Le samedi 23 février 2013 à 17:14 -0800, Ondřej Čertík a écrit : > On Sat, Feb 23, 2013 at 8:35 AM, Sébastien Villemot > <sebast...@debian.org> wrote: > > Le samedi 23 février 2013 à 17:17 +0100, Julien Cristau a écrit : > >> On Sun, Feb 17, 2013 at 13:42:45 +0100, Andreas Beckmann wrote: > >> > >> > I have now observed this behavior also while upgrading the following > >> > packages from squeeze -> sid (and verified that these upgrades fail on > >> > squeeze -> wheezy(+new numpy) in the same way). > >> > > >> If this is due to the breaks added in the new numpy, then I'd say let's > >> revert those. Aborting the upgrade like that seems much worse than > >> possible incompatibilities for partial upgrades. > > > > Yes the problem is introduced by the new Breaks. However, as noted in my > > previous message to this bug, this is the consequence of a real bug > > somewhere else. > > > > If reverting the Breaks still has your preference, I can do it. > > I am the release manager of NumPy 1.7.0. Do you think this is some bug > in NumPy itself?
I think the bug is not in NumPy, see: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=700718#17 This bug has been exposed when some Breaks have been added in python-numpy 1:1.6.2-1.1. -- .''`. Sébastien Villemot : :' : Debian Developer `. `' http://www.dynare.org/sebastien `- GPG Key: 4096R/381A7594
signature.asc
Description: This is a digitally signed message part
