OpenPGP digital signature -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Saturday 18 July 2015 12:46:51 Jonathan Wakely wrote: > Hi, > > I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which > will require all packages that depend on Boost to be rebuilt. The plan > was to update to 1.59.0 but that isn't going to be released in time > for the F23 schedule, so I'll update rawhide to 1.59.0 at a later > date. > > For the F23 rebuilds please use the tag f23-boost e.g. > > fedpkg build --target f23-boost Hi Jonathan, what is the status of this update for Fedora 23? Regards, -- José Abílio -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thursday, 23 July 2015 at 19:21, Jonathan Wakely wrote: > On 23/07/15 14:27 +0200, David Tardon wrote: > >Hi, > > > >On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > >>Any problems rebuilding either open a bug or feel free to email me or > >>ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. > > > >This is a work-in-progress list of FTBFS packages: > > > >* Build failures: > > > >- F23 + Rawhide > > > >mkvtoolnix > > For mkvtoolnix I get this error in rawhide: > > In file included from ./src/mkvtoolnix-gui/util/files_drag_drop_widget.h:6:0, > from src/mkvtoolnix-gui/util/files_drag_drop_widget.moc:9: > src/mkvtoolnix-gui/util/files_drag_drop_handler.h:12:44: error: expected > class-name before '{' token > class FilesDragDropHandler: public QObject { >^ > > That seems to be another missing header (as with rhbz1234405) not a > Boost problem. This is fixed upstream. I've updated to 8.2.0 for f23+. Regards, Dominik -- Fedora http://fedoraproject.org/wiki/User:Rathann RPMFusion http://rpmfusion.org "Faith manages." -- Delenn to Lennier in Babylon 5:"Confessions and Lamentations" -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 31/07/15 14:49 +0100, Richard W.M. Jones wrote: Ceph failed to build with some impenetrable C++ error: In file included from /usr/include/boost/optional/optional.hpp:28:0, from /usr/include/boost/optional/optional_io.hpp:19, from ./include/encoding.h:289, from ./include/uuid.h:8, from ./include/types.h:21, from mon/OSDMonitor.h:28, from mon/OSDMonitor.cc:21: /usr/include/boost/variant/get.hpp: In instantiation of 'typename boost::add_reference::type boost::strict_get(const boost::variant&) [with U = int; T0 = std::__cxx11::basic_string; T1 = bool; T2 = long long int; T3 = double; T4 = std::vector >; T5 = boost::detail::variant::void_; T6 = boost::detail::variant::void_; T7 = boost::detail::variant::void_; T8 = boost::detail::variant::void_; T9 = boost::detail::variant::void_; T10 = boost::detail::variant::void_; T11 = boost::detail::variant::void_; T12 = boost::detail::variant::void_; T13 = boost::detail::variant::void_; T14 = boost::detail::variant::void_; T15 = boost::detail::variant::void_; T16 = boost::detail::variant::void_; T17 = boost::detail::variant::void_; T18 = boost::detail::variant::void_; T19 = boost::detail::variant::void_; typename boost::add_reference::type = const int&]': /usr/include/boost/variant/get.hpp:299:25: required from 'typename boost::add_reference::type boost::get(const boost::variant&) [with U = int; T0 = std::__cxx11::basic_string; T1 = bool; T2 = long long int; T3 = double; T4 = std::vector >; T5 = boost::detail::variant::void_; T6 = boost::detail::variant::void_; T7 = boost::detail::variant::void_; T8 = boost::detail::variant::void_; T9 = boost::detail::variant::void_; T10 = boost::detail::variant::void_; T11 = boost::detail::variant::void_; T12 = boost::detail::variant::void_; T13 = boost::detail::variant::void_; T14 = boost::detail::variant::void_; T15 = boost::detail::variant::void_; T16 = boost::detail::variant::void_; T17 = boost::detail::variant::void_; T18 = boost::detail::variant::void_; T19 = boost::detail::variant::void_; typename boost::add_reference::type = const int&]' ./common/cmdparse.h:47:26: required from 'bool cmd_getval(CephContext*, const cmdmap_t&, std::__cxx11::string, T&) [with T = int; cmdmap_t = std::map, boost::variant, bool, long long int, double, std::vector > > >; std::__cxx11::string = std::__cxx11::basic_string]' mon/OSDMonitor.cc:3002:54: required from here /usr/include/boost/variant/get.hpp:195:5: error: invalid application of 'sizeof' to incomplete type 'boost::STATIC_ASSERTION_FAILURE' BOOST_STATIC_ASSERT_MSG( ^ Makefile:15876: recipe for target 'mon/OSDMonitor.lo' failed -- Any ideas on that one? This blocks qemu and all the rest of the virt stack. It's a static assertion failure. Line 195 in boost/variant/get.hpp is BOOST_STATIC_ASSERT_MSG( (boost::detail::variant::holds_element, const U >::value), "boost::variant does not contain specified type U, " "call to boost::get(const boost::variant*) will always return NULL" ); Which is pretty descriptive. It's the same problem as described at https://lists.fedoraproject.org/pipermail/devel/2015-July/212789.html i.e. caused by the breaking change to Boost.Variant, which can be fixed by changing the source or defining a macro to use relaxed_get() intstead of strict_get(). -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
-BEGIN PGP SIGNED MESSAGE- Hash: SHA256 On 07/29/2015 12:09 AM, Adam Williamson wrote: Mukundan, > can you please decide whether you want to: > > i) patch out the #error and keep building against wxGTK 3 against > upstream's advice ii) revert to building against wxGTK 2 > > and then make the necessary changes and submit builds for the > f23-boost and f24-boost targets (as explained in the first post in > this thread), please? Thanks! > Slowly catching up after vacation - Thanks Adam for the troubleshooting work. I had started work on fityk 1.3.0 before my vacation but could not finish and rebuilds all showed it to be broken. I have not built it against compat- pkgs and f24 and f23 should be good to go. After testing a bit, I will submit it for update. - -- GPG key: 00E8658D "The heights by great men reached and kept were not attainted by sudden flight, but they while their companions slept, were toiling upward in the night" -- Henry Wadsworth Longfellow -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQIcBAEBCAAGBQJVwYD9AAoJEIfjPv0A6GWN5t8QAKOIdj74N65iy4Zo73MZ5/PZ i12FrLoeQnbnSBlTV04j5D/FQdSz40jUkSbDFWrjozUhpJuxVTpsNIXLPUetwJFI qg76b90DC96c1bkUPa4wev2pOOhqR/GCJWZqVVLukjReiAb6OSQb0/XJp138Hz82 LdSB8y+ZysWTB/hytVPbL9lxe5G1sNk0yXKS087yLUw85uI3vzvdWEfdpv1E0/hM JSw1bIipUqF5gghqLOaH62SgQ69fbLW7GZGAJQ+wZ5CEwHAitMQc0/Y7UqfwvIk7 JukPl+yLhBPmiqOQQ5SkmdkhTqK28cCD1KdfE/ivP6ygdRgfTA/6jSWoPKzrft0q Sl8XyUj5ev9iWwlo7JRMB0cqCLOA3wK8ahqwKh9a5RteokxhQPmgvBZA9iWzI1Ki WzMHMO2uhSC191RbVTj6CzfmI9Q01a8KWT/7WqSTTN+DvBFR4oaaSVjDQXoIhYpk 846X8NEyOqx5bfn0aRBs+3UD00OO9k/S489nHVbTEsSXv/5vZdBnCERsjvqJ1DDp WTLVW6Oq6zSBgEN8gXhPkjacv48aNYqba1yTjyYimfyKXumiz47WlGOhuGvTdEu/ xAjLotcqjvNiUTuF/xbAZQGrOzK+mv8Mys/OFxpaGe85od9PRolqhEwsKs/OZcLP mAqeY67BHNVZNiHbSbGO =qIi+ -END PGP SIGNATURE- -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, Jul 31, 2015 at 02:49:26PM +0100, Richard W.M. Jones wrote: > Any ideas on that one? This blocks qemu and all the rest of the virt > stack. Don't worry - I found the upstream bug with a fix: https://github.com/ceph/ceph/pull/4616 I'll try that. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-builder quickly builds VMs from scratch http://libguestfs.org/virt-builder.1.html -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Ceph failed to build with some impenetrable C++ error: In file included from /usr/include/boost/optional/optional.hpp:28:0, from /usr/include/boost/optional/optional_io.hpp:19, from ./include/encoding.h:289, from ./include/uuid.h:8, from ./include/types.h:21, from mon/OSDMonitor.h:28, from mon/OSDMonitor.cc:21: /usr/include/boost/variant/get.hpp: In instantiation of 'typename boost::add_reference::type boost::strict_get(const boost::variant&) [with U = int; T0 = std::__cxx11::basic_string; T1 = bool; T2 = long long int; T3 = double; T4 = std::vector >; T5 = boost::detail::variant::void_; T6 = boost::detail::variant::void_; T7 = boost::detail::variant::void_; T8 = boost::detail::variant::void_; T9 = boost::detail::variant::void_; T10 = boost::detail::variant::void_; T11 = boost::detail::variant::void_; T12 = boost::detail::variant::void_; T13 = boost::detail::variant::void_; T14 = boost::detail::variant::void_; T15 = boost::detail::variant::void_; T16 = boost::detail::variant::void_; T17 = boost::detail::variant::void_; T18 = boost::detail::variant::void_; T19 = boost::detail::variant::void_; typename boost::add_reference::type = const int&]': /usr/include/boost/variant/get.hpp:299:25: required from 'typename boost::add_reference::type boost::get(const boost::variant&) [with U = int; T0 = std::__cxx11::basic_string; T1 = bool; T2 = long long int; T3 = double; T4 = std::vector >; T5 = boost::detail::variant::void_; T6 = boost::detail::variant::void_; T7 = boost::detail::variant::void_; T8 = boost::detail::variant::void_; T9 = boost::detail::variant::void_; T10 = boost::detail::variant::void_; T11 = boost::detail::variant::void_; T12 = boost::detail::variant::void_; T13 = boost::detail::variant::void_; T14 = boost::detail::variant::void_; T15 = boost::detail::variant::void_; T16 = boost::detail::variant::void_; T17 = boost::detail::variant::void_; T18 = boost::detail::variant::void_; T19 = boost::detail::variant::void_; typename boost::add_reference::type = const int&]' ./common/cmdparse.h:47:26: required from 'bool cmd_getval(CephContext*, const cmdmap_t&, std::__cxx11::string, T&) [with T = int; cmdmap_t = std::map, boost::variant, bool, long long int, double, std::vector > > >; std::__cxx11::string = std::__cxx11::basic_string]' mon/OSDMonitor.cc:3002:54: required from here /usr/include/boost/variant/get.hpp:195:5: error: invalid application of 'sizeof' to incomplete type 'boost::STATIC_ASSERTION_FAILURE' BOOST_STATIC_ASSERT_MSG( ^ Makefile:15876: recipe for target 'mon/OSDMonitor.lo' failed -- Any ideas on that one? This blocks qemu and all the rest of the virt stack. Rich. -- Richard Jones, Virtualization Group, Red Hat http://people.redhat.com/~rjones Read my programming and virtualization blog: http://rwmj.wordpress.com virt-top is 'top' for virtual machines. Tiny program with many powerful monitoring features, net stats, disk stats, logging, etc. http://people.redhat.com/~rjones/virt-top -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, 2015-07-24 at 19:26 +0100, Jonathan Wakely wrote: > > > scantailor > > This has some weird CMake error about being unable to figure out how > to use -pthread. This one boiled down to crappy upstream build scripts which disregard the provided CFLAGS and CXXFLAGS values: https://github.com/scantailor/scantailor/pull/160 > > > xmlcopyeditor > > This fails with a "recompile with -fPIC" error that doesn't seem > related to Boost. I think the failure in this case was much the same as scantailor's, but there's a much newer upstream release which compiles successfully and works fine (they fixed their build scripts, it looks like), so I'm just going to bump to that. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, 2015-07-24 at 19:26 +0100, Jonathan Wakely wrote: > >> fityk > > Not Boost related: > > app.cpp:12:2: error: #error "Not everything is working with wxGTK3. > Use default wxGTK instead, " "based on GTK+2. If you want to test it, > just remove this #error." So this one does exactly what it says on the tin, basically. nonamedotc changed fityk to build against wxGTK 3 when bumping it to 1.2.9. In 1.3.0 this warning/error was added upstream: https://github.com/wojdyr/fityk/commit/365cc0f8748d36788f46a81461c717d8 27363e89 to me that strongly suggests we should revert and build fityk against wxGTK 2, but I didn't want to just go ahead and do that, I thought it'd be best to see what nonamedotc wants to do. Mukundan, can you please decide whether you want to: i) patch out the #error and keep building against wxGTK 3 against upstream's advice ii) revert to building against wxGTK 2 and then make the necessary changes and submit builds for the f23-boost and f24-boost targets (as explained in the first post in this thread), please? Thanks! -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, 2015-07-24 at 19:26 +0100, Jonathan Wakely wrote: > >> condor > > Another BuildRequires failure with libpoppler.so In fact you managed to get an attempt through which failed in compilation: https://kojipkgs.fedoraproject.org//work/tasks/8134/10448134/build.log I've been figuring that one out and marvelling at the insanity that is this 'globus' project, which has a website with more links than wikipedia but doesn't manage to link to their SCM anywhere I could spot, which I just found via google ( https://github.com/globus/globus-toolkit - I mean, for pete's sake, if you're on github just freaking well say so). They also appear to have shipped an update to globus-gsi-credential which changes its interfaces (which is what's causing this problem) without documenting its *existence* at all - show me where you see globus_gsi_credential-7.9 onhttp://toolkit.globus.org/toolkit/advisories.html ? Anyone? Nope. Oh well, I'll deal with it. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Tue, 2015-07-28 at 14:49 +0100, Jonathan Wakely wrote: > On 28/07/15 14:48 +0100, Jonathan Wakely wrote: > > On 27/07/15 19:35 -0700, Adam Williamson wrote: > > > On Mon, 2015-07-27 at 13:31 +0100, Jonathan Wakely wrote: > > > > On 23/07/15 14:33 +0200, David Tardon wrote: > > > > > Hi, > > > > > > > > > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > > > > > Hi, > > > > > > > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely > > > > > > wrote: > > > > > > > Any problems rebuilding either open a bug or feel free to > > > > > > > email > > > > > > > me or > > > > > > > ping me on IRC (my freenode nick is 'redi') and I'll be > > > > > > > happy > > > > > > > to help. > > > > > > > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > > > > > > > * Build failures: > > > > > > > > > > > > - F23 + Rawhide > > > > > > > > > > > > ledger > > > > > > > > This patch is needed to make Ledger compile with Boost 1.58.0 > > > > > > I submitted a build using the patch you submitted as an upstream > > > PR - > > > https://github.com/ledger/ledger/pull/422 - but it still failed: > > > > > > https://kojipkgs.fedoraproject.org//work/tasks/9527/10499527/buil > > > d.log > > > > Yes, as well as my patch for #422 you still need the second half of > > the patch for the upstream #417 (which was not reverted), which > > I've > > attached to this mail. > > Oh, and then it *still* fails because of unpackaged files, see > https://kojipkgs.fedoraproject.org//work/tasks/1167/10501167/build.lo > g > which I did yesterday using the constness patch from my previous mail > and my upstream fix for #422. For the record - with many thanks to nirik - this appears to be due to some shenanigans with RPM's handling of %doc: https://bugzilla.redhat.com/show_bug.cgi?id=728959 in Ye Olde Olde Days, %doc ANY_RELATIVE_PATH would wipe %_pkgdocdir before installing the listed files to it - so if you wanted to keep stuff installed there during %install you couldn't use %doc in that way, you had to do it yourself. In Ye Somewhat Less Olde Days - between 2011 and four days ago - %doc didn't do that any more, and would happily include any files that were put in %_pkgdocdir during %install, without them needing to be listed in %files. For Rawhide, since four days ago (downstream - the change, I think, arrived in rpm-4.12.90-1.fc24; it landed on upstream master in April), we have a sort of happy medium apparently, where %doc does not clean %_pkgdocdir but the 'unpackaged files' check does apply to it - so if %install puts files there, you either have to delete them or list them in %files. And that's what broke the build here; the package's %install installs various things to %_pkgdocdir , but the %files section doesn't list them. I'll fix that. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Mon, 2015-07-27 at 17:35 +0100, Jonathan Wakely wrote: > On 23/07/15 14:33 +0200, David Tardon wrote: > > Hi, > > > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > > Hi, > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > > Any problems rebuilding either open a bug or feel free to email > > > > me or > > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > > to help. > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > * Build failures: > > > > > > - Rawhide only > > > > > > pion-net > > The attached patch is needed to fix pion-net's %check step, due to a > breaking change in Boost.Variant in 1.58.0, documented at > http://www.boost.org/users/history/version_1_58_0.html and > http://www.boost.org/doc/libs/1_58_0/doc/html/boost/get_idp295310448. > html > > Also I noticed that the URL in the pion-net.spec file no longer > works, > I guess the project was hosted at Google Code which is gone now. The > new home seems to be https://github.com/splunk/pion although that > doesn't have any releases before 5.0.1 (Fedora uses 4.0.9). Again, it seems absurd to keep reanimating the zombified corpse of an ancient release, so instead I bumped this to 5.0.7. It should really be renamed pion, but I'm damned if I'm doing a rename review request. I have builds going through now. For the record, both this and roboptim don't seem to have any other packages in the repos that require them, so nothing else should need rebuilding. I did check, and I'm pretty sure neither of this was stuck on an old version for some genuine reason, they both just seem to have become neglected by their maintainers. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 28/07/15 14:48 +0100, Jonathan Wakely wrote: On 27/07/15 19:35 -0700, Adam Williamson wrote: On Mon, 2015-07-27 at 13:31 +0100, Jonathan Wakely wrote: On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > Hi, > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > Any problems rebuilding either open a bug or feel free to email > > me or > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > to help. > > This is a work-in-progress list of FTBFS packages: > > * Build failures: > > - F23 + Rawhide > > ledger This patch is needed to make Ledger compile with Boost 1.58.0 I submitted a build using the patch you submitted as an upstream PR - https://github.com/ledger/ledger/pull/422 - but it still failed: https://kojipkgs.fedoraproject.org//work/tasks/9527/10499527/build.log Yes, as well as my patch for #422 you still need the second half of the patch for the upstream #417 (which was not reverted), which I've attached to this mail. Oh, and then it *still* fails because of unpackaged files, see https://kojipkgs.fedoraproject.org//work/tasks/1167/10501167/build.log which I did yesterday using the constness patch from my previous mail and my upstream fix for #422. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 27/07/15 19:35 -0700, Adam Williamson wrote: On Mon, 2015-07-27 at 13:31 +0100, Jonathan Wakely wrote: On 23/07/15 14:33 +0200, David Tardon wrote: > Hi, > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > Hi, > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > Any problems rebuilding either open a bug or feel free to email > > > me or > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > to help. > > > > This is a work-in-progress list of FTBFS packages: > > > > * Build failures: > > > > - F23 + Rawhide > > > > ledger This patch is needed to make Ledger compile with Boost 1.58.0 I submitted a build using the patch you submitted as an upstream PR - https://github.com/ledger/ledger/pull/422 - but it still failed: https://kojipkgs.fedoraproject.org//work/tasks/9527/10499527/build.log Yes, as well as my patch for #422 you still need the second half of the patch for the upstream #417 (which was not reverted), which I've attached to this mail. Thanks for doing these fixes. >From 68c9d649caa2c7c7f222613efe86576c379a5a7a Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?Johann=20Kl=C3=A4hn?= Date: Sun, 10 May 2015 13:41:26 +0200 Subject: [PATCH 2/2] fix build for boost 1.58 --- src/filters.cc | 2 +- src/iterators.cc | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-) diff --git a/src/filters.cc b/src/filters.cc index 2f97a0e..b6530c0 100644 --- a/src/filters.cc +++ b/src/filters.cc @@ -707,7 +707,7 @@ namespace { insert_prices_in_map(price_map_t& _all_prices) : all_prices(_all_prices) {} -void operator()(datetime_t& date, const amount_t& price) { +void operator()(const datetime_t& date, const amount_t& price) { all_prices.insert(price_map_t::value_type(date, price)); } }; diff --git a/src/iterators.cc b/src/iterators.cc index 21bec5d..0225e21 100644 --- a/src/iterators.cc +++ b/src/iterators.cc @@ -96,7 +96,7 @@ namespace { TRACE_DTOR(create_price_xact); } -void operator()(datetime_t& date, const amount_t& price) { +void operator()(const datetime_t& date, const amount_t& price) { xact_t * xact; string symbol = price.commodity().symbol(); -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Mon, 2015-07-27 at 13:33 +0100, Jonathan Wakely wrote: > On 23/07/15 14:33 +0200, David Tardon wrote: > > Hi, > > > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > > Hi, > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > > Any problems rebuilding either open a bug or feel free to email > > > > me or > > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > > to help. > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > * Build failures: > > > > > > - F23 + Rawhide > > > > > > roboptim-core > > This is needed to make the %configure step work with GCC-5 Well, there's that, and also there's the fact that the roboptim we're shipping is five years old and there've been like four major releases since then. So instead of keeping an ancient roboptim on life support, I figured let's just go with the latest version...and spent all afternoon dealing with build issues in 3.1. Still, it's actually built now (well, still waiting on ARM to finish up): http://koji.fedoraproject.org/koji/taskinfo?taskID=10503753 (I'm sending an f23-boost build now). there are a couple of issues shown up by the tests on i686 if anyone's at all interested in looking into those: https://github.com/roboptim/roboptim-core/issues/102 I'll update the sole plugin we have packaged in a bit. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Mon, 2015-07-27 at 13:31 +0100, Jonathan Wakely wrote: > On 23/07/15 14:33 +0200, David Tardon wrote: > > Hi, > > > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > > Hi, > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > > Any problems rebuilding either open a bug or feel free to email > > > > me or > > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > > to help. > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > * Build failures: > > > > > > - F23 + Rawhide > > > > > > ledger > > This patch is needed to make Ledger compile with Boost 1.58.0 I submitted a build using the patch you submitted as an upstream PR - https://github.com/ledger/ledger/pull/422 - but it still failed: https://kojipkgs.fedoraproject.org//work/tasks/9527/10499527/build.log In file included from /usr/include/boost/function/detail/maybe_include.hpp:23:0, from /usr/include/boost/function/detail/function_iterate.hpp:14, from /usr/include/boost/preprocessor/iteration/detail/iter/forward1.hpp:57, from /usr/include/boost/function.hpp:64, from /usr/include/boost/algorithm/string/detail/find_iterator.hpp:18, from /usr/include/boost/algorithm/string/find_iterator.hpp:24, from /usr/include/boost/algorithm/string/iter_find.hpp:27, from /usr/include/boost/algorithm/string/split.hpp:16, from /usr/include/boost/algorithm/string.hpp:23, from /builddir/build/BUILD/ledger-720c03b139d251c53f9deef491f5953e2fdb97ca/system.hh:173: /usr/include/boost/function/function_template.hpp: In instantiation of 'static void boost::detail::function::void_function_obj_invoker2::invoke(boost::detail::function::function_buffer&, T0, T1) [with FunctionObj = ledger::{anonymous}::insert_prices_in_map; R = void; T0 = boost::posix_time::ptime; T1 = const ledger::amount_t&]': /usr/include/boost/function/function_template.hpp:940:38: required from 'void boost::function2::assign_to(Functor) [with Functor = ledger::{anonymous}::insert_prices_in_map; R = void; T0 = boost::posix_time::ptime; T1 = const ledger::amount_t&]' /usr/include/boost/function/function_template.hpp:728:7: required from 'boost::function2::function2(Functor, typename boost::enable_if_c::value>::value, int>::type) [with Functor = ledger::{anonymous}::insert_prices_in_map; R = void; T0 = boost::posix_time::ptime; T1 = const ledger::amount_t&; typename boost::enable_if_c::value>::value, int>::type = int]' /usr/include/boost/function/function_template.hpp:1077:16: required from 'boost::function::function(Functor, typename boost::enable_if_c::value>::value, int>::type) [with Functor = ledger::{anonymous}::insert_prices_in_map; R = void; T0 = boost::posix_time::ptime; T1 = const ledger::amount_t&; typename boost::enable_if_c::value>::value, int>::type = int]' /builddir/build/BUILD/ledger-720c03b139d251c53f9deef491f5953e2fdb97ca/src/filters.cc:788:69: required from here /usr/include/boost/function/function_template.hpp:159:11: error: no match for call to '(ledger::{anonymous}::insert_prices_in_map) (boost::posix_time::ptime, const ledger::amount_t&)' BOOST_FUNCTION_RETURN((*f)(BOOST_FUNCTION_ARGS)); ^ /builddir/build/BUILD/ledger-720c03b139d251c53f9deef491f5953e2fdb97ca/src/filters.cc:710:10: note: candidate: void ledger::{anonymous}::insert_prices_in_map::operator()(ledger::datetime_t&, const ledger::amount_t&) void operator()(datetime_t& date, const amount_t& price) { ^ /builddir/build/BUILD/ledger-720c03b139d251c53f9deef491f5953e2fdb97ca/src/filters.cc:710:10: note: conversion of argument 1 would be ill-formed: In file included from /usr/include/boost/function/detail/maybe_include.hpp:23:0, from /usr/include/boost/function/detail/function_iterate.hpp:14, from /usr/include/boost/preprocessor/iteration/detail/iter/forward1.hpp:57, from /usr/include/boost/function.hpp:64, from /usr/include/boost/algorithm/string/detail/find_iterator.hpp:18, from /usr/include/boost/algorithm/string/find_iterator.hpp:24, from /usr/include/boost/algorithm/string/iter_find.hpp:27, from /usr/include/boost/algorithm/string/split.hpp:16, from /usr/include/boost/algorithm/string.hpp:23, from /builddir/build/BUILD/ledger-720c03b139d251c53f9deef491f5953e2fdb97ca/system.hh:173: /usr/include/boost/function/function_template.hpp:159:11: error: invalid initialization of non-const reference of type 'ledger::datetime_t& {aka boost::posix_time::ptime&}' from an rvalue of type 'boost::posix_time::ptime' BOOST_FUNCTION_RETURN((*f)(BOOST_FUNCTION_ARGS)); ^ -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | X
Re: Boost updated to 1.58.0 in rawhide and f23
On Mon, 2015-07-27 at 09:48 +0100, Jonathan Wakely wrote: > On 24/07/15 19:26 +0100, Jonathan Wakely wrote: > > > > flamerobin > > > > This looks like a Boost problem: > > > > configure: error: invalid value: boost_major_version= > > > > But actually it's caused by a GCC change that puts a # preprocessor > > marker before macro expansions: > > > > boost-lib-version = # 2 "/tmp/conftest.cpp" 3 4 > > "1_58" > > > > The configure script needs to use -P as described at > > https://gcc.gnu.org/gcc-5/porting_to.html > > This patch fixes the flamerobin build. > > I don't understand why flamerobin isn't shown in the recent FTBFS > email when this failure is nothing to do with the Boost upgrade and > should be happening for all f23 and rawhide builds using GCC-5. boost.m4 upstream fixed this differently: https://github.com/tsuna/boost.m4/commit/32553aaf4d5090da19aa0ec33b9369 82c685009f the test comes into flamerobin from boost.m4 -> aclocal.m4 -> configure . I've filed a bug on upstream flamerobin to update to latest boost.m4: https://sourceforge.net/p/flamerobin/bugs/256/ I was gonna change the fix in our flamerobin to 'pull in the latest boost.m4 and regenerate the build scripts', but the snapshot we have is so ancient that doesn't work very well, so I'll stick with more or less the current patch (only make it apply properly). -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - Rawhide only pion-net The attached patch is needed to fix pion-net's %check step, due to a breaking change in Boost.Variant in 1.58.0, documented at http://www.boost.org/users/history/version_1_58_0.html and http://www.boost.org/doc/libs/1_58_0/doc/html/boost/get_idp295310448.html Also I noticed that the URL in the pion-net.spec file no longer works, I guess the project was hosted at Google Code which is gone now. The new home seems to be https://github.com/splunk/pion although that doesn't have any releases before 5.0.1 (Fedora uses 4.0.9). commit 935dcf35040883dadf9e8d583946d30ce644ef49 Author: Jonathan Wakely Date: Mon Jul 27 16:47:43 2015 +0100 Add patch to work with strict_get from Boost.Variant 1.58.0. diff --git a/pion-net-boost-variant.patch b/pion-net-boost-variant.patch new file mode 100644 index 000..6638244 --- /dev/null +++ b/pion-net-boost-variant.patch @@ -0,0 +1,24 @@ +Adapt to breaking change in Boost.Variant where boost::get<> is strict. + +Author: Jonathan Wakely + +--- a/common/tests/PionBlobTests.cpp~ 2015-07-27 16:41:47.604056525 +0100 b/common/tests/PionBlobTests.cpp 2015-07-27 16:45:59.882935301 +0100 +@@ -114,7 +114,7 @@ + BOOST_CHECK(b1 == goodbye_str); + + boost::variant valueA(b1); +- BOOST_CHECK_EQUAL(goodbye_str, boost::get(valueA).get()); ++ BOOST_CHECK_EQUAL(goodbye_str, boost::get(valueA).get()); + BOOST_CHECK(! b1.unique()); + BOOST_CHECK_EQUAL(b1.use_count(), 2); + +@@ -123,7 +123,7 @@ + BOOST_CHECK_EQUAL(b1.use_count(), 1); + + boost::variant valueB(p2); +- BOOST_CHECK_EQUAL(goodbye_str, boost::get(valueB).get()); ++ BOOST_CHECK_EQUAL(goodbye_str, boost::get(valueB).get()); + + } + diff --git a/pion-net.spec b/pion-net.spec index 33e92fa..c13813e 100644 --- a/pion-net.spec +++ b/pion-net.spec @@ -2,7 +2,7 @@ Name: pion-net Version: 4.0.9 -Release: 19%{?dist} +Release: 20%{?dist} Summary: C++ library for building lightweight HTTP interfaces License: Boost @@ -19,6 +19,7 @@ Patch3: pion-net-boost-linking.patch Patch4:pion-net-log4cpp-headers.patch Patch5:pion-net-boost-compatibility.patch Patch6:pion-net-boost-time-utc.patch +Patch7:pion-net-boost-variant.patch %description Pion Network Library is a C++ framework for building lightweight HTTP @@ -114,6 +115,9 @@ make check %changelog +* Mon Jul 27 2015 Jonathan Wakely 4.0.9-20 +- Add patch to work with strict_get from Boost.Variant 1.58.0. + * Wed Jul 22 2015 David Tardon - 4.0.9-19 - rebuild for Boost 1.58 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide roboptim-core This is needed to make the %configure step work with GCC-5 commit 44966bae9369d4e2caf1ae778dd5a64739667ee6 Author: Jonathan Wakely Date: Mon Jul 27 11:14:37 2015 +0100 Use -P flag during configure tests. diff --git a/roboptim-core.spec b/roboptim-core.spec index ba84ee1..e904cd9 100644 --- a/roboptim-core.spec +++ b/roboptim-core.spec @@ -1,6 +1,6 @@ Name: roboptim-core Version: 0.5 -Release: 16%{?dist} +Release: 17%{?dist} Summary: The RobOptim core C++ library Group: Development/Libraries @@ -54,7 +54,7 @@ Documentation for the RobOptim core library touch -r aclocal.m4 configure configure.ac %build -%configure --docdir=%{_pkgdocdir} --disable-static +%configure --docdir=%{_pkgdocdir} --disable-static CPPFLAGS=-P sed -i 's|^hardcode_libdir_flag_spec=.*|hardcode_libdir_flag_spec=""|g' libtool sed -i 's|^runpath_var=LD_RUN_PATH|runpath_var=DIE_RPATH_DIE|g' libtool @@ -107,6 +107,9 @@ rm -rf $RPM_BUILD_ROOT %postun -p /sbin/ldconfig %changelog +* Mon Jul 27 2015 Jonathan Wakely 0.5-17 +- Use -P flag during configure tests. + * Wed Jul 22 2015 David Tardon - 0.5-16 - rebuild for Boost 1.58 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide ledger This patch is needed to make Ledger compile with Boost 1.58.0 commit ff32007cc80c489f516bc6ec03009e9b9c56017b Author: Jonathan Wakely Date: Mon Jul 27 12:23:02 2015 +0100 Add patch to fix build failure with Boost 1.58.0. diff --git a/ledger-3.1-0010-boost-optional-none.patch b/ledger-3.1-0010-boost-optional-none.patch new file mode 100644 index 000..1549827 --- /dev/null +++ b/ledger-3.1-0010-boost-optional-none.patch @@ -0,0 +1,133 @@ +From 48aec0f093ff6494a3e4f7cd5166cb4a27c16814 Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Johann=20Kl=C3=A4hn?= +Date: Sun, 10 May 2015 12:45:28 +0200 +Subject: [PATCH 1/2] boost::none_t no longer convertible from literal 0 in + 1.58 + +Instead we use boost::none, which has been documented since boost 1.34.0. +--- + src/account.h | 6 +- + src/item.h| 6 +- + src/parser.h | 2 +- + src/post.h| 6 +- + src/times.h | 6 +- + 5 files changed, 5 insertions(+), 21 deletions(-) + +diff --git a/src/account.h b/src/account.h +index 7fae93e..8d0fb1d 100644 +--- a/src/account.h b/src/account.h +@@ -261,11 +261,7 @@ class account_t : public supports_flags<>, public scope_t + mutable optional xdata_; + + bool has_xdata() const { +-#if BOOST_VERSION >= 105600 +-return xdata_ != NULL; +-#else +-return xdata_; +-#endif ++return xdata_ != boost::none; + } + void clear_xdata(); + xdata_t& xdata() { +diff --git a/src/item.h b/src/item.h +index ba81217..2c349bd 100644 +--- a/src/item.h b/src/item.h +@@ -174,11 +174,7 @@ class item_t : public supports_flags, public scope_t + static bool use_aux_date; + + virtual bool has_date() const { +-#if BOOST_VERSION >= 105600 +-return _date != NULL; +-#else +-return _date; +-#endif ++return _date != boost::none; + } + + virtual date_t date() const { +diff --git a/src/parser.h b/src/parser.h +index e46fc71..25c4a7e 100644 +--- a/src/parser.h b/src/parser.h +@@ -118,7 +118,7 @@ class expr_t::parser_t : public noncopyable + + ptr_op_t parse(std::istream& in, + const parse_flags_t&flags = PARSE_DEFAULT, +- const optional& original_string = NULL); ++ const optional& original_string = boost::none); + }; + + } // namespace ledger +diff --git a/src/post.h b/src/post.h +index 0fb45e9..3fa67e5 100644 +--- a/src/post.h b/src/post.h +@@ -205,11 +205,7 @@ class post_t : public item_t + mutable optional xdata_; + + bool has_xdata() const { +-#if BOOST_VERSION >= 105600 +-return xdata_ != NULL; +-#else +-return xdata_; +-#endif ++return xdata_ != boost::none; + } + void clear_xdata() { + xdata_ = none; +diff --git a/src/times.h b/src/times.h +index 421d146..cc98085 100644 +--- a/src/times.h b/src/times.h +@@ -500,11 +500,7 @@ class date_interval_t : public equality_comparable + void stabilize(const optional& date = none); + + bool is_valid() const { +-#if BOOST_VERSION >= 105600 +-return start != NULL; +-#else +-return start; +-#endif ++return start != boost::none; + } + + /** Find the current or next period containing date. Returns false if + +From 68c9d649caa2c7c7f222613efe86576c379a5a7a Mon Sep 17 00:00:00 2001 +From: =?UTF-8?q?Johann=20Kl=C3=A4hn?= +Date: Sun, 10 May 2015 13:41:26 +0200 +Subject: [PATCH 2/2] fix build for boost 1.58 + +--- + src/filters.cc | 2 +- + src/iterators.cc | 2 +- + 2 files changed, 2 insertions(+), 2 deletions(-) + +diff --git a/src/filters.cc b/src/filters.cc +index 2f97a0e..b6530c0 100644 +--- a/src/filters.cc b/src/filters.cc +@@ -707,7 +707,7 @@ namespace { + insert_prices_in_map(price_map_t& _all_prices) + : all_prices(_all_prices) {} + +-void operator()(datetime_t& date, const amount_t& price) { ++void operator()(const datetime_t& date, const amount_t& price) { + all_prices.insert(price_map_t::value_type(date, price)); + } + }; +diff --git a/src/iterators.cc b/src/iterators.cc +index 21bec5d..0225e21 100644 +--- a/src/iterators.cc b/src/iterators.cc +@@ -96,7 +96,7 @@ namespace { + TRACE_DTOR(create_price_xact); + } + +-void operator()(datetime_t& date, const amount_t& price) { ++void operator()(const datetime_t& date, const amount_t& price) { + xact_t * xact; + string symbol = price.commodity().symbol(); + diff --git a/ledger.spec b/ledger.spec index 64be657..a3ab0ec 100644 --- a/ledger.spec +++ b/ledger.spec @@ -4,7 +4,7 @@ Name: ledger Version: 3.1 -Release: 7%{?dist}
Re: Boost updated to 1.58.0 in rawhide and f23
On 24/07/15 19:26 +0100, Jonathan Wakely wrote: flamerobin This looks like a Boost problem: configure: error: invalid value: boost_major_version= But actually it's caused by a GCC change that puts a # preprocessor marker before macro expansions: boost-lib-version = # 2 "/tmp/conftest.cpp" 3 4 "1_58" The configure script needs to use -P as described at https://gcc.gnu.org/gcc-5/porting_to.html This patch fixes the flamerobin build. I don't understand why flamerobin isn't shown in the recent FTBFS email when this failure is nothing to do with the Boost upgrade and should be happening for all f23 and rawhide builds using GCC-5. commit 79905dc02dd19265d235c0cb37b0360a783c8eeb Author: Jonathan Wakely Date: Mon Jul 27 09:26:07 2015 +0100 Use -P flag during configure tests. diff --git a/flamerobin.spec b/flamerobin.spec index b7aeb95..beae887 100644 --- a/flamerobin.spec +++ b/flamerobin.spec @@ -3,7 +3,7 @@ Summary: Graphical client for Firebird Name: flamerobin Version: 0.9.3 -Release: 11.%{checkout}%{?dist} +Release: 12.%{checkout}%{?dist} License: BSD Group: Applications/Databases #Source0: http://downloads.sourceforge.net/%{name}/%{name}-%{version}-src.tar.gz @@ -30,7 +30,7 @@ toolkit. %setup -q -c -n %{name} %build -%configure +%configure CPPFLAGS=-P make %{?_smp_mflags} %install @@ -59,6 +59,9 @@ rm -rf %{buildroot} %changelog +* Mon Jul 27 2015 Jonathan Wakely 0.9.3-12.20130401snap +- Use -P flag during configure tests. + * Wed Jul 22 2015 David Tardon - 0.9.3-11.20130401snap - rebuild for Boost 1.58 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, 2015-07-24 at 19:26 +0100, Jonathan Wakely wrote: > On 23/07/15 14:33 +0200, David Tardon wrote: > > Hi, > > > > On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > > > Hi, > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > > Any problems rebuilding either open a bug or feel free to email > > > > me or > > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > > to help. > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > * Build failures: > > > > > > - F23 + Rawhide > > > > > > adobe-source-libraries > > This won't even install the BuildRequires, it fails with an error > about libpoppler.so > > Error: Package: 4:texlive-pdftex-bin-svn30845.0 > -11.20140525_r34255.fc23.1.x86_64 (build) >Requires: libpoppler.so.52()(64bit) > Error: Package: 4:texlive-luatex-bin-svn31084.0 > -11.20140525_r34255.fc23.1.x86_64 (build) >Requires: libpoppler.so.52()(64bit) > > Doesn't look related to Boost. poppler got an soname bump at approximately the same time as boost, which is confusing things. Obviously texlive has not yet been rebuilt for the new poppler. For any poppler case, I'd say just go ahead and fire the rebuild. Well, possibly except for texlive, it's never wise to touch texlive until you've said the sacred incantations and sacrificed at least six chickens. > > > inkscape > > Fails due to a glibmm problem, see adamw's email in this thread. I've already worked around that and rebuilt it for main f23 and f24, you should be able to just bump again and build against f23-boost and f24-boost and it'll either succeed or fail for some more boost-y reason. :) > > > xmlcopyeditor > > This fails with a "recompile with -fPIC" error that doesn't seem > related to Boost. fPIC stuff is likely to be related to the newish hardening flags, see https://fedoraproject.org/wiki/Changes/Harden_All_Packages . -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, Jul 24, 2015 at 19:26:09 +0100, Jonathan Wakely wrote: blobby needs this patch to build: http://git.engineering.redhat.com/git/users/jwakely/fedora/blobby/commit/?h=ostream Could you attach a copy of that patch to https://bugzilla.redhat.com/show_bug.cgi?id=1239387 (the blobby FTBFS f23 bug) as the patch doesn't appear to be on a public server. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Fri, Jul 24, 2015 at 19:26:09 +0100, Jonathan Wakely wrote: blobby blobby needs this patch to build: http://git.engineering.redhat.com/git/users/jwakely/fedora/blobby/commit/?h=ostream (I've sent the patch to one of the upstream project maintainers). But it still fails anyway with: /usr/bin/ld: CMakeFiles/blobby.dir/BlobbyDebug.cpp.o: relocation R_X86_64_32 against `.bss' can not be used when making a shared object; recompile with -fPIC CMakeFiles/blobby.dir/BlobbyDebug.cpp.o: error adding symbols: Bad value which doesn't seem to be a Boost problem. I have taken a couple of looks at blobby and I'll take a look at your patch. Unfortunately the svn repo at sourceforge is broken right now, so I can't see if there is a later snapshot than we are using that might be better. It looks like the issue is something might be fixed in a few days or weeks. Blobby was FTBFS before the boost update, so likely there isn't a boost issue. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide adobe-source-libraries This won't even install the BuildRequires, it fails with an error about libpoppler.so Error: Package: 4:texlive-pdftex-bin-svn30845.0-11.20140525_r34255.fc23.1.x86_64 (build) Requires: libpoppler.so.52()(64bit) Error: Package: 4:texlive-luatex-bin-svn31084.0-11.20140525_r34255.fc23.1.x86_64 (build) Requires: libpoppler.so.52()(64bit) Doesn't look related to Boost. blobby blobby needs this patch to build: http://git.engineering.redhat.com/git/users/jwakely/fedora/blobby/commit/?h=ostream (I've sent the patch to one of the upstream project maintainers). But it still fails anyway with: /usr/bin/ld: CMakeFiles/blobby.dir/BlobbyDebug.cpp.o: relocation R_X86_64_32 against `.bss' can not be used when making a shared object; recompile with -fPIC CMakeFiles/blobby.dir/BlobbyDebug.cpp.o: error adding symbols: Bad value which doesn't seem to be a Boost problem. condor Another BuildRequires failure with libpoppler.so elektra /builddir/build/BUILD/elektra-0.8.7/src/bindings/swig/python3/kdb.i:77: Error: Unknown SWIG preprocessor directive: TODO (if this is a block of target language code, delimit it with %{ and %}) src/bindings/swig/python3/CMakeFiles/_swig-python3.dir/build.make:64: recipe for target 'src/bindings/swig/python3/kdbPYTHON_wrap.cxx' failed Probably not Boost related. fityk Not Boost related: app.cpp:12:2: error: #error "Not everything is working with wxGTK3. Use default wxGTK instead, " "based on GTK+2. If you want to test it, just remove this #error." flamerobin This looks like a Boost problem: configure: error: invalid value: boost_major_version= But actually it's caused by a GCC change that puts a # preprocessor marker before macro expansions: boost-lib-version = # 2 "/tmp/conftest.cpp" 3 4 "1_58" The configure script needs to use -P as described at https://gcc.gnu.org/gcc-5/porting_to.html inkscape Fails due to a glibmm problem, see adamw's email in this thread. scantailor This has some weird CMake error about being unable to figure out how to use -pthread. xmlcopyeditor This fails with a "recompile with -fPIC" error that doesn't seem related to Boost. xs This is fixed. I'll keep checking the remaining failures, but maybe not until Monday. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, Jul 23, 2015 at 6:27 AM, David Tardon wrote: > This is a work-in-progress list of FTBFS packages: > > * Build failures: > > - F23 + Rawhide ... > polymake The polymake build was already broken for other reasons (namely, the latest perl update broke it, which happens with pretty much every perl update). I am working with polymake upstream to resolve the issue. -- Jerry James http://www.jamezone.org/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 07/24/2015 02:31 AM, Jonathan Wakely wrote: > On 23/07/15 14:39 -0700, Josh Stone wrote: >> On 07/23/2015 02:33 PM, Josh Stone wrote: >>> Is this a general failure in f23-boost, or just a few packages? >> >> The f23-boost target tag is using build tag f23-build! >> http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=150 >> >> vs f24-boost has itself for both build and destination tag: >> http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=154 > > Thanks for spotting that. I should read more koji docs so I actually > understand these things and I might have noticed it myself. > > Am I right in thinking the build tag defines which other packages it > uses to build against, and the destination tag defines where it goes > after it's built? So everything that was rebuilt was going to the > right tag, but nothing else looked at those rebuilt packages? That's the way I understand it, yes. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thursday, July 23, 2015 02:39:53 PM Josh Stone wrote: > On 07/23/2015 02:33 PM, Josh Stone wrote: > > Is this a general failure in f23-boost, or just a few packages? > > The f23-boost target tag is using build tag f23-build! > http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=150 I have fixed this, everything built for f23-boost will need to be rebuilt Dennis > vs f24-boost has itself for both build and destination tag: > http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=154 signature.asc Description: This is a digitally signed message part. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 22:33, Josh Stone wrote: On 07/22/2015 07:50 AM, Jonathan Wakely wrote: On 22/07/15 15:30 +0100, Tom Hughes wrote: On 22/07/15 14:54, Jonathan Wakely wrote: mapnik-0:2.2.1-0.4.20150127git0639d54.fc23.src nodejs-mapnik-vector-tile-0:0.6.2-6.fc23.src I did these for both branches when you announced it at the weekend. Great, thanks. nodejs-mapnik-0:1.4.17-7.fc23.src This one is proving more interesting... It builds fine for F23 but fails some tests on F24 although it still builds fine with the old boost. As best I can tell it's down to boost::spirit::karma but that hasn't actually changed in the new boost, and in any case the new boost works in F23. I'm kind of stumped at the moment. I'll look into it and see if I can figure it out. It's good that the F23 build is OK, as that's obviously the one with more time pressure. Beware! I was just looking into dyninst, which also succeeded on F23 but failed on F24. But looking at the root.log for F23 reveals it still had 1.57! f23-boost: http://koji.fedoraproject.org/koji/buildinfo?buildID=671537 f24-boost: http://koji.fedoraproject.org/koji/buildinfo?buildID=670925 nodejs-mapnik on f23-boost also had boost 1.57: http://koji.fedoraproject.org/koji/buildinfo?buildID=669822 Now that I have rebuilt everything again both builds of nodejs-mapnik are failing, so at least there is some consistency... Tom -- Tom Hughes (t...@compton.nu) http://compton.nu/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 19:06 +0100, Jonathan Wakely wrote: The attached patch fixes the xs build for rawhide. Upstream fixed it differently: https://github.com/frytvm/XS/commit/034bab9965ab554c4e51053d65cf18a9f40192db I'll add that patch to the spec file instead. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 24/07/15 11:00 +0100, Jonathan Wakely wrote: On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Any problems rebuilding either open a bug or feel free to email me or ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide xs Gah, sorry, this is for the xmlcopyeditor package, not xs (I'm in the process of reporting the xs bug upstream and got my packages beginning with 'x' mixed up :-) This fails with: /usr/bin/ld: xmlcopyeditor.o: relocation R_X86_64_32 against `_ZN7MyFrame13sm_eventTableE' can not be used when making a shared object; recompile with -fPIC xmlcopyeditor.o: error adding symbols: Bad value which doesn't seem related to Boost, and indeed I see the same failure on plain rawhide, without the f4-boost tag. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:33 +0200, David Tardon wrote: Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide xs This fails with: /usr/bin/ld: xmlcopyeditor.o: relocation R_X86_64_32 against `_ZN7MyFrame13sm_eventTableE' can not be used when making a shared object; recompile with -fPIC xmlcopyeditor.o: error adding symbols: Bad value which doesn't seem related to Boost, and indeed I see the same failure on plain rawhide, without the f4-boost tag. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:39 -0700, Josh Stone wrote: On 07/23/2015 02:33 PM, Josh Stone wrote: Is this a general failure in f23-boost, or just a few packages? The f23-boost target tag is using build tag f23-build! http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=150 vs f24-boost has itself for both build and destination tag: http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=154 Thanks for spotting that. I should read more koji docs so I actually understand these things and I might have noticed it myself. Am I right in thinking the build tag defines which other packages it uses to build against, and the destination tag defines where it goes after it's built? So everything that was rebuilt was going to the right tag, but nothing else looked at those rebuilt packages? -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, 2015-07-23 at 14:39 -0700, Josh Stone wrote: > On 07/23/2015 02:33 PM, Josh Stone wrote: > > Is this a general failure in f23-boost, or just a few packages? > > The f23-boost target tag is using build tag f23-build! > http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=150 > > vs f24-boost has itself for both build and destination tag: > http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=154 Thanks for spotting this - I reported it to dgilmore and he fixed it. All the builds that were run in f23-boost will need to be re-done (with a release bump). I tagged libreoffice-5.0.0.3-2.fc23 from f23-boost into f23. If anyone's wondering why - yes it didn't actually build against boost 1.58, but it *did* build against the updated poppler, and we needed a rebuild against the updated poppler for F23 Alpha TC composes. LO builds take so goddamn long it seemed a shame to waste one. I'll kick off a re-rebuild in f23-boost now. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, 2015-07-23 at 13:27 -0700, Adam Williamson wrote: > It seems odd to land the new build into the F23 and Rawhide repos and > *then* set up a side tag to do rebuilds. What's the point of that? Well, I must have misunderstood here, because looking at the actual builds, it seems that indeed boost-1.58.0 was only built in the f23 -boost and f24-boost tags, not in the main tags. That's good, thanks. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 07/23/2015 02:33 PM, Josh Stone wrote: > Is this a general failure in f23-boost, or just a few packages? The f23-boost target tag is using build tag f23-build! http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=150 vs f24-boost has itself for both build and destination tag: http://koji.fedoraproject.org/koji/buildtargetinfo?targetID=154 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 07/22/2015 07:50 AM, Jonathan Wakely wrote: > On 22/07/15 15:30 +0100, Tom Hughes wrote: >> On 22/07/15 14:54, Jonathan Wakely wrote: >> >>> mapnik-0:2.2.1-0.4.20150127git0639d54.fc23.src >>> nodejs-mapnik-vector-tile-0:0.6.2-6.fc23.src >> >> I did these for both branches when you announced it at the weekend. > > Great, thanks. > >>> nodejs-mapnik-0:1.4.17-7.fc23.src >> >> This one is proving more interesting... >> >> It builds fine for F23 but fails some tests on F24 although it still >> builds fine with the old boost. >> >> As best I can tell it's down to boost::spirit::karma but that hasn't >> actually changed in the new boost, and in any case the new boost works >> in F23. I'm kind of stumped at the moment. > > I'll look into it and see if I can figure it out. It's good that the > F23 build is OK, as that's obviously the one with more time pressure. Beware! I was just looking into dyninst, which also succeeded on F23 but failed on F24. But looking at the root.log for F23 reveals it still had 1.57! f23-boost: http://koji.fedoraproject.org/koji/buildinfo?buildID=671537 f24-boost: http://koji.fedoraproject.org/koji/buildinfo?buildID=670925 nodejs-mapnik on f23-boost also had boost 1.57: http://koji.fedoraproject.org/koji/buildinfo?buildID=669822 Is this a general failure in f23-boost, or just a few packages? -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, 2015-07-23 at 13:39 -0700, Adam Williamson wrote: > On Thu, 2015-07-23 at 19:55 +0100, Jonathan Wakely wrote: > > On 23/07/15 14:27 +0200, David Tardon wrote: > > > Hi, > > > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > > Any problems rebuilding either open a bug or feel free to email > > > > > > > > me or > > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy > > > > to > > > > help. > > > > > > This is a work-in-progress list of FTBFS packages: > > > > > > * Build failures: > > > > > > - F23 + Rawhide > > > > > > inkscape > > > > This seems to be a gtkmm or inkscape bug (I have no idea why it > > only > > shows up > > when rebuilding with the new Boost). > > > > /usr/include/gtkmm-2.4/gtkmm/box.h includes > > and > > uses Glib::HelperList unconditionally, but the definition of that > > class template is guarded by: > > > > #ifndef GLIBMM_DISABLE_DEPRECATED > > > > That macro is defined when building Inkscape, so including > > gives an error. > > HelperList's deprecation is fairly recent: > > https://git.gnome.org/browse/glibmm/commit/?id=08c6cc2ca8dfdab8c1294b > c9 > 25ea31df0b6ff8ff > > (and obviously predicated on a wrong belief - "nothing uses it > anymore" > - as we're finding out. :>) Hum, I guess it's not really glibmm's fault. glibmm deprecated it because only really old gtkmm uses it...like the gtkmm inkscape uses. inkscape sets GLIBMM_DISABLE_DEPRECATED because it wants to avoid using deprecated symbols. That's a laudable goal and maybe it was true at some point, but it's kinda silly when it's using such an ancient gtkmm...see below: it's only been working up till now because gtkmm 2.4 ignores the DISABLE_DEPRECATED flags a lot of the time, when it knows it would break otherwise. The quick fix should be to rebuild inkscape with --disable-strict-build . A proper 'fix' might be one or both of two things: 1. inkscape configure.ac already hedges its DISABLE_DEPRECATED stuff in one case it knows about (some particular version of pango); it could also skip the DISABLE_DEPRECATED if the glibmm version is sufficiently new (2.45.3 or later). See: https://github.com/inkscape/inkscape/blob/master/inkscape -launchpad/configure.ac#L864-L893 2. gtkmm 2.4 already quite often temporarily undefines GTK_DISABLE_DEPRECATED and GDK_DISABLE_DEPRECATED when it knows the thing it's about to try and use is deprecated. It could presumably undefine GLIBMM_DISABLE_DEPRECATED right before it tries to use HelperList. I'll file some bugs upstream for this, and I guess check if an inkscape build with --disable-strict-build works. I suspect it may not - I think I see another build attempt which actually did fail on something in boost... -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, Jul 23, 2015 at 7:27 AM, David Tardon wrote: > blender: esound-devel (retired) The blender (root) issue is 2.75 is not building with either the openCOLLADA in Fedora or locally with the latest checkout. I've filed an issue with openCOLLADA github, but no response as of yet. As far as the esound-devel it can be removed, I think all the BR: for blender needs to be scrubbed. Thanks, Richard -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, 2015-07-23 at 19:55 +0100, Jonathan Wakely wrote: > On 23/07/15 14:27 +0200, David Tardon wrote: > > Hi, > > > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > > Any problems rebuilding either open a bug or feel free to email > > > me or > > > ping me on IRC (my freenode nick is 'redi') and I'll be happy to > > > help. > > > > This is a work-in-progress list of FTBFS packages: > > > > * Build failures: > > > > - F23 + Rawhide > > > > inkscape > > This seems to be a gtkmm or inkscape bug (I have no idea why it only > shows up > when rebuilding with the new Boost). > > /usr/include/gtkmm-2.4/gtkmm/box.h includes and > uses Glib::HelperList unconditionally, but the definition of that > class template is guarded by: > > #ifndef GLIBMM_DISABLE_DEPRECATED > > That macro is defined when building Inkscape, so including > gives an error. HelperList's deprecation is fairly recent: https://git.gnome.org/browse/glibmm/commit/?id=08c6cc2ca8dfdab8c1294bc9 25ea31df0b6ff8ff (and obviously predicated on a wrong belief - "nothing uses it anymore" - as we're finding out. :>) -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Wed, 2015-07-22 at 16:18 +0200, David Tardon wrote: > > Also, the timing is really unfortunate. The update should have > happened > before F-23 branch-off, not a week after it :-) Now every package has > to > be rebuilt twice. It's also unfortunate from another angle. We're trying to build Fedora 23 Alpha composes and we're already late; this landed just as anaconda and the compose toolchain was more or less ready, and caused a bunch of dependency issues which break installs and live image creation. It seems odd to land the new build into the F23 and Rawhide repos and *then* set up a side tag to do rebuilds. What's the point of that? The usual way a side tag is used is that the new boost is built *in the side tag* and all the package rebuilds are also done in the side tag; then the new boost and the packages built against it can be tagged into the main repo together, without disrupting things. I don't see that having a side tag for rebuilds against an soname bump that's already in the main repo achieves anything at all. -- Adam Williamson Fedora QA Community Monkey IRC: adamw | Twitter: AdamW_Fedora | XMPP: adamw AT happyassassin . net http://www.happyassassin.net -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:27 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Any problems rebuilding either open a bug or feel free to email me or ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide inkscape This seems to be a gtkmm or inkscape bug (I have no idea why it only shows up when rebuilding with the new Boost). /usr/include/gtkmm-2.4/gtkmm/box.h includes and uses Glib::HelperList unconditionally, but the definition of that class template is guarded by: #ifndef GLIBMM_DISABLE_DEPRECATED That macro is defined when building Inkscape, so including gives an error. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 18:50 +0100, Jonathan Wakely wrote: On 23/07/15 14:27 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Any problems rebuilding either open a bug or feel free to email me or ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide xs This seems to be a bug caused by the new std::string in GCC-5, the error I get is: initial.cxx:2590:34: error: redefinition of 'const char X_7ffd70a41bb0 []' static const char X_7ffd70a41bb0[] = "max-eval-depth"; ^ initial.cxx:104:19: note: 'const char X_7ffd70a41bb0 [15]' previously defined here static const char X_7ffd70a41bb0[] = "fn-%background"; ^ which is due to this: const char *name = str("S_%F", string); if (strlen(name) > MAXVARNAME) name = str("X_%zulx", (long long) string); print("static const char %s[] = ", name); This assumes that every const char* being printed has a unique address, which is not a valid assumption, but probably worked with the Copy-on-Write strings before GCC-5. Or maybe it's a change in Boost.Foreach that causes it. Either way, the code makes invalid assumptions. I'm testing this patch. The attached patch fixes the xs build for rawhide. commit 2da49df85234094fa3f9c9dfabd25dcbb2740293 Author: Jonathan Wakely Date: Thu Jul 23 19:03:32 2015 +0100 Patch dump.cxx to make foreach loop use references not copies. diff --git a/xs-dump-cxx.patch b/xs-dump-cxx.patch new file mode 100644 index 000..18cee75 --- /dev/null +++ b/xs-dump-cxx.patch @@ -0,0 +1,11 @@ +--- dump.cxx~ 2015-07-23 18:48:30.115033012 +0100 dump.cxx 2015-07-23 18:48:31.567038423 +0100 +@@ -263,7 +263,7 @@ + const List *title = runfd(0, "initial.xs", 0); + + printheader(title); +- foreach (Dict::value_type var, vars) dumpvar(var.first.c_str(), var.second); ++ foreach (Dict::value_type& var, vars) dumpvar(var.first.c_str(), var.second); + + /* these must be assigned in this order, or things just won't work */ + varbuf << "\nstatic const struct { const char *name; List *value; } defs[] = {\n"; diff --git a/xs.spec b/xs.spec index 6b78840..7e69d5b 100644 --- a/xs.spec +++ b/xs.spec @@ -2,7 +2,7 @@ Name: xs Version: 0.1 -Release: 15.git%{githash}%{?dist} +Release: 16.git%{githash}%{?dist} Summary: Shell supporting functional programming License: Public Domain @@ -10,6 +10,7 @@ URL: https://github.com/frytvm # downloaded from https://github.com/frytvm/XS/tarball/master Source0: frytvm-XS-%{githash}.tar.gz Patch0: xs-automake-1.12-bison-hxx.patch +Patch1: xs-dump-cxx.patch BuildRequires: automake BuildRequires: bison @@ -36,6 +37,7 @@ Frederic Koehler. # patch for automake-1.12 %patch0 -p1 -b .orig %endif +%patch1 -p0 -b .orig %build @@ -59,6 +61,9 @@ make testxs %changelog +* Thu Jul 23 2015 Jonathan Wakely 0.1-16.git +- Patch dump.cxx to make foreach loop use references not copies. + * Wed Jul 22 2015 David Tardon - 0.1-15.git9c19777 - rebuild for Boost 1.58 -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:27 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Any problems rebuilding either open a bug or feel free to email me or ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide xs This seems to be a bug caused by the new std::string in GCC-5, the error I get is: initial.cxx:2590:34: error: redefinition of 'const char X_7ffd70a41bb0 []' static const char X_7ffd70a41bb0[] = "max-eval-depth"; ^ initial.cxx:104:19: note: 'const char X_7ffd70a41bb0 [15]' previously defined here static const char X_7ffd70a41bb0[] = "fn-%background"; ^ which is due to this: const char *name = str("S_%F", string); if (strlen(name) > MAXVARNAME) name = str("X_%zulx", (long long) string); print("static const char %s[] = ", name); This assumes that every const char* being printed has a unique address, which is not a valid assumption, but probably worked with the Copy-on-Write strings before GCC-5. Or maybe it's a change in Boost.Foreach that causes it. Either way, the code makes invalid assumptions. I'm testing this patch. --- dump.cxx~ 2015-07-23 18:48:30.115033012 +0100 +++ dump.cxx2015-07-23 18:48:31.567038423 +0100 @@ -263,7 +263,7 @@ const List *title = runfd(0, "initial.xs", 0); printheader(title); - foreach (Dict::value_type var, vars) dumpvar(var.first.c_str(), var.second); + foreach (Dict::value_type& var, vars) dumpvar(var.first.c_str(), var.second); /* these must be assigned in this order, or things just won't work */ varbuf << "\nstatic const struct { const char *name; List *value; } defs[] = {\n"; -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 23/07/15 14:27 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Any problems rebuilding either open a bug or feel free to email me or ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide mkvtoolnix For mkvtoolnix I get this error in rawhide: In file included from ./src/mkvtoolnix-gui/util/files_drag_drop_widget.h:6:0, from src/mkvtoolnix-gui/util/files_drag_drop_widget.moc:9: src/mkvtoolnix-gui/util/files_drag_drop_handler.h:12:44: error: expected class-name before '{' token class FilesDragDropHandler: public QObject { ^ That seems to be another missing header (as with rhbz1234405) not a Boost problem. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, Jul 23, 2015 at 1:27 PM, David Tardon wrote: > Hi, > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: >> Any problems rebuilding either open a bug or feel free to email me or >> ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. > > This is a work-in-progress list of FTBFS packages: > > * Build failures: > > - F23 + Rawhide > > abiword > adobe-source-libraries > blobby > condor > elektra > fityk > flamerobin > gazebo > glogg > inkscape > ledger > libyui-qt > mkvtoolnix > paraview > polymake > psi4 > qt-gstreamer > roboptim-core > scantailor > stellarium > xmlcopyeditor > xs > > - Rawhide only > > IQmol > ceph > lucene++ > ompl > pion-net > swift > > * Build failures, probably unrelated > > erlang-bashno_metrics > libreoffice > mesos > pdns > tintii > > * Missing dependencies > > blender: esound-devel (retired) Looking at the last successful build it doesn't even support esound any more so I've dropped the dep and pushed a rebuild. Peter -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On Thu, Jul 23, 2015 at 8:27 AM, David Tardon wrote: > * Build failures: > > - F23 + Rawhide > condor I'll try to take a peek at this one in the next day or two if no one beats me to it. -- Ben Cotton -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Hi, On Thu, Jul 23, 2015 at 02:27:33PM +0200, David Tardon wrote: > Hi, > > On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > > Any problems rebuilding either open a bug or feel free to email me or > > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. > > This is a work-in-progress list of FTBFS packages: > > * Build failures: > > - F23 + Rawhide > > abiword > adobe-source-libraries > blobby > condor > elektra > fityk > flamerobin > gazebo > glogg > inkscape > ledger > libyui-qt > mkvtoolnix > paraview > polymake > psi4 > qt-gstreamer > roboptim-core > scantailor > stellarium > xmlcopyeditor > xs > > - Rawhide only > > IQmol > ceph > lucene++ > ompl > pion-net > swift > > * Build failures, probably unrelated > > erlang-bashno_metrics > libreoffice > mesos > pdns > tintii > > * Missing dependencies > > blender: esound-devel (retired) > rocs: grantlee-qt5-devel > > * Other > > osm2pgsql: missing tarball * Depending on a FTBFS package openoffice.org-diafilter: libreoffice poedit: lucene++ tomahawk: lucene++ D. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Any problems rebuilding either open a bug or feel free to email me or > ping me on IRC (my freenode nick is 'redi') and I'll be happy to help. This is a work-in-progress list of FTBFS packages: * Build failures: - F23 + Rawhide abiword adobe-source-libraries blobby condor elektra fityk flamerobin gazebo glogg inkscape ledger libyui-qt mkvtoolnix paraview polymake psi4 qt-gstreamer roboptim-core scantailor stellarium xmlcopyeditor xs - Rawhide only IQmol ceph lucene++ ompl pion-net swift * Build failures, probably unrelated erlang-bashno_metrics libreoffice mesos pdns tintii * Missing dependencies blender: esound-devel (retired) rocs: grantlee-qt5-devel * Other osm2pgsql: missing tarball D. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 15:30 +0100, Tom Hughes wrote: On 22/07/15 14:54, Jonathan Wakely wrote: mapnik-0:2.2.1-0.4.20150127git0639d54.fc23.src nodejs-mapnik-vector-tile-0:0.6.2-6.fc23.src I did these for both branches when you announced it at the weekend. Great, thanks. nodejs-mapnik-0:1.4.17-7.fc23.src This one is proving more interesting... It builds fine for F23 but fails some tests on F24 although it still builds fine with the old boost. As best I can tell it's down to boost::spirit::karma but that hasn't actually changed in the new boost, and in any case the new boost works in F23. I'm kind of stumped at the moment. I'll look into it and see if I can figure it out. It's good that the F23 build is OK, as that's obviously the one with more time pressure. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 14:54, Jonathan Wakely wrote: mapnik-0:2.2.1-0.4.20150127git0639d54.fc23.src nodejs-mapnik-vector-tile-0:0.6.2-6.fc23.src I did these for both branches when you announced it at the weekend. nodejs-mapnik-0:1.4.17-7.fc23.src This one is proving more interesting... It builds fine for F23 but fails some tests on F24 although it still builds fine with the old boost. As best I can tell it's down to boost::spirit::karma but that hasn't actually changed in the new boost, and in any case the new boost works in F23. I'm kind of stumped at the moment. Tom -- Tom Hughes (t...@compton.nu) http://compton.nu/ -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 16:18 +0200, David Tardon wrote: I've learnt from IRC that I would need to be a provenpackager to do it myself, which I'm not, so either package owners need to rebuild, or if that doesn't happen in time I'll have to rely on the kindness of a provenpackager to help out. I can help. But you should really apply for provenpackager yourself, so you do not have to rely on other people the next time. I'll apply, though I don't know if I'll get approved, as I'm still wet behind the ears at this. If you could help that would be great. Or if package owners want to (maybe temporarily) give me commit privs to their packages I'm happy to do it that way too. Also, the timing is really unfortunate. The update should have happened before F-23 branch-off, not a week after it :-) Now every package has to be rebuilt twice. Yeah, I know :-\ The plan was to rebase to Boost 1.59.0 but it's not out yet, so I went with 1.58.0 but could have made that call sooner and not had to build everything again. I'll learn from the mistake. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Hi, On Wed, Jul 22, 2015 at 01:47:49PM +0100, Jonathan Wakely wrote: > On 22/07/15 13:30 +0100, Jonathan Wakely wrote: > >On 22/07/15 14:08 +0200, David Tardon wrote: > >>Hi, > >> > >>On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > >>>Hi, > >>> > >>>I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which > >>>will require all packages that depend on Boost to be rebuilt. The plan > >>>was to update to 1.59.0 but that isn't going to be released in time > >>>for the F23 schedule, so I'll update rawhide to 1.59.0 at a later > >>>date. > >>> > >>>For the F23 rebuilds please use the tag f23-boost e.g. > >>> > >>>fedpkg build --target f23-boost > >>> > >>>For the rawhide rebuilds please use the tag f24-boost e.g. > >>> > >>>fedpkg build --target f24-boost > >> > >>Does this mean that maintainers are expected to rebuild their packages > >>themselves? > > > >What's the usual policy? Should I be doing it? (Can I do it, if I > >don't have privs to commit a new %release to the spec file and push > >that?) There is no policy, but usually it is that the maintainer of the rebased package who handles rebuild of depending packages. Especially if the rebuild is done in a side tag. > > > >If I'm supposed to be doing it then I'll get started immediately :-) > > I've learnt from IRC that I would need to be a provenpackager to do it > myself, which I'm not, so either package owners need to rebuild, or if > that doesn't happen in time I'll have to rely on the kindness of a > provenpackager to help out. I can help. But you should really apply for provenpackager yourself, so you do not have to rely on other people the next time. Also, the timing is really unfortunate. The update should have happened before F-23 branch-off, not a week after it :-) Now every package has to be rebuilt twice. D. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 13:47 +0100, Jonathan Wakely wrote: On 22/07/15 13:30 +0100, Jonathan Wakely wrote: On 22/07/15 14:08 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Hi, I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which will require all packages that depend on Boost to be rebuilt. The plan was to update to 1.59.0 but that isn't going to be released in time for the F23 schedule, so I'll update rawhide to 1.59.0 at a later date. For the F23 rebuilds please use the tag f23-boost e.g. fedpkg build --target f23-boost For the rawhide rebuilds please use the tag f24-boost e.g. fedpkg build --target f24-boost Does this mean that maintainers are expected to rebuild their packages themselves? What's the usual policy? Should I be doing it? (Can I do it, if I don't have privs to commit a new %release to the spec file and push that?) If I'm supposed to be doing it then I'll get started immediately :-) I've learnt from IRC that I would need to be a provenpackager to do it myself, which I'm not, so either package owners need to rebuild, or if that doesn't happen in time I'll have to rely on the kindness of a provenpackager to help out. I think the list of F23 packages that need rebuilding is: That list was wrong, the (much longer) list for rawhide is shown below. 0ad-0:0.0.18-4.fc23.src CGAL-0:4.6.1-1.fc23.src Coin3-0:3.1.3-9.fc23.src Field3D-0:1.6.1-5.fc23.src FlightCrew-0:0.7.2-16.fc23.src FlightGear-0:3.4.0-5.fc23.src FlightGear-Atlas-0:0.5.0-0.11.cvs20141002.fc23.src IQmol-0:2.3.0-6.fc23.src LuxRender-0:1.3.1-20.fc23.src Macaulay2-0:1.6-13.fc23.src MyPasswordSafe-0:0.6.7-25.20061216.fc23.src OpenImageIO-0:1.5.17-1.fc24.src QuantLib-0:1.6-2.fc23.src Shinobi-0:0.9-8.fc22.src SimGear-0:3.4.0-3.fc23.src SkyX-0:0.4-11.fc23.src abiword-1:3.0.1-4.fc23.src adobe-source-libraries-0:1.0.43-24.fc22.src airinv-0:1.00.1-3.fc23.src airrac-0:1.00.1-2.fc23.src airtsp-0:1.01.2-2.fc23.src akonadi-0:1.13.0-16.fc23.src anyterm-0:1.1.29-27.fc23.src apt-cacher-ng-0:0.8.3-2.fc23.src aqsis-0:1.8.2-18.fc23.src ardour-0:2.8.16-15.fc23.src ardour2-0:2.8.16-15.fc23.src ardour3-0:3.5.403-4.fc23.src ardour4-0:4.1.0-1.fc23.src asc-0:2.6.0.0-9.fc23.src asio-0:1.10.4-4.fc23.src assimp-0:3.1.1-1.fc23.src autowrap-0:0.6.1-4.20150209gitd0e9a5.fc23.src avogadro-0:1.1.1-12.fc23.src barry-0:0.18.4-9.fc23.src bastet-0:0.43.1-6.fc23.src bibletime-0:2.10.1-8.fc23.src blender-1:2.74-6.fc23.src blobby-0:1.0-2svn1541.fc22.src bookkeeper-0:4.2.1-16.fc23.src boost-gdb-printers-0:5-1.fc23.src calligra-0:2.9.6-1.fc24.src cclive-0:0.9.3-8.fc23.src cegui-0:0.8.4-9.fc23.src ceph-1:0.94.2-1.fc24.src clementine-0:1.2.3-10.fc24.src clucene-0:2.3.3.4-19.20130812.e8e3d20git.fc23.src codeblocks-0:13.12-17.fc23.src collada-dom-0:2.4.3-8.fc23.src compat-qpid-cpp-0:0.24-22.fc23.src condor-0:8.3.6-1.fc23.src console-bridge-0:0.2.7-4.fc23.src csdiff-0:1.2.3-2.fc23.src csound-0:6.03.2-6.fc23.src curlpp-0:0.7.3-22.fc23.src cvc4-0:1.4-5.fc23.src cyphesis-0:0.6.2-6.fc23.src dans-gdal-scripts-0:0.23-8.fc23.src davix-0:0.4.1-4.fc23.src device-mapper-persistent-data-0:0.5.4-1.fc24.src diet-0:2.9-2.fc23.src dmlite-0:0.7.2-5.fc23.src dmlite-plugins-s3-0:0.5.1-8.fc23.src dolphin-connector-0:1.2-13.fc23.src dyninst-0:8.2.1-5.fc23.src easystroke-0:0.6.0-10.fc23.src edb-0:0.9.18-14.fc23.src ekiga-0:4.0.1-20.fc23.src elektra-0:0.8.6-2.fc22.src enblend-0:4.1.3-5.fc23.src erlang-basho_metrics-0:1.0.0-15.fc23.src exempi-0:2.2.1-10.fc23.src fawkes-0:0.5.0-23.fc23.src fgrun-0:3.4.0-3.fc23.src fife-2:0.3.3r3-14.fc23.src fityk-0:1.2.9-6.fc23.src flamerobin-0:0.9.3-9.20130401snap.fc22.src flowcanvas-0:0.7.1-19.fc23.src freecad-1:0.15-6.fc23.src fritzing-0:0.9.0b-4.fc23.src fts-0:3.2.32-3.fc23.src fuse-encfs-0:1.7.4-19.fc23.src galera-0:25.3.10-2.fc23.src gappa-0:1.2.0-2.fc23.src gazebo-0:5.1.0-2.fc23.src gearmand-0:1.1.12-12.fc23.src gecode-0:4.4.0-2.fc23.src gfal2-plugin-xrootd-0:0.4.0-2.fc23.src gfal2-python-0:1.8.3-1.fc23.src glob2-0:0.9.4.4-28.fc23.src glogg-0:1.0.2-4.fc23.src glom-0:1.28.4-3.fc23.src gnash-1:0.8.10-17.fc23.src gnote-0:3.17.0-1.fc23.src gnuradio-0:3.7.7.1-2.fc23.src gource-0:0.43-2.fc23.src gpick-0:0.2.5-10.fc23.src gpsdrive-0:2.11-29.fc23.src gqrx-0:2.3.2-7.fc23.src gr-air-modes-0:0-0.33.20140312gitcc0fa180.fc23.src gr-osmosdr-0:0.1.3-6.20141023git42c66fdd.fc23.src grfcodec-0:6.0.5-2.fc23.src gromacs-0:5.0.5-3.fc23.src guitarix-0:0.32.3-4.fc23.src hamlib-0:1.2.15.3-17.fc23.src highlight-0:3.22-2.fc23.src hokuyoaist-0:3.0.2-7.fc23.src hugin-0:2014.0.0-8.fc23.src inkscape-0:0.91-9.fc23.src iwhd-0:1.6-15.fc23.src k3d-0:0.8.0.3-5.fc23.src kcm_systemd-0:1.1.0-3.fc23.src kdenetwork-strigi-analyzers-7:15.04.2-2.fc23.src kdepim-7:4.14.10-1.fc23.src kdepimlibs-0:4.14.10-1.fc23.src kdevelop-9:4.7.1-4.fc23.src kdevplatform-0:1.7.1-3.fc23.src kea-0:0.9.2-0.1.beta.fc23.src kf5-kactivities-0:5.12.0-1.fc23.src kgraphviewer-0:2.2.0-4.fc23.src kicad-0:2015.03.21-3.rev5528.fc23.src kig-0:15.04.3-1.fc23.src kmymo
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 13:30 +0100, Jonathan Wakely wrote: On 22/07/15 14:08 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Hi, I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which will require all packages that depend on Boost to be rebuilt. The plan was to update to 1.59.0 but that isn't going to be released in time for the F23 schedule, so I'll update rawhide to 1.59.0 at a later date. For the F23 rebuilds please use the tag f23-boost e.g. fedpkg build --target f23-boost For the rawhide rebuilds please use the tag f24-boost e.g. fedpkg build --target f24-boost Does this mean that maintainers are expected to rebuild their packages themselves? What's the usual policy? Should I be doing it? (Can I do it, if I don't have privs to commit a new %release to the spec file and push that?) If I'm supposed to be doing it then I'll get started immediately :-) I've learnt from IRC that I would need to be a provenpackager to do it myself, which I'm not, so either package owners need to rebuild, or if that doesn't happen in time I'll have to rely on the kindness of a provenpackager to help out. I think the list of F23 packages that need rebuilding is: asio-0:1.10.4-4.fc23.src barry-0:0.18.4-9.fc23.src CGAL-0:4.6.1-1.fc23.src compat-qpid-cpp-0:0.24-22.fc23.src curlpp-0:0.7.3-22.fc23.src dmlite-0:0.7.2-5.fc23.src dolphin-connector-0:1.2-13.fc23.src dyninst-0:8.2.1-5.fc23.src flann-0:1.8.4-6.fc23.src flowcanvas-0:0.7.1-19.fc23.src gazebo-0:5.1.0-2.fc23.src gnuradio-0:3.7.7.1-2.fc23.src kdepimlibs-0:4.14.10-1.fc23.src kdevplatform-0:1.7.1-3.fc23.src libkolabxml-0:1.0.3-2.fc23.src libyui-0:3.1.5-3.fc23.src luabind-0:0.9.1-17.fc23.src mdds-0:0.12.1-2.fc23.src milia-0:1.0.0-9.fc23.src mlpack-0:1.0.11-5.fc23.src ogre-0:1.9.0-10.fc23.src ompl-0:1.0.0-4.fc23.src permlib-0:0.2.8-11.fc23.src pion-net-0:4.0.9-18.fc23.src player-0:3.0.2-43.fc23.src polybori-0:0.8.3-19.fc23.src qpid-cpp-0:0.32-6.fc23.src qt-gstreamer-0:1.2.0-4.fc23.src rb_libtorrent-0:1.0.4-2.fc24.src rcsslogplayer-0:15.1.1-7.fc23.src rcssserver-0:15.2.2-12.fc23.src rcssserver3d-0:0.6.10-1.fc23.src roboptim-core-0:0.5-13.fc22.src roboptim-trajectory-0:0.5-12.fc22.src simspark-0:0.2.4-13.fc23.src snapper-0:0.2.5-4.fc23.src source-highlight-qt-0:0.2.3-12.fc23.src sympol-0:0.1.8-20.fc23.src thrift-0:0.9.1-16.fc23.2.src urg-0:0.8.18-9.fc23.src vigra-0:1.10.0-12.fc23.src votca-tools-0:1.2.4-5.fc23.src websocketpp-0:0.4.0-4.fc23.src xmms2-0:0.8-27.fc23.src xylib-0:1.4-3.fc23.src yaml-cpp-0:0.5.1-8.fc23.src yaml-cpp03-0:0.3.0-8.fc23.src -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
On 22/07/15 14:08 +0200, David Tardon wrote: Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: Hi, I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which will require all packages that depend on Boost to be rebuilt. The plan was to update to 1.59.0 but that isn't going to be released in time for the F23 schedule, so I'll update rawhide to 1.59.0 at a later date. For the F23 rebuilds please use the tag f23-boost e.g. fedpkg build --target f23-boost For the rawhide rebuilds please use the tag f24-boost e.g. fedpkg build --target f24-boost Does this mean that maintainers are expected to rebuild their packages themselves? What's the usual policy? Should I be doing it? (Can I do it, if I don't have privs to commit a new %release to the spec file and push that?) If I'm supposed to be doing it then I'll get started immediately :-) -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
Re: Boost updated to 1.58.0 in rawhide and f23
Hi, On Sat, Jul 18, 2015 at 12:46:51PM +0100, Jonathan Wakely wrote: > Hi, > > I've pushed a new version of Boost, 1.58.0, to rawhide and f23, which > will require all packages that depend on Boost to be rebuilt. The plan > was to update to 1.59.0 but that isn't going to be released in time > for the F23 schedule, so I'll update rawhide to 1.59.0 at a later > date. > > For the F23 rebuilds please use the tag f23-boost e.g. > > fedpkg build --target f23-boost > > For the rawhide rebuilds please use the tag f24-boost e.g. > > fedpkg build --target f24-boost Does this mean that maintainers are expected to rebuild their packages themselves? D. -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct