| Commit message (Collapse) | Author | Age | Files | Lines | ||
|---|---|---|---|---|---|---|
| ... | ||||||
| * | 3K: BUG: do not convert Bytes/Unicode via PEP 3118 buffers -- use 'S/U#' instead | Pauli Virtanen | 2010-02-20 | 1 | -5/+3 | |
| | | ||||||
| * | BUG: core: prefer byte strings over PEP 3118 buffers when determining array ↵ | Pauli Virtanen | 2010-02-20 | 1 | -15/+15 | |
| | | | | | type in _array_find_type | |||||
| * | 3K: BUG: core: fix some bytes/str issues in defchararray.py | Pauli Virtanen | 2010-02-20 | 1 | -2/+4 | |
| | | ||||||
| * | 3K: ENH: core: audit some PyString -> PyBytes conversions | Pauli Virtanen | 2010-02-20 | 1 | -3/+3 | |
| | | ||||||
| * | ENH: core: make PyArray_FromAny error message more clear | Pauli Virtanen | 2010-02-20 | 1 | -1/+14 | |
| | | | | | | | | | | | | | | | When building string/unicode arrays, PyArray_FromAny would try building a 0-d array as a last resort. If the object is a finite-length sequence, this always fails with an error "ValueError: setting an array element with a sequence", as STRING/UNICODE_setitem do not allow this. In some cases this message is confusing: for example on Py3, array([['a', 'b'], ['c', '\u03a3']], dtype='S2') would fail because of this, even though the real error is that '\u03a3' cannot be unambiguously cast to bytes(). | |||||
| * | 3K: BUG: cast any object to unicode/bytes in UNICODE/STRING_setitem also for ↵ | Pauli Virtanen | 2010-02-20 | 1 | -4/+34 | |
| | | | | | | | | | Py3, similarly as it is done in Py2 This makes UNICODE_setitem and STRING_setitem work similarly as on Python 2 -- uses ASCII codec -- I judge this useful for compatibility reasons. | |||||
| * | 3K: BUG: make test_defchararray convert properly via 2to3 | Pauli Virtanen | 2010-02-20 | 1 | -3/+6 | |
| | | ||||||
| * | 3K: core: adjust some tests vs. str/bytes and int inheritance issues | Pauli Virtanen | 2010-02-20 | 6 | -25/+48 | |
| | | ||||||
| * | 3K: core: fix one str/bytes issue in records.py | Pauli Virtanen | 2010-02-20 | 1 | -2/+3 | |
| | | ||||||
| * | 3K: BUG: fix tofile/fromfile on Py3 by dup-ing the file descriptor before ↵ | Pauli Virtanen | 2010-02-20 | 3 | -8/+31 | |
| | | | | | fdopening -- works around issues in flushing data | |||||
| * | 3K: BUG: core: fix some tests for Py3 | Pauli Virtanen | 2010-02-20 | 2 | -8/+10 | |
| | | ||||||
| * | ENH: core: implement support for PEP 3118 buffer objects in asarray() and ↵ | Pauli Virtanen | 2010-02-20 | 2 | -5/+145 | |
| | | | | | | | | | array() This is available only for Python >= 2.7, since the MemoryView object is not present in earlier versions of Python. (It makes releasing the buffer much easier, via garbage collection.) | |||||
| * | ENH: core: add some support routines needed for consuming PEP 3118 buffers | Pauli Virtanen | 2010-02-20 | 5 | -1/+193 | |
| | | ||||||
| * | ENH: add some tests for PEP 3118 buffer interface | Pauli Virtanen | 2010-02-20 | 1 | -5/+124 | |
| | | ||||||
| * | ENH: core: require PEP 3118 buffer contiguity only with ND and not STRIDES | Pauli Virtanen | 2010-02-20 | 1 | -8/+21 | |
| | | ||||||
| * | ENH: core: cleanups in PEP 3118 provider | Pauli Virtanen | 2010-02-20 | 1 | -36/+38 | |
| | | ||||||
| * | ENH: core: support sub-arrays in PEP 3118 buffers | Pauli Virtanen | 2010-02-20 | 1 | -6/+27 | |
| | | ||||||
| * | ENH: rewrite PEP 3118 interface without adding items to structs | Pauli Virtanen | 2010-02-20 | 4 | -97/+211 | |
| | | | | | | | | | | | | | | | | | | | | | | Implementing PEP 3118 is somewhat convoluted because of the desirata: - Don't add new members to ndarray or descr structs, to preserve binary compatibility. (Also, adding the items is actually not very useful, since mutability issues prevent an 1 to 1 relationship between arrays and buffer views.) - Don't use bf_releasebuffer, because it prevents PyArg_ParseTuple("s#", ...) from working. Breaking this would cause several backward compatibility issues already on Python 2.6. - Behave correctly when array is reshaped in-place, or it's dtype is altered. This commit works around these issues by keeping copies of buffer format strings and shape/stride information in a global dictionary. Elements in the dictionary are reused when possible, and cleared when an array is deallocated. | |||||
| * | 3K: core: sort using the key= method | Pauli Virtanen | 2010-02-20 | 1 | -1/+1 | |
| | | ||||||
| * | 3K: BUG: bytes as dtype should map to 1-byte strings, not long | Pauli Virtanen | 2010-02-20 | 1 | -6/+0 | |
| | | ||||||
| * | updated documentation from pydoc website (thanks to everyone who contributed!) | Jarrod Millman | 2010-02-17 | 3 | -14/+30 | |
| | | ||||||
| * | more docstring updates from pydoc website (thanks to everyone who contributed!) | Jarrod Millman | 2010-02-17 | 17 | -626/+1281 | |
| | | ||||||
| * | updated docstrings from pydoc website (thanks to everyone who contributed!) | Jarrod Millman | 2010-02-17 | 6 | -46/+122 | |
| | | ||||||
| * | STY: Remove hard tabs, reindent. | Charles Harris | 2010-02-17 | 1 | -144/+150 | |
| | | ||||||
| * | STY: Remove hardtabs, reindent. | Charles Harris | 2010-02-17 | 1 | -145/+136 | |
| | | ||||||
| * | STY: Remove hardtabs, reindent. | Charles Harris | 2010-02-17 | 1 | -182/+250 | |
| | | ||||||
| * | STY: Remove hardtabs, reindent. | Charles Harris | 2010-02-17 | 1 | -15/+19 | |
| | | ||||||
| * | STY: Remove hard tabs, reindent. | Charles Harris | 2010-02-17 | 1 | -8/+9 | |
| | | ||||||
| * | STY: Remove hard tabs. Reindent. | Charles Harris | 2010-02-17 | 1 | -43/+43 | |
| | | ||||||
| * | STY: Remove hard tabs. | Charles Harris | 2010-02-17 | 1 | -7/+7 | |
| | | ||||||
| * | STY: Remove hard tab. | Charles Harris | 2010-02-17 | 1 | -3/+2 | |
| | | ||||||
| * | BUG: Import warnings module in polynomial modules. | Charles Harris | 2010-02-17 | 2 | -1/+2 | |
| | | ||||||
| * | removed old behavior for the histogram function. | dhuard | 2010-02-16 | 3 | -158/+67 | |
| | | ||||||
| * | Fix Order of scalartypes. | Travis Oliphant | 2010-02-16 | 1 | -3/+3 | |
| | | ||||||
| * | Update ordering of cast functions which was incorrectly placed. | Travis Oliphant | 2010-02-15 | 1 | -1/+1 | |
| | | ||||||
| * | API and ABI changes: Removed BigArrayType, Fixed hasobject to an integer ↵ | Travis Oliphant | 2010-02-15 | 10 | -314/+325 | |
| | | | | | (and inserted an unused character for alignment) and renamed it to flags. Re-organized the ArrFuncs structure. Moved NPY_DATETIME and friends to their 'correct' place in the type order. | |||||
| * | Fix #1367 (round 2) | pierregm | 2010-02-14 | 1 | -1/+1 | |
| | | ||||||
| * | Fix #1367 | pierregm | 2010-02-14 | 1 | -1/+1 | |
| | | ||||||
| * | BUG: Apply patch to fix ticket #1326. | Charles Harris | 2010-02-14 | 1 | -1/+1 | |
| | | ||||||
| * | Fix _dotblas ob_type pointer for py3k. | Charles Harris | 2010-02-13 | 1 | -4/+4 | |
| | | ||||||
| * | Update trunk version to 2.0 and fix descriptor pickle code so that it ↵ | Travis Oliphant | 2010-02-12 | 1 | -42/+53 | |
| | | | | | produces 1.3-compatible pickle unless it cannot. | |||||
| * | ENH: handle complex input for assert_array_almost_equal_nulp. | David Cournapeau | 2010-02-09 | 1 | -3/+6 | |
| | | ||||||
| * | BUG: fix typo. | David Cournapeau | 2010-02-09 | 1 | -1/+1 | |
| | | ||||||
| * | TST: add simple test for complex arrays input to assert_array_almost_equal_nulp. | David Cournapeau | 2010-02-09 | 1 | -0/+13 | |
| | | ||||||
| * | BUG: Check input to poly for zero-dimensional arrays. | Stefan van der Walt | 2010-02-08 | 2 | -2/+8 | |
| | | ||||||
| * | * allow fill_value to be np.void | pierregm | 2010-02-08 | 1 | -4/+1 | |
| | | ||||||
| * | * Make sure _fill_value is never None for structured masked arrays | pierregm | 2010-02-08 | 2 | -6/+38 | |
| | | ||||||
| * | * Force the fill_value of a structured masked array to be defined (bug #1332) | pierregm | 2010-02-07 | 2 | -0/+10 | |
| | | ||||||
| * | ctypeslib: Correct tabs to spaces | Pauli Virtanen | 2010-02-04 | 1 | -2/+2 | |
| | | ||||||
| * | BUG: Fix #1376, indentation problem. | Travis Oliphant | 2010-02-04 | 1 | -2/+2 | |
| | | ||||||
 |
index : delta/python-packages/numpy.git | |
| github.com: numpy/numpy.git |
| summaryrefslogtreecommitdiff |