>>> Install gentoo-syntax-20070506 into /var/tmp/portage/app-vim/gentoo-syntax-20070506/image/ category app-vim * Fixing file permissions ... * QA Notice: USE Flag 'userland_BSD' not in IUSE for app-vim/gentoo-syntax-20070506 * QA Notice: USE Flag 'userland_Darwin' not in IUSE for app-vim/gentoo-syntax-20070506 [ ok ] In the eclass: if use userland_BSD || use userland_Darwin ; then find ${S} -group 'portage' -exec chgrp wheel '{}' \; || die "chgrp failed" else find ${S} -group 'portage' -exec chgrp root '{}' \; || die "chgrp failed" fi I think you could just use chgrp 0
chgrp 0 would cause an LDAP or similar lookup if the group name 0 did not exist as group name 0 could equal group id 10000 Hence, we have to use names and not id's.
Isn't it time for get_root_user and get_root_grp functions then? I think there are some ebuilds/eclasses that do assume 0 is fine. The Vim eclasses were the only ones not having it, since the maintainers of those eclasses didn't like the chgrp 0.
(In reply to comment #1) > chgrp 0 would cause an LDAP or similar lookup if the group name 0 did not exist > as group name 0 could equal group id 10000 > > Hence, we have to use names and not id's. > Then you need to fix IUSE.
The lines above that use user 'portage' are breaking all ebuilds for vim-plugins for me (Mac OS X prefix), because there is no portage user in the environment. I'm commenting those lines as a work around, but I don't know if that's the right thing to do or not.
I believe this has been fixed a while ago. Should we close this bug?
(In reply to Guilherme Amadio from comment #5) > I believe this has been fixed a while ago. Should we close this bug? Yep, I'll mark this as fixed.