summaryrefslogtreecommitdiff
path: root/numpy/numarray
Commit message (Collapse)AuthorAgeFilesLines
* Rename PyArray_RequireWriteable to PyArray_FailUnlessWriteableNathaniel J. Smith2012-05-221-3/+3
| | | | | | Also clean up its API slightly so that the caller passes in a name describing the array being checked which is used to generate an error message, rather than writing an error message from scratch.
* Funnel all assignments to PyArrayObject->base through a single pointNathaniel J. Smith2012-05-151-3/+1
| | | | | | | | | | | | | | | | | | | | | | | This patch removes all direct assignments of non-NULL values to the 'base' field on PyArrayObjects. A new utility function was created to set up UPDATEIFCOPY arrays, and all assignments to 'base' were adjusted to use either PyArray_SetBaseObject or the new PyArray_SetUpdateIfCopyBase. One advantage of this is that it produces more consistent error handling. This error handling revealed a bug in the nditer code, which this patch does *not* yet fix. The bug is that npyiter_new_temp_array sometimes (when dealing with reversed axes) creates an array and then returns a view onto it. But, npyiter_allocate_arrays assumes that the array returned by npyiter_new_temp_array can have UPDATEIFCOPY set, which requires reassigning its 'base' field. Previously, this meant that the temporary array leaked. Now, it produces a ValueError. This code path doesn't seem to actually be hit very often; only one of the nditer tests fails because of the change. See numpy/core/tests/test_nditer.py:test_iter_array_cast_buggy
* Consolidate all array writeability checking in new PyArray_RequireWriteableNathaniel J. Smith2012-05-151-10/+8
| | | | | | | | | This is mostly a code cleanup, but it does have a user-visible effect in that attempting to write to a unwriteable array now consistently raises ValueError. (It used to randomly raise either ValueError or RuntimeError.) Passes numpy.test("full").
* DEP: Update all the '#define NPY_NO_DEPRECATED_API' instances to beMark Wiebe2012-04-061-1/+1
| | | | versioned
* UPDATE: Replace macros in old_defines.h by new.Charles Harris2012-02-042-18/+18
| | | | | With the exception of the numarray fixups these were all instances that were results of code generation.
* STY: Remove trailing whitespaceMark Wiebe2011-07-264-17/+17
|
* ENH: core: Rename PyArray_SetBase to PyArray_SetBaseObject to be more clearMark Wiebe2011-07-221-1/+1
|
* ENH: core: More cleanups removing direct PyArrayObject field accessMark Wiebe2011-07-192-141/+141
|
* ENH: core: Progress getting NumPy building without direct field accessMark Wiebe2011-07-191-13/+19
|
* STY: Replace remaining old style classes with classes subclassing object.Charles Harris2011-04-051-2/+2
|
* 3K: numarray: Python3 fixes in _capi.cPauli Virtanen2010-07-171-20/+24
| | | | Thanks to Christoph Gohlke.
* BUG: core: use PyCapsule objects only on Python >= 3.0, stay with PyCObjects ↵Pauli Virtanen2010-07-172-2/+2
| | | | on Python 2.x
* BUG: Fix problem with numarray _capi.c and python 2.7.Charles Harris2010-05-241-2/+6
|
* ENH, BUG: PyCObject will be deprecated in python 2.7. So use the NpyCapsuleCharles Harris2010-05-032-2/+2
| | | | | | | compatibility functions in npy_3kcompat.h to replace the current calls. This gets rid of a number of version checks and is easier to maintain. Fix bug that was present in the ufunc _loop1d_list_free destructor in the python3k case.
* ENH: get rid of #warning directives, either by fixing the issue or changing ↵Pauli Virtanen2010-04-041-3/+3
| | | | them to comments
* ENH: Try to fix numarray _capi.c for Python >= 3.1. Needs testing that I am notCharles Harris2010-02-251-10/+40
| | | | in a position to do.
* BUG: NPY_PY3K probably isn't defined when libnumarray.h is included.Charles Harris2010-02-231-1/+1
|
* BUG: Replace deprecated PyCObject by PyCapsule for Python >= 3.1.Charles Harris2010-02-232-21/+44
|
* Fix bad usage of namespace alias changechanley2010-02-221-5/+5
|
* STY: Use import numpy as np. This seems to fix an import error introduced byCharles Harris2010-02-211-5/+13
| | | | | 2to3, but that may have been an artifact from a previous build. In anycase, no harm done.
* BUG: fix numarray._capi compilationPauli Virtanen2010-02-211-1/+1
|
* 3K: ENH: make numpy.numarray to importPauli Virtanen2010-02-211-0/+3
|
* 3K: ENH: move numarray includes under numarray/include so that the 'numpy' ↵Pauli Virtanen2010-02-219-3/+3
| | | | directory does not confuse 2to3's relative import conversion
* BUG: fix numscons build.David Cournapeau2009-12-101-2/+1
|
* 3K: numarray: replace buffer APIs by dummy+error. Some ob_types.Pauli Virtanen2009-12-061-4/+31
|
* ENH: Fix some more uninitialized vars.David Cournapeau2009-09-161-18/+52
|
* Fix installation of numarray headers on Windows.Stefan van der Walt2009-06-091-1/+1
|
* Reindent and cleanup whitespace.Charles Harris2009-06-031-2316/+2314
|
* More signed/unsigned comp fixes.David Cournapeau2009-03-091-6/+6
|
* Fix wrong typedef for UInt64.David Cournapeau2009-03-031-1/+1
|
* ran reindentJarrod Millman2008-12-311-1/+1
|
* More unused vars tagged.David Cournapeau2008-10-081-15/+16
|
* Removed unused/duplicate imports.Alan McIntyre2008-09-135-27/+18
| | | | | | | | Removed repeated members of __all__. Fixed reference to undefined "out" in functions.py:take function. Fixed references to undefined "N" in functions.py. Rewrapped lines to conform to PEP8. Fixed references to undefined FPE_* constants (from numpy) in util.py.
* Simplify a bit more scons scripts.David Cournapeau2008-07-091-1/+1
|
* Use NumpyPythonExtension instead of DistutilsPythonExtension where possible.David Cournapeau2008-07-091-3/+2
|
* Restore old test framework classes.Alan McIntyre2008-06-211-1/+1
| | | | | | | | | | | Added numpy.testing.run_module_suite to simplify "if __name__ == '__main__'" boilerplate code in test modules. Removed numpy/testing/pkgtester.py since it just consisted of an import statement after porting SciPy r4424. Allow numpy.*.test() to accept the old keyword arguments (but issue a deprecation warning when old arguments are seen). numpy.*.test() returns a test result object as before. Fixed typo in distutils doc.
* Switched to use nose to run tests. Added test and bench functions to all ↵Alan McIntyre2008-06-171-0/+4
| | | | modules.
* Adapt numpy.numarray to new build dir convention.David Cournapeau2008-06-122-9/+11
|
* Do not mess with __NUMPY_SETUP__ in scons scripts anymore: this is handled inDavid Cournapeau2008-05-201-4/+1
| | | | | numscons.
* numarray now builds with sconsDavid Cournapeau2008-01-072-0/+26
|
* typoJarrod Millman2007-12-291-1/+1
|
* typoJarrod Millman2007-12-291-1/+1
|
* janitorial workJarrod Millman2007-12-291-50/+55
|
* Fix building on freebsd [patch by Joe Peterson].Stefan van der Walt2007-12-141-3/+7
|
* use 'in' keyword to test dictionary membershipJarrod Millman2007-11-282-4/+4
|
* ran reindent.py to cleanup whitespacesJarrod Millman2007-11-191-4/+4
|
* cleaning up after reindent.pyJarrod Millman2007-10-291-18/+18
|
* ran reindent.py to clean up whitespaceJarrod Millman2007-10-293-17/+15
|
* Small fix for numpy.info(), which was unconditionally brokenFernando Perez2007-06-201-1/+1
|
* #513: fix up include of fenv.c in numarray for cygwincookedm2007-05-152-12/+8
|
View Lorry Controller Status