On Monday 19 July 2010 01:15:43 Andreas Schwab wrote: > John5342 <john5342@xxxxxxxxx> writes: > > Well the problem is obvious. > > Very obvious, indeed. > > > The line in the test program should be: > > > > boost::program_options::variables_map dummy() > > > > and not: > > > > boost::program_options::variables_map::variables_map dummy() > > > > I however know very little about autotools (and don't intend on ever > > learning either) so how to actually fix it is another matter. > > A trivial grep. > > define([testprog], [AC_LANG_PROGRAM([#include <boost/program_options.hpp>], > > [boost::program_options::variables_map::variables_map dummy()])]) > > Andreas. Thanks guys for the help. I'll get a patch in this morning. However....the question remains, why did it build just fine with the function header as-is on a rawhide machine and in my local mockbuild, but not when the build was run on the koji server? Some difference in the autotools, g++ compiler, boost? Any theories? It doesn't seem good at all to have unknown differences in behavior between the developer build environment and and official build environment. -Zach -- devel mailing list devel@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/devel
