Created attachment 340786 [details, diff] Patch to mercurial-2.5.1.ebuild Attaching a patch.
Seems pretty okay. Is something like PYTHON_MODNAME="${PN} hgext" no longer needed at all? We'd sure want to install the extensions...
(In reply to comment #1) > Seems pretty okay. Is something like PYTHON_MODNAME="${PN} hgext" no longer > needed at all? We'd sure want to install the extensions... Yes, that was necessary for pkg_postinst() module compilation. Now the modules are compiled in src_install().
+*mercurial-2.5.2 (08 Mar 2013) + + 08 Mar 2013; Lars Wendler <polynomial-c@gentoo.org> -mercurial-2.1.1.ebuild, + -files/mercurial-2.4-dont_grep_indefinitely.patch, -mercurial-2.5.ebuild, + +mercurial-2.5.2.ebuild: + Version bump. Removed old. Converted bumped ebuild to use distutils-r1 (bug + #460020). + Let's see if this bombs out :) @mgorny: I had to tweak one line added by your patch a bit. Without that tweak the ebuild failed for me.
(In reply to comment #3) > +*mercurial-2.5.2 (08 Mar 2013) > + > + 08 Mar 2013; Lars Wendler <polynomial-c@gentoo.org> > -mercurial-2.1.1.ebuild, > + -files/mercurial-2.4-dont_grep_indefinitely.patch, -mercurial-2.5.ebuild, > + +mercurial-2.5.2.ebuild: > + Version bump. Removed old. Converted bumped ebuild to use distutils-r1 > (bug > + #460020). > + > > Let's see if this bombs out :) > > @mgorny: I had to tweak one line added by your patch a bit. Without that > tweak the ebuild failed for me. You should've told me. You probably found a bug and worked it around rather than reporting it, resulting in possibly random results depending on eselected Python...
Created attachment 341314 [details, diff] Patch to the current ebuild Please 'cvs up -dP eclass' and try with this patch. With the current eclass, it should work just fine.
(In reply to comment #5) > Created attachment 341314 [details, diff] [details, diff] > Patch to the current ebuild > > Please 'cvs up -dP eclass' and try with this patch. With the current eclass, > it should work just fine. Indeed. I reverted my tweak so the ebuild now has your patch from this bug and works. Thank you :)
Well, I wanted you to commit the whole patch but committed the remaining part myself ;). Now it should all be fixed and clean.
(In reply to comment #7) > Well, I wanted you to commit the whole patch but committed the remaining > part myself ;). Now it should all be fixed and clean. Sorry, I was in a hurry and didn't see your new patch.