Hi Avogadro team, I'm the avogadro maintainer in Arch Linux. We are updating boost to 1.48.0, but Qt moc has a bug with this boost version (you can find more info here[1]). Avogadro uses moc, so avogadro has this bug too. In fact when you build avogadro 1.0.3 with boost 1.48.0 you get:
[ 4%] Generating pythonextension_p.moc usr/include/boost/type_traits/detail/has_binary_operator.hp:50: Parse error at "BOOST_JOIN" make[2]: *** [libavogadro/src/pythonextension_p.moc] Error 1 make[1]: *** [libavogadro/src/CMakeFiles/avogadro.dir/all] Error 2 In the qt bug report an user said that passing a specific option to moc the bug can be avoided. But the only one method to pass some option to moc in cmake is the QT4_WRAP_CPP macro. You use this macro in the libavogadro/CMakeLists.txt, but you don't in libavogadro/src/CMakeLists.txt and libavogadro/src/python/CMakeLists.txt; you use qt4_automoc there. So I replaced qt4_automoc usage with the qt4_wrap_cpp macro in those CMakeLists.txt. The goal of this post is that you could merge my patch if you want, is here[2]. Note, I had to move the ReadFileThread class declaration in another header (moleculefile_p.h) to make the macro work. Cheers. [1] https://bugreports.qt.nokia.com/browse/QTBUG-22829 [2] https://projects.archlinux.org/svntogit/packages.git/tree/trunk/replace-qt4_automoc-with-qt4_wrap_cpp.patch?h=packages/avogadro -- Andrea ------------------------------------------------------------------------------ All the data continuously generated in your IT infrastructure contains a definitive record of customers, application performance, security threats, fraudulent activity, and more. Splunk takes this data and makes sense of it. IT sense. And common sense. http://p.sf.net/sfu/splunk-novd2d _______________________________________________ Avogadro-devel mailing list [email protected] https://lists.sourceforge.net/lists/listinfo/avogadro-devel
