[gentoo-dev] Commiting of ~arch virtual/* ebuilds causes deptree issues

2005-12-21 Thread Jakub Moc
Hello here, the virtual/ thingy broke the deptree again with virtual/libstdc++ (see Bug 116253), essentially the same issue like with virtual/x11. These virtuals need to go straight stable if any of their RDEPEND atoms is stable for a particular arch. Betelgeuse is working on a repoman check for

Re: [gentoo-dev] Commiting of ~arch virtual/* ebuilds causes deptree issues

2005-12-21 Thread Donnie Berkholz
Jakub Moc wrote: the virtual/ thingy broke the deptree again with virtual/libstdc++ (see Bug 116253), essentially the same issue like with virtual/x11. These virtuals need to go straight stable if any of their RDEPEND atoms is stable for a particular arch. Shouldn't it be the reverse? Rather,

Re: [gentoo-dev] Commiting of ~arch virtual/* ebuilds causes deptree issues

2005-12-21 Thread Diego 'Flameeyes' Pettenò
On Wednesday 21 December 2005 19:28, Donnie Berkholz wrote: IOW, it doesn't matter if an ~arch virtual depends on stable packages. It matters if stable packages depend on an ~arch virtual. So that's like any other package in the tree.. -- Diego Flameeyes Pettenò -

Re: [gentoo-dev] Commiting of ~arch virtual/* ebuilds causes deptree issues

2005-12-21 Thread Mark Loeser
Jakub Moc [EMAIL PROTECTED] said: Hello here, the virtual/ thingy broke the deptree again with virtual/libstdc++ (see Bug 116253), essentially the same issue like with virtual/x11. These virtuals need to go straight stable if any of their RDEPEND atoms is stable for a particular arch.

Re: [gentoo-dev] Commiting of ~arch virtual/* ebuilds causes deptree issues

2005-12-21 Thread Mark Loeser
Jakub Moc [EMAIL PROTECTED] said: Hello here, the virtual/ thingy broke the deptree again with virtual/libstdc++ (see Bug 116253), essentially the same issue like with virtual/x11. These virtuals need to go straight stable if any of their RDEPEND atoms is stable for a particular arch. Just