| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | BLD: add support for the new X11 directory structure on Ubuntu & co. | Raymond Roberts | 2011-10-29 | 1 | -1/+8 |
| | | |||||
| * | BUG: Fix maximum/minimum for object arrays in Python 3. | Charles Harris | 2011-10-23 | 3 | -26/+64 |
| | | | | | Use PyObject_RichCompareBool instead of PyObject_Cmp. | ||||
| * | STY: Some coding style fixups. | Charles Harris | 2011-10-23 | 1 | -14/+22 |
| | | |||||
| * | ENH: Explicitly coded argmin for timedelta | Ben Root | 2011-10-23 | 5 | -19/+341 |
| | | |||||
| * | ENH: Add 'pgfortran' as a valid compiler in the Portland Group | tim cera | 2011-10-23 | 1 | -10/+10 |
| | | | | | suite of fortran compilers. Adjust the version string command. | ||||
| * | BLD: fix up detection of Intel CPU on OS X in system_info.py. | Ralf Gommers | 2011-10-17 | 1 | -3/+5 |
| | | |||||
| * | ENH: add context manager for NpzFile class. | David Cournapeau | 2011-10-10 | 1 | -0/+13 |
| | | |||||
| * | DOC: mention that NpzFile instances must be closed. | David Cournapeau | 2011-10-10 | 1 | -1/+13 |
| | | |||||
| * | BUG: core: fix compilation for UCS-narrow | Pauli Virtanen | 2011-10-08 | 2 | -1/+2 |
| | | |||||
| * | BUG: Fix version comparison bug on Py25. | Han | 2011-10-04 | 1 | -2/+2 |
| | | |||||
| * | BUG: Fixes for Python3 and some further enhancements. | hangenuit@gmail.com | 2011-10-04 | 1 | -17/+19 |
| | | |||||
| * | ENH: Incorporated review advice and changed macro hack to MinGW hint. | hangenuit@gmail.com | 2011-10-04 | 2 | -28/+7 |
| | | | | | | | | | When doing the typedef change, it conflicted with existing typedefs. I realized that MinGW did not know which MSVC runtime it would link with, but it had conditional definitions for MSVC 8 and up, which can be activated by defining __MSVCRT_VERSION__. So I added it to the compiler macros, which made the fallback hack and extra typedef unnecessary. | ||||
| * | BUG: Add another preprocessor statement for 64-bits fallback mechanism. | hangenuit@gmail.com | 2011-10-04 | 2 | -1/+22 |
| | | |||||
| * | BUG: DLL finder should also look in WinSxS directory. | hangenuit@gmail.com | 2011-10-04 | 1 | -2/+13 |
| | | |||||
| * | BUG: Also build MSVC debug library if MINGW_USE_CUSTOM_MSVCR is defined ↵ | hangenuit@gmail.com | 2011-10-04 | 1 | -10/+6 |
| | | | | | (best effort). | ||||
| * | ENH: In debug mode, do not use customized MSVC runtime library. | Han | 2011-10-04 | 1 | -0/+3 |
| | | |||||
| * | ENH: Clean up symbol definition file. | Han | 2011-10-04 | 1 | -0/+3 |
| | | |||||
| * | ENH: Some commments. Skip customization for MSVC < 8.0. | Han | 2011-10-04 | 1 | -3/+14 |
| | | |||||
| * | ENH: Use customized MSVC runtime library if present. | Han | 2011-10-04 | 1 | -0/+6 |
| | | |||||
| * | ENH: Generate custom MSCV runtime library for mingw32. | Han | 2011-10-04 | 1 | -0/+46 |
| | | |||||
| * | BUG: Fixing reference counting bug in array comparison | Christopher L. Farrow | 2011-10-02 | 1 | -1/+1 |
| | | |||||
| * | BUG: fixing reference leak from the numpy refactor @ 4c21144 | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | |||||
| * | BUG: porting bugfix from refactor @ 708d6be | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | |||||
| * | BUG: porting bugfix from refactor @ 3da77df | Christopher L. Farrow | 2011-10-02 | 1 | -0/+3 |
| | | |||||
| * | BUG: Fixed reference count bug where scalar value assignment by field to an ↵ | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | | | | array would increment the reference count of the scalar. | ||||
| * | BUG: Fixing reference leak in datetime metadata retrieval | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | |||||
| * | BUG: fixed reference leak when new record dtype squashed old one in sort | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | |||||
| * | BUG: fixed reference counting bug in PyArray_FromScalar that occurs when ↵ | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | | | | using an extended dtype | ||||
| * | BUG: Fixed reference leak when gen_arraytype_subscript fails | Christopher L. Farrow | 2011-10-02 | 1 | -0/+1 |
| | | |||||
| * | BUG: fixing reference leak in OBJECT_fill | Christopher L. Farrow | 2011-10-02 | 1 | -1/+3 |
| | | |||||
| * | PyArray_FromBuffer: Allow creating arrays from empty buffers or empty slices. | Hrvoje Niksic | 2011-10-01 | 2 | -3/+6 |
| | | |||||
| * | BUG: Attempt Y2038 fix, and restrict local timezone parsing to 1970 and ↵ | Mark Wiebe | 2011-10-01 | 1 | -37/+77 |
| | | | | | | | | later, because of Win32 On MS Windows, dates earlier than 1970 cause mktime to error. This patch also applies to parsing the same Y2038 hack that was done for printing. | ||||
| * | BUG: Attempt to fix the Y2038 problem parsing dates >= 2038 | Mark Wiebe | 2011-10-01 | 1 | -70/+167 |
| | | | | | | | Also incorporated part of Han Genuit's pull request #23, to try and reduce the difficulty dealing with the merge conflict that will result from these concurrent changes | ||||
| * | TST: Add tests for the Y2038 problem | Mark Wiebe | 2011-10-01 | 1 | -0/+18 |
| | | | | | | This only seems to be cropping up on 32-bit Linux, where time_t is 32 bits, and there appears to be no 64-bit time_t extension like on Windows. | ||||
| * | BUG: Remove floor_divide support for timedelta64 // timedelta64 | Mark Wiebe | 2011-10-01 | 3 | -3/+4 |
| | | |||||
| * | TST: Add true_divide and floor_divide tests for the mm_d timedelta case | Mark Wiebe | 2011-10-01 | 1 | -0/+3 |
| | | |||||
| * | STY: Fix a few cases of spaces not following conventions | Mark Wiebe | 2011-10-01 | 2 | -2/+2 |
| | | |||||
| * | BUG: Add %lld exception for Python 3.1 as well | Mark Wiebe | 2011-10-01 | 1 | -6/+14 |
| | | |||||
| * | WRN: Remove many warnings when building at a higher warning level | Mark Wiebe | 2011-10-01 | 18 | -175/+161 |
| | | | | | | This includes warnings about shadowed variables, some bad casts, and others. Fixing these warnings also fixed a memory leak in one instance. | ||||
| * | BUG: datetime: Fix timedelta / timedelta -> float64 on Python 3 | Mark Wiebe | 2011-10-01 | 3 | -6/+12 |
| | | |||||
| * | BUG: datetime: Fix str() function of datetime arrays | Mark Wiebe | 2011-10-01 | 1 | -3/+18 |
| | | | | | | Done by making the unit given by datetime_data be a string instead of bytes in Python 3. | ||||
| * | BUG: datetime: Make datetime_as_string produce a UNICODE array instead of ↵ | Mark Wiebe | 2011-10-01 | 5 | -28/+45 |
| | | | | | STRING array on Python3 | ||||
| * | BUG: Fix polynomial tests. | Stefan van der Walt | 2011-09-14 | 1 | -2/+2 |
| | | |||||
| * | ENH: Add weights and covariance estimate to standard polyfit. | Travis E. Oliphant | 2011-09-13 | 2 | -20/+72 |
| | | |||||
| * | TST: skip umath complex tests on all platforms except linux. Closes #1926. | Rui Pereira | 2011-09-08 | 1 | -5/+1 |
| | | |||||
| * | BUG: Fix reference leak in PyArray_DTypeFromObject | Han | 2011-09-08 | 1 | -0/+1 |
| | | |||||
| * | MISC: cloadable is not useful anymore. | David Cournapeau | 2011-09-08 | 1 | -1/+1 |
| | | |||||
| * | Merge pull request #149 from 87/ref-fix | Mark | 2011-08-31 | 2 | -2/+2 |
| |\ | | | | | Bugfix in PyArray_SetBaseObject | ||||
| | * | BUG: PyArray_SetBaseObject didn't set proper base when collapsing views | 87 | 2011-08-31 | 2 | -2/+2 |
| | | | |||||
| * | | FIX: Make tril/triu return the same dtype as the original array. | Fabian Pedregosa | 2011-08-31 | 2 | -2/+14 |
| | | | | | | | | | | | | | | | This should fix: * http://projects.scipy.org/numpy/ticket/1848 * http://projects.scipy.org/scipy/ticket/1449 | ||||
 |
index : delta/python-packages/numpy.git | |
| github.com: numpy/numpy.git |
| summaryrefslogtreecommitdiff |