Control: tags -1 + pending
Since pygtk finally compiled on ia64, I have uploaded to DELAYED/2 a NMU
of python-numpy. The debdiff is attached.
--
.''`.Sébastien Villemot
: :' :Debian Developer
`. `' http://www.dynare.org/sebastien
`- GPG Key: 4096R/381A7594
diff -Nru python-n
Update: all reverse dependencies are now fixed, except pygtk which FTBFS
on ia64. Once this is sorted out, I plan to NMU python-numpy with the
patch previously sent, and then finally close this bug.
--
.''`.Sébastien Villemot
: :' :Debian Developer
`. `' http://www.dynare.org/sebasti
Le mercredi 21 novembre 2012 à 22:35 +0100, Sébastien Villemot a écrit :
> This makes a total of 7 packages possibly affected. The brute force
> solution would be to binNMU all of them. A more elegant solution would
> be to investigate their source code to see if they are really affected,
> but th
Package: python-numpy
Followup-For: Bug #685812
Upstream discussed this issue [1]. It looks like they consider NPY_CHAR
to be an obsolete interface. However they did not advertise it clearly,
and they did not intentionally break the API. It is clear that they
won't fix this issue, since NPY_CHAR
I found the cause of this bug: in
numpy/core/include/numpy/ndarraytypes.h, in the enum NPY_TYPES, the
value NPY_CHAR has changed between NumPy 1.5 and 1.6 (more precisely, it
was increased by 3, because new types were introduced).
python-scientific uses this enum value at several places (under its
control: severity -1 important
The 2.8-3+b3 binnmu solved the breakage problem originally reported.
An abi bump is still useful, but since there is no longer breakage,
this is no longer release-critical.
Best wishes,
Mike
___
Python-modules-team mailin
6 matches
Mail list logo
