A new version is out: Release notes (https://github.com/cryos/avogadro/releases) Avogadro 1.2.0 @ghutchis ghutchis released this on 8 Jun · 13 commits to master since this release This release marks a new stable release of Avogadro. It fixes countless bugs and adds important new features as we work on Avogadro v2. Key Updates with 1.2.0: Support for the ORCA quantum chemistry package, thanks to Dagmar Lenk, including input generation and output parsing Improved support for MO calculations, including orbitals with F, G, H, and I angular momentum, thanks to Dagmar Lenk and Albert DeFusco Support for exporting VRML models of atoms, bonds, surfaces, and orbitals (e.g. for 3D printing), thanks to Ethan Pavolik Support for perceiving molecular symmetry on Mac and Linux using the libmsym library (i.e., Properties -> Symmetry), thanks to Marcus Johansson Updated links to the new Avogadro website, manual, and discussion forum Fixed support for compiling with the Eigen3 library up to version 3.2.8 Improved support for space groups through spglib Updated translations, now including over 25 languages in addition to English Fixed a bug downloading from the Protein Data Bank Fixed a bug fetching molecules from the network, including the "chemical by name" Fixed a bug when naming molecules from the NIH chemical resolver website We anticipate several more new releases of Avogadro v1 before we finish Avogadro 2.0 at the end of 2016.
It would be great to have it in the tree, as the old v1.1.1 does not compile after making the transition to the new 17.0 profile (default/linux/amd64/17.0/desktop/plasma/systemd).
See also bug # 619544
Created attachment 511608 [details] avogadro-1.2.0.tar.bz2 A working avogadro-1.2.0.ebuild in my case. Some bits were taken from Debian and Arch. I also added a user hack patch.
(In reply to Attila Tóth from comment #3) > Created attachment 511608 [details] > avogadro-1.2.0.tar.bz2 > > A working avogadro-1.2.0.ebuild in my case. Some bits were taken from Debian > and Arch. I also added a user hack patch. Happy New Year and thanks for your ebuild! Unfortunately, it fails this way in my case: === [ 48%] Linking CXX shared module ../../../lib64/vrmlextension.so cd /var/tmp/portage/sci-chemistry/avogadro-1.2.0/work/avogadro-1.2.0_build/libavogadro/src/extensions && /usr/bin/cmake -E cmake_link_script CMakeFiles/vrmlextension.dir/link.txt --verbose=1 /usr/bin/x86_64-pc-linux-gnu-g++ -fPIC -march=native -O2 -pipe -fvisibility=hidden -fvisibility-inlines-hidden -fpermissive -msse2 -Wl,--enable-new-dtags -Wl,--enable-new-dtags -Wl,--fatal-warnings -Wl,--no-undefined -lc -Wl,-O1 -Wl,--as-needed -shared -o ../../../lib64/vrmlextension.so CMakeFiles/vrmlextension.dir/vrmlextension.cpp.o CMakeFiles/vrmlextension.dir/vrmlpainter.cpp.o CMakeFiles/vrmlextension.dir/vrmldialog.cpp.o CMakeFiles/vrmlextension.dir/moc_vrmlextension.cxx.o CMakeFiles/vrmlextension.dir/moc_vrmldialog.cxx.o -Wl,-rpath,/var/tmp/portage/sci-chemistry/avogadro-1.2.0/work/avogadro-1.2.0_build/lib64:/usr/lib64/qt4: ../../../lib64/libavogadro.so.1.2.0 /usr/lib64/qt4/libQtCore.so /usr/lib64/qt4/libQtGui.so -lopenbabel make[2]: Leaving directory '/var/tmp/portage/sci-chemistry/avogadro-1.2.0/work/avogadro-1.2.0_build' [ 48%] Built target vrmlextension make[1]: Leaving directory '/var/tmp/portage/sci-chemistry/avogadro-1.2.0/work/avogadro-1.2.0_build' make: *** [Makefile:152: all] Error 2 === It seems to me it fails when reaching some openbabel dependency. Here are the openbabel packages installed in my box: [I] sci-chemistry/openbabel Available versions: 2.3.2-r1 2.4.1(0/5.0.0) {doc openmp test wxwidgets} Installed versions: 2.4.1(20:51:54 08/12/17)(openmp wxwidgets -doc -test) Homepage: http://openbabel.sourceforge.net/ Description: Interconverts file formats used in molecular modeling [I] sci-chemistry/openbabel-python Available versions: 2.3.2 2.4.1-r1(0/5) {PYTHON_TARGETS="python2_7 python3_4 python3_5 python3_6"} Installed versions: 2.4.1-r1(21:07:40 08/12/17)(PYTHON_TARGETS="python2_7 python3_5 -python3_4 -python3_6") Homepage: http://openbabel.sourceforge.net/ Description: Python bindings for OpenBabel (including Pybel)
Created attachment 514460 [details] sci-chemistry/avogadro-1.2.0_p20170207 ebuild This is the ebuild I use on my systems. avogadro 1.2.0 has some issues, mainly concerning eigen and gcc. These issues were fixed upstream; this ebuild include the fixes up to the introduction of yaehmop dependency. There is also a python issue, fixed by a simple patch.
Created attachment 514462 [details, diff] renamed file
Created attachment 514464 [details, diff] avogadro-1.20-boost-join-moc.patch renamed file
Created attachment 514466 [details, diff] avogadro-1.2.0-mkspecs-dir.patch renamed file
Created attachment 514468 [details, diff] avogadro-1.2.0-no-strip.patch renamed file
Created attachment 514470 [details, diff] avogadro-1.2.0-numpy.patch fix for python3
Created attachment 514472 [details, diff] avogadro-1.2.0-openbabel.patch renamed file
Created attachment 514474 [details, diff] avogadro-1.2.0-pkgkconfig_eigen.patch renamed file
Qt4 is dead and going away, is there any good reason to have this bump in tree rather than move to avogadro2?
(In reply to Andreas Sturmlechner from comment #13) > Qt4 is dead and going away, is there any good reason to have this bump in > tree rather than move to avogadro2? Current avogadro is working and is actively developed. Avogadro2 is still in beta: https://avogadro.cc/news/avogadro-1-90-released/ states: "We are very proud to announce the availability of Avogadro 1.90.0, the latest Beta for Avogadro 2.0." And, according to authors, avogadro2 has less features than avogadro: "The Avogadro 1.x series currently has more features, and can be found here. We will be porting more features to the Avogadro 2 code base, and making regular releases to get feedback from the community." (from: https://github.com/OpenChemistry/avogadrolibs https://github.com/OpenChemistry/avogadroapp )
If upstream is too slow to port to Qt5 then we will have to drop the Qt4-depending version no matter what. We can probably do one more bump though, as I am unable to even build avogadrolibs/avogadro2 right now.
Your ebuild has a much bigger delta over 1.1.1 than necessary and looks like it may come from an outdated version of the ebuild. This is why attaching unified diffs over the latest version in portage is always preferred rather than full ebuild files. Can you confirm that the only additional patch over 1.1.1 is actually 1.2.0-numpy.patch and the others are identical to 1.1.1-* patches?
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=de6524a1b675617ebcf587d9fb5ba515eb675026 commit de6524a1b675617ebcf587d9fb5ba515eb675026 Author: Andreas Sturmlechner <asturm@gentoo.org> AuthorDate: 2018-01-16 13:03:09 +0000 Commit: Andreas Sturmlechner <asturm@gentoo.org> CommitDate: 2018-01-16 13:04:51 +0000 sci-chemistry/avogadro: 1.2.0_p20170207 version bump Thanks-to: Michelangelo Scopelliti <kernelpanic@gmx.it> Closes: https://bugs.gentoo.org/594630 Closes: https://bugs.gentoo.org/604604 Package-Manager: Portage-2.3.19, Repoman-2.3.6 sci-chemistry/avogadro/Manifest | 1 + .../avogadro/avogadro-1.2.0_p20170207.ebuild | 80 ++++++++++++++++++++++ .../avogadro/files/avogadro-1.2.0-numpy.patch | 13 ++++ 3 files changed, 94 insertions(+)
