Package: release.debian.org Severity: normal User: release.debian.org@packages.debian.org Usertags: transition * Sandro Tosi <morph@debian.org>, 2011-02-28, 19:44:
Hi all,
(Let me turn this mail into a bug report.)
Jakub wants to upload new Sphinx to unstable, but in order to do that, we need also Numpy to go in, and so here it comes this email.Numpy 1.5.1 was uploaded in experimental in late December and it successfully built on every architecture, so the same should happen now (modulo new gcc/ld). Jakub kindly rebuilt these packages, successfully: brian h5py libvigraimpex magics++ mayavi2 nipy petsc4py plplot pygtk pytables python-scipy scikit-learn shogun .
All of these (maybe except scikit-learn) will need binNMUs. I can schedule them myself if that's OK for you.
scikit-learn currently FTBFS with python-sphinx >= 1.0 (bug #588630, already fixed in experimental). I CCed this mail to scikit-learn maintainers.
-- Jakub Wilk