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@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct
