On 2016-05-28 18:51:19 -0300, Lisandro Damián Nicanor Pérez Meyer wrote: > Vincent: if I add that breaks then qt in unstable/testing becomes > uninstalable.
Even with versioned dependencies? But... > The bug is *solved* in both unstable and experimental, as both > versions of qtbase already have the fix. Well... I wasn't aware of this, and the culprit was apt-show-versions, which was saying that 5.6.0+dfsg-2 was uptodate, though it knew about 5.6.0+dfsg-3. I have a cron command[*] that normally tells me which packages are manually upgradeable (useful for experimental), so here it saw nothing. :( [*] apt-show-versions -u | grep manually > The real bug was an uncoordination due to us being preparing qt 5.6 > in experimental and it only affects experimental users. > > I accept the severity of this bug to be at least important, although I > still don't think it's a bug at all, and in the worst case it was a bug in > qtbase in experimental. If you insist on this i can update qtchooser's B+R > whenenever we get to upload qt 5.6 to unstable. That's now OK for me. Now that the experimental libqt5core5a is no longer broken, the "Breaks" is perhaps not much useful any longer (well, it might be now for potential users who still have a buggy libqt5core5a but this will be less and less the case). -- Vincent Lefèvre <vinc...@vinc17.net> - Web: <https://www.vinc17.net/> 100% accessible validated (X)HTML - Blog: <https://www.vinc17.net/blog/> Work: CR INRIA - computer arithmetic / AriC project (LIP, ENS-Lyon)
