I have made and attached new ebuild ( just a epatch line added) and a patch. Patch works with 1.9.2.20,21,22 although I did not full ytest all three versions. They compile fine- that much I can say. Reproducible: Always
Created attachment 306229 [details] failed emerge log
Created attachment 306233 [details] emerge --info
Created attachment 306235 [details, diff] patchfile
The fix is bug 403415 soon enough. Just saying.
Created attachment 306237 [details] new ebuild for 1.9.2.20-r1 versions .21-r1 and .22-r1 can be made from regular .21 and .22 in main tree in similar way- just by adding that epatch line...
I hope I did not screw something up. It compiles fine...
@Samuli Suominen: I didn't see that. In any case, this could take two months if I understood it correctly and in that time someone might find himself in a need for it still...
also, dev-java/swt-3.5.2 and app-text/acroread-9.4.7 seem to need it.
(In reply to comment #7) > @Samuli Suominen: > > I didn't see that. In any case, this could take two months if I understood > it correctly and in that time someone might find himself in a need for it > still... Not likely, gcc-4.7 will not make stable that fast by any means. I am not interested in xulrunner bugs any longer, people have had more then enough time to migrate to npapi-sdk and or webkit.
+ 03 Jun 2012; Justin Lecher <jlec@gentoo.org> xulrunner-2.0.1-r1.ebuild, + +files/xulrunner-2.0.1-gcc47.patch, metadata.xml: + Patch for gcc-47; don't run elibtoolize twice
net-libs/xulrunner was punted from tree wrt bug 403415 so closing this bug too.