Arches, please test and mark stable the latest Qt 4.5.3 ebuilds. This is a priority request as this release fixes a security issue (bug 283810). We will remove all ebuilds for previous versions in a few weeks, or sooner if possible.
Please note that the meta-ebuild (x11-libs/qt-4.5.3) is hardmasked, but we'd
still like it stable for arches where all Qt modules build, so users can unmask and use it if they want to install all Qt modules.
Hmm, something must be broken probably due to this, since I just decided to update my world and install KDE 4 amongst other things, and getting following errors. It's also clear that e.g. qt-opengl-4.5.3 has not yet been marked stable (when looking at ebuild).
I emerge --synced 5 minutes ago and running x86.
!!! One or more updates have been skipped due to a dependency conflict:
('ebuild', '/', 'x11-libs/qt-gui-4.5.3-r1', 'merge') conflicts with
~x11-libs/qt-gui-4.5.2[qt3support,-debug] required by ('ebuild', '/', 'x11-libs/qt-opengl-4.5.2-r1', 'merge')
('ebuild', '/', 'x11-libs/qt-core-4.5.3-r1', 'merge') conflicts with
~x11-libs/qt-core-4.5.2[qt3support,-debug] required by ('ebuild', '/', 'x11-libs/qt-opengl-4.5.2-r1', 'merge')
('ebuild', '/', 'x11-libs/qt-qt3support-4.5.3', 'merge') conflicts with
~x11-libs/qt-qt3support-4.5.2[-debug] required by ('ebuild', '/', 'x11-libs/qt-gui-4.5.2-r2', 'merge')
!!! The following update(s) have been skipped due to unsatisfied dependencies
!!! triggered by backtracking:
And what about Opera 10 bug with Qt > 4.5.2 ? It is fixed ?
(In reply to comment #3)
> And what about Opera 10 bug with Qt > 4.5.2 ? It is fixed ?
Opera bug supposedly fixed in 10.10 snapshot today. No ebuild for it yet, and it is just a snapshot...
Nevermind my last comment. Only partial fix for the issues made it into today's Opera snapshot build :(
Dear arches, due to applying a patch fixing a regression in Qt's event loop several ebuilds got revbumped. Here's the most up-to-date list of ebuilds to be marked stable:
x86 team, please stabilize the revbumped packages again.
Somehow the "Add archs" button refused to work... x86 team, please mark stable the abovementioned Qt 4.5.3 ebuilds as well.
(In reply to comment #7)
> Somehow the "Add archs" button refused to work... x86 team, please mark stable
> the abovementioned Qt 4.5.3 ebuilds as well.
Grrr, if you have an x86 system, do that yourself after a little test, I spent too much time with the initial revisions. Else I will not come to it in the next days.
Anyway, Opera 10.01 works fine here, so I see no issues in stable land.
Yes, Opera 10 build 4685 and Qt 4.5.3 latest "r" works fine. No more loop issue.
(In reply to comment #2)
> Hmm, something must be broken probably due to this, since I just decided to
It could be your your settings. Start from fixing your current profile:
eselect profile set 2
(In reply to comment #6)
> Dear arches, due to applying a patch fixing a regression in Qt's event loop
I've spend more than an hour compiling my stable qt-*-r1 today, and now you've stabilized -r2. This is painful on the old machine. Please allocate more time for stabilizing big packages like QT.
(In reply to comment #12)
> (In reply to comment #2)
> > Hmm, something must be broken probably due to this, since I just decided to
> It could be your your settings. Start from fixing your current profile:
> eselect profile set 2
No - all I had to do was unmask qt-opengl-4.5.3. Now, after a few days and --syncing again everything works. I guess I synced mid-stabilization.
Why is qt-gui on x86 still keyworded?
[I--] [ ~] x11-libs/qt-gui-4.5.3-r2 (4)
After emerge --sync and emerge --deep --update --newuse --verbose --pretend world ... All my qt 4.5.2 packages are blocking 4.5.3 ... What shall I do ?
(In reply to comment #15)
> After emerge --sync and emerge --deep --update --newuse --verbose --pretend
> world ... All my qt 4.5.2 packages are blocking 4.5.3 ... What shall I do ?
Please don't ask for user support in a stabling bug. Go to #gentoo or #gentoo-kde on Freenode IRC or ask on the forums. Those are are first-line support channels.
That being said, Qt has softblocks that should be automatically resolved by portage, if your useflags are in order, and there is nothing wrong with either your world file or your package.mask.
Arches, please mark x11-libs/qt-webkit-4.5.3-r1 stable instead of the earlier revision.
Archs, can we get your attention on this *please* ?
This is an important release, as yngwin pointed out it fixes a security issue.
x86 arch, please stable =x11-libs/qt-phonon-4.5.3 as well, it somehow got away :)
ppc/ppc64 marked stable. I compiled the libraries and tested a number of the demos in qt-demo.
Stable for HPPA and closing.