summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
* DOC: nditer: Add links to the nditer introductory doc to make it more ↵Mark Wiebe2011-08-251-0/+2
| | | | discoverable
* Changed to follow PEP 7Chris Jordan-Squire2011-08-221-1/+1
|
* DOCS: New ufunc creation docsChris Jordan-Squire2011-08-221-1/+1
|
* BUG: Fix segfault when PyArray_VOID and no minitype in _array_find_type.Ben Walsh2011-08-212-3/+9
| | | | | The segfault showed up in numpy/testing/print_coercion_tables.py and 'lookfor'.
* BUG: use npy_isnan, not isnan.David Cournapeau2011-08-211-3/+3
|
* BUG: fix dummy module build.David Cournapeau2011-08-211-1/+1
|
* STY: define all config.h macro to value 1 to match autoheader convention.David Cournapeau2011-08-211-6/+6
|
* ENH: Introduce new options extra_f77_compile_args and extra_f90_compile_args toPearu Peterson2011-08-205-2/+32
| | | | | | Configuration.add_extension. Configuration.add_library, and Extension. These options allow specifying extra compile options for compiling Fortran sources within a setup.py file.
* BUG: Revert commit that causes many tests not to run.Charles Harris2011-08-205-32/+2
| | | | | | Revert "Introduce new options extra_f77_compiler_args and extra_f90_compiler_args to Configuration.add_extension. Configuration.add_library, and Extension. These options allow specifying extra compile options for compiling Fortran sources within a setup.py file." This reverts commit 43862759384a86cb4a95e8adb4d39fa1522acb28.
* Merge branch 'pull-140'Charles Harris2011-08-202-5/+30
|\ | | | | | | | | * pull-140: BUG: loadtxt: There was some extra nesting for subarray dtypes (Ticket #1936)
| * BUG: loadtxt: There was some extra nesting for subarray dtypes (Ticket #1936)Mark Wiebe2011-08-202-5/+30
| |
* | BLD: follow build_ext coding convention in build_clib to catch strange ↵Pearu Peterson2011-08-191-12/+14
| | | | | | | | fcompiler value reported in ml.
* | ENH: add useful repr's for finfo, iinfo.Brent Pedersen2011-08-182-0/+20
| |
* | Introduce new options extra_f77_compiler_args and extra_f90_compiler_args to ↵Pearu Peterson2011-08-165-2/+32
|/ | | | Configuration.add_extension. Configuration.add_library, and Extension. These options allow specifying extra compile options for compiling Fortran sources within a setup.py file.
* ENH: refactor doctest plugin to help subclassingMatthew Brett2011-08-161-45/+71
| | | | | | Move numpy-specific parts of the plugin into their own methods, or into class-level defines. This makes it easier to subclass the plugin. This in turn may help keep more eyes on the code.
* ENH: remove unused class definitionMatthew Brett2011-08-161-22/+1
| | | | | | | NumpyDocTestCase definition overwritten further down the file. The deleted class only redefined the ``id`` method with the same code as that in the parent class since before nose 0.10.
* ENH: skip doctests for testsMatthew Brett2011-08-162-11/+9
| | | | | | | | | | | | There are various docstrings show examples of how to run the tests, and give example test output. Obviously the test output changes, and running the doctests for the testing package: import numpy.testing as npt npt.test(doctests=True) will cause several large sets of tests to be run in the rest of the tree. So I skipped these.
* FIX: fix doctest error with empty outputMatthew Brett2011-08-162-1/+21
| | | | | | | | | | | The numpy doctest extension generates an error with empty doctest output; this most often comes about with the +SKIP option. The numpy doctest plugin exposed a nose bug because it accidentally used a different default for the 'doctest-result-variable'. nose bug report here: http://code.google.com/p/python-nose/issues/detail?id=445
* ENH: move doctest tests to own file with ifmainMatthew Brett2011-08-162-29/+35
| | | | | | | The doctesting tests were in the code file, and (for me) rather difficult to run without running lots of other tests. With this change you can run the doctest tests in isolation by executing the test_doctesting.py file.
* ENH: refactor testing to improve subclass supportMatthew Brett2011-08-161-23/+39
| | | | | | | Refactor ``prepare_test_args`` method to make it easier for subclasses to adapt its behavior. This should make it easier for nipy and other projects to use the numpy testing machinery without wholesale copies into their source trees.
* ENH: refactor of docteset plugin managementMatthew Brett2011-08-162-90/+71
| | | | | | | | | | | | | | We previously had a baroque inheritance scheme to deal with the case where the user had normal nose doctests enabled in their environment. However, this scheme didn't deal with bench() routine, and was complicated. This commit uses a null Unplugger plugin to pull the doctest plugin off the nose configuration after it has been initialized. We can use this for bench() and test(), and it allows the doctest module to be enabled (by the user environment) and then thrown away. Also rejigged the docstrings and removed the automated docstring addition as the docstrings have already been copied and adapted in the code.
* ENH: Add provision for headers and footers to savetxt, fixes ticket 1236.Paul Anton Letnes2011-08-152-4/+57
| | | | | I suggest using a separate keyword argument for structured arrays. It might also be nice to be able to add a manual header.
* BUG: print format fix for datetime/timedeltaAlok Singhal2011-08-131-0/+14
|
* STY: Rewrite some macros to look better. This code could probably use aCharles Harris2011-08-131-8/+41
| | | | bigger refactor.
* BUG: Fix import_umath macro to work with python 3.Arfrever Frehtes Taifersar Arahesis2011-08-131-1/+7
|
* ENH: Faster asarray_chkfiniteChris Jordan-Squire2011-08-131-2/+1
|
* BLD: fix build for py3k + pip. Closes #1857. Thanks to Erik Bray.Ralf Gommers2011-08-131-1/+5
| | | | Also works inside a virtualenv.
* BUG: fix failing npyio test under py3k. Thanks to Derek Homeier. Closes #1793.Ralf Gommers2011-08-102-1/+3
|
* BUG: lib: use Py_TYPE to access ob_type, so it works also on Py3Pauli Virtanen2011-07-311-7/+8
|
* ENH: let genfromtxt return empty array for empty input file instead of an error.Paul Anton Letnes2011-07-312-7/+10
| | | | A warning for empty files is issued, including file name. Closes #1793.
* BUG: core: Fix memory leak, add a test which catches it (Ticket #1917)Mark Wiebe2011-07-292-1/+8
|
* BUG: don't use quotes in "dtype=name" when printing arrays for "simple" dtypes'Alok Singhal2011-07-281-1/+5
| | | | See: http://mail.scipy.org/pipermail/numpy-discussion/2011-July/057797.html
* In the old API, PyArray_DATA and PyArray_BYTES differ in that DATA returns a ↵Michael Droettboom2011-07-271-3/+4
| | | | | | (void *) and BYTES returns a (char *). matplotlib builds were failing because PyArray_BYTES was returning a (void *).
* Merge branch 'deprecate_array_field_access'Mark Wiebe2011-07-2667-2371/+2974
|\
| * STY: Remove trailing whitespaceMark Wiebe2011-07-2621-76/+76
| |
| * ENH: core: Rename PyArray_SetBase to PyArray_SetBaseObject to be more clearMark Wiebe2011-07-2221-40/+39
| |
| * ENH: dtype: Make handling of struct dtype align= flag more rigorous (also ↵Mark Wiebe2011-07-223-31/+111
| | | | | | | | | | | | | | fixes ticket #1912) This adds an 'aligned'= property to the format dict, which gets put in the str() representation when necessary to preserve it.
| * STY: Updates for pull request feedback from Chuck and BenMark Wiebe2011-07-201-1/+5
| |
| * ENH: core: Add access macros back in conditionally, for backwards ↵Mark Wiebe2011-07-191-39/+84
| | | | | | | | | | | | compatibilities Sticking with inline functions when NPY_NO_DEPRECATED_API is defined.
| * BUG: mmap: Make the memmap subclass rely on the Python mmap's destructor ↵Mark Wiebe2011-07-192-36/+22
| | | | | | | | | | | | | | | | | | | | | | | | | | being correct There was some fishy code flushing and closing the _mmap property, which looked as if it was paranoid in not trusting mmap.mmap to behave properly. This was done in a way which assumed the .base attribute isn't collapsed, something which I've changed in the refactoring. The easiest way to fix this is to trust mmap.mmap - if this is incorrect, the specific platform on which this fails should have been already documented in the comments!
| * BUG: core: Needed to initialize 'obj' to NULL for Py_XDECREF_ERR in fail sectionMark Wiebe2011-07-192-7/+7
| |
| * TST: rec: DType in join_by test was inconsistentMark Wiebe2011-07-191-2/+3
| |
| * BUG: core: PyArray_GetArrayParamsFromObject was treating __array_interface__ ↵Mark Wiebe2011-07-191-1/+7
| | | | | | | | | | | | | | incorrectly I'm not sure why this came up after the arrayobject field access changes, since it looks like the bug was in there before already.
| * BUG: dtype: comma-list dtype formats didn't accept M8[] parameterized dtypesMark Wiebe2011-07-194-54/+41
| |
| * ENH: core: Fix more test failures post-field access deprecationMark Wiebe2011-07-198-113/+107
| |
| * ENH: core: Got the tests running after the ArrayObject field access deprecationMark Wiebe2011-07-199-55/+61
| |
| * ENH: core: More cleanups removing direct PyArrayObject field accessMark Wiebe2011-07-198-242/+260
| |
| * WRN: nditer: remove some warningsMark Wiebe2011-07-191-4/+4
| |
| * ENH: core: Get it fully building without direct field accessMark Wiebe2011-07-194-78/+81
| |
| * ENH: core: more progress removing direct ArrayObject field accessMark Wiebe2011-07-1914-461/+487
| |
View Lorry Controller Status