Please do not reply directly to this email. All additional comments should be made in the comments box of this bug. https://bugzilla.redhat.com/show_bug.cgi?id=694287 --- Comment #45 from Richard Shaw <hobbes1069@xxxxxxxxx> 2011-04-27 11:17:40 EDT --- (In reply to comment #44) > Almost there, but still needs some work: > > * Please drop everything in the spec file above Name: except for the %define of > AGE, the rest is no longer needed Done. > * Please drop Patch1 and Patch2 they are not needed, they patch SConscript > files which are only used by scons ... I think you mean patch 0 and patch1 since patch2 is your patch. Done. > * Please drop "License: MIT" from the -devel subpackage, since it > is the same as for the main package (I missed that before) Done. > * Remove %{?_smp_mflags} from your make invocation, otherwise the build fails > at least it does so consistently on my quad core. > Funny how you kept the comment from me snippet saying that the build breaks, > but re-added the %{?_smp_mflags} :) When you said it didn't work I didn't think you meant failed to build, just that it didn't do anything. It works for me on my dual core with -j4. Maybe we should spec a max number of threads conditional? > * You're still installing header files from the common dir, resulting in > unittest and performance test headers ending up under /usr/include, but > see the next item for a more radical suggestion for re-arranging the > headers. Should I remove common in addition to the below suggestion? What about Externals/MathMLSolver? > * And last, one slightly larger issue (which I should have checked before). > I'm not really happy with putting a bunch of the .h files directly under > /usr/include. Ideally (IMHO) COLLADAfoo/include/* should end up as > /usr/include/COLLADAfoo/* for all variants of foo Hmm. This sounds like a job for bash and my bashfoo is not that strong :) I'm thinking some sort of for loop that either inverts foo/include to include/foo or that strips the ./include off before copying the files. Another option would be to create a /usr/include/openCOLLADA and then just dump all the files and/or directories under that... What do you think? Richard -- Configure bugmail: https://bugzilla.redhat.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug. _______________________________________________ package-review mailing list package-review@xxxxxxxxxxxxxxxxxxxxxxx https://admin.fedoraproject.org/mailman/listinfo/package-review