summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
* ma.extraspierregm2010-08-092-29/+52
| | | | * fixed the documentation of (flat)notmasked_xxx
* * fixed numpy.ma.extras.flatnotmasked_contiguous (bug #1576)pierregm2010-08-091-10/+8
|
* ENH: Make atleast_3d respect ndarray subclasses. Closes #1560.rgommers2010-08-081-2/+2
|
* TST: fix memmap test error. Closes #1550. Thanks to Christoph Gohlke.rgommers2010-08-081-0/+1
|
* DEP: Remove ma.extras functions that were deprecated in 1.4.0.rgommers2010-08-051-53/+3
|
* 3K: fix core/dotblas module initialization.Pauli Virtanen2010-08-041-3/+30
|
* 3K: distutils: fix __code__ object access in mingw32ccompilerPauli Virtanen2010-08-031-1/+4
|
* BUG: core/3K: use bytes.maketrans instead of the deprecated string.maketrans ↵Pauli Virtanen2010-08-031-1/+1
| | | | (fixes #1571)
* BUG: core: fix a missed PyCObject call (#1571)Pauli Virtanen2010-08-031-1/+1
|
* BUG: Fix typo in commit r8564-66.rgommers2010-08-011-1/+1
|
* BUG: core: fix refcount error in PyArray_Take (#939)Pauli Virtanen2010-07-312-0/+15
|
* BUG: core/umath: fix powers of complex 0 (#1271)Pauli Virtanen2010-07-312-1/+41
|
* remove unnecessary import and reload in test_getlimits, which was causingDarren Dale2010-07-311-8/+0
| | | | errors when running the test suite.
* TST: Add tests for matlib. Closes #1242.Pauli Virtanen2010-07-312-4/+57
| | | | (From Ralf Gommers.)
* BUG: fix numscons build issue introduced in r8541.rgommers2010-07-311-0/+2
| | | | | The problem was that if endian.h was not detected, @DEFINE_NPY_HAVE_ENDIAN_H@ was still present in _numpyconfig.h, causing compilation to fail.
* DOC: wiki merge, matrlixlib and some ndarray methods.rgommers2010-07-313-28/+30
|
* DOC: wiki merge, twodim_base and a few loose ones.rgommers2010-07-314-52/+54
|
* DOC: wiki merge, npyio, format and function_basergommers2010-07-313-64/+91
|
* DOC: wiki merge, core.umathrgommers2010-07-311-131/+165
|
* DOC: wiki merge, core/numeric.pyrgommers2010-07-311-22/+26
|
* DOC: wiki merge, fromnumeric.pyrgommers2010-07-311-107/+121
|
* DOC: wiki merge, add itemset method doc.rgommers2010-07-311-13/+59
|
* DOC: wiki merge, add distutils.npy_pkg_config.read_config docstring.rgommers2010-07-311-2/+40
|
* BUG: fix base_repr for py3k. strings can not be indexed with floats anymore.rgommers2010-07-311-1/+1
|
* ENH: Make trapz work with ndarray subclasses. Thanks to Ryan May. Closes #1438.rgommers2010-07-312-5/+37
|
* BUG: core/umath: do not create views unnecessarily in ↵Pauli Virtanen2010-07-292-0/+14
| | | | ndarray.__array_prepare__ (fixes #1548)
* BUG: make sure npy_cpu.h is always included in ndarraytypes.hDavid Cournapeau2010-07-291-0/+1
|
* BUG: look for endian.h.David Cournapeau2010-07-293-0/+10
| | | | We should use this header in npy_endian.h if available.
* ENH: testing: add assert_tol_equal for testing array equality with specified ↵Pauli Virtanen2010-07-282-1/+60
| | | | tolerances
* BUG: support assignment x[(0,)] = foo in ndarray subtypes (#1563)Pauli Virtanen2010-07-282-3/+17
| | | | | | The problem was a code path trying to assign x[0], which returns a scalar -- for actual ndarrays a internal function always returning arrays was used.
* Fix copy-and-paste error discovered by re-factor testing...Travis Oliphant2010-07-281-1/+1
|
* BUG: fix usability bugs in lookforPauli Virtanen2010-07-281-8/+9
| | | | | | | | - allow numbers in function signature - print the items in relevance order (not reverse relevance) - include ufunc docstrings
* BUG: New implementation of base_repr. Fixes #1549, is several times faster andrgommers2010-07-282-23/+31
| | | | adds tests. Thanks to K. Maglione.
* numpy.distutils: Fixed bug in Configuration._get_svn_revision. Introduced ↵Pearu Peterson2010-07-271-3/+95
| | | | make_hg_version_py method, get_version supports hg revision.
* BUG: attempt at fixing alpha build failure (quad prec little endian).David Cournapeau2010-07-271-0/+35
|
* DOC: Merge wiki changes for RandomState as far as possible. Closes #1503.rgommers2010-07-261-57/+69
|
* f2py: fixed issue 1533 (scanning pyf files will report lines that do not ↵Pearu Peterson2010-07-241-1/+1
| | | | match known patterns).
* BUG: quick and ugly fix for long double on linux ppc.David Cournapeau2010-07-202-1/+126
|
* ENH: fix long double detection for linux ppcDavid Cournapeau2010-07-193-1/+9
| | | | | | This does not fix build on linux ppc, as the implementation for long double functions is missing.
* BUG: core: fix argmax and argmin NaN handling to conform with max/min (#1429)Pauli Virtanen2010-07-182-1/+70
| | | | | | This makes `argmax` and `argmix` treat NaN as a maximal element. Effectively, this causes propagation of NaNs, which is consistent with the current behavior of amax & amin.
* BUG: core/umath: make complex number comparisons False when *either* element ↵Pauli Virtanen2010-07-183-8/+35
| | | | | | | | | | | is nan This also fixes NaN propagation in np.maximum/minimum, for example in the case for maximum(1, complex(0, nan)) -> (0, nan) which previously yielded (1, 0).
* BUG: core: make set_string_function(None) to restore the functions to what ↵Pauli Virtanen2010-07-183-48/+73
| | | | they are on import (#1130)
* BUG: core: format 'nan' and 'inf' also in array repr by default (#1050)Pauli Virtanen2010-07-182-4/+14
|
* BUG: core: make .std() and .var() respect the out= keyword (#1434)Pauli Virtanen2010-07-182-3/+21
|
* Merge branch 'wgt'Charles Harris2010-07-185-75/+212
|
* 3K: numarray: Python3 fixes in _capi.cPauli Virtanen2010-07-171-20/+24
| | | | Thanks to Christoph Gohlke.
* 3K: f2py: catch a misc import errorPauli Virtanen2010-07-171-1/+1
|
* 3K: lib/tests: fix test_io.RoundtripTest on Python3 + WindowsPauli Virtanen2010-07-171-1/+2
|
* BUG: core: finfo.tiny, resolution, and epsneg should be scalars (#1538)Pauli Virtanen2010-07-172-2/+7
|
* BUG: core: harmonize complex number formatting with Python 2.7 (#1534)Pauli Virtanen2010-07-171-1/+1
|
View Lorry Controller Status