[Numpy-svn] [numpy/numpy] de5156: convert_datatype.c: Check for null pointer before ...
noreply at github.com
noreply at github.com
Fri Dec 23 08:31:42 EST 2011
Branch: refs/heads/master
Home: https://github.com/numpy/numpy
Commit: de5156079754f4021553cbaa0d870c168911adf4
https://github.com/numpy/numpy/commit/de5156079754f4021553cbaa0d870c168911adf4
Author: Geoffrey Irving <irving at naml.us>
Date: 2011-12-21 (Wed, 21 Dec 2011)
Changed paths:
M numpy/core/src/multiarray/convert_datatype.c
Log Message:
-----------
convert_datatype.c: Check for null pointer before calling NpyCapsule_Check
Commit: 1b7673d7ddf3671e5ffd05fef864708dc6dd4752
https://github.com/numpy/numpy/commit/1b7673d7ddf3671e5ffd05fef864708dc6dd4752
Author: Geoffrey Irving <irving at naml.us>
Date: 2011-12-21 (Wed, 21 Dec 2011)
Changed paths:
M numpy/core/src/umath/ufunc_type_resolution.c
Log Message:
-----------
BUG: ufunc_object.c: Fix a bug in user-defined loop resolution
Commit: 31eccabfd84dd1058afa0ce8908e00dfd74cfa71
https://github.com/numpy/numpy/commit/31eccabfd84dd1058afa0ce8908e00dfd74cfa71
Author: Geoffrey Irving <irving at naml.us>
Date: 2011-12-21 (Wed, 21 Dec 2011)
Changed paths:
M numpy/core/src/umath/ufunc_type_resolution.c
Log Message:
-----------
BUG: ufunc_type_resolution.c: Fix another user defined type ufunc bug
Commit: 166d0819dc86fb68380b66ffa91c966fa77cbc26
https://github.com/numpy/numpy/commit/166d0819dc86fb68380b66ffa91c966fa77cbc26
Author: Geoffrey Irving <irving at naml.us>
Date: 2011-12-21 (Wed, 21 Dec 2011)
Changed paths:
M numpy/core/include/numpy/npy_3kcompat.h
Log Message:
-----------
REV: Revert check for NULL pointer in NpyCapsule_Check
Essentially all functions in the Python C API require nonnull arguments,
and crash if given nulls. PyCapsule_Check is no exception, and numpy
should probably mirror the standard behavior. As of change e873d58492,
we don't need this check for null, so revert.
Specifically, this reverts commit e9c42cab112d94dbcade9f1afc64b26af0521e14.
Compare: https://github.com/numpy/numpy/compare/3b3735d...166d081
More information about the Numpy-svn
mailing list