summaryrefslogtreecommitdiff
path: root/numpy/core
Commit message (Collapse)AuthorAgeFilesLines
* commit BLD: AIX uses the flag _LARGE_FILES to ensure proper prototype ↵Michael Felt2016-10-211-3/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | declarations The problem this fix resolves is to ensure that 32-bit and 64-bit functions (e.g., fclear() and fclear64()) to access/manipulate "large files" are defined properly - much as GNU and other platforms use one or more of the defines _FILE_OFFSET_BITS, _LARGEFILE_SOURCE, and _LARGEFILE64_SOURCE. Without this fix the numpy code only defines flags that are not recognized in the AIX environment and have no effect in anyway. The fix applies the AIX convention and does not "export" any of the flags currently exported. For all other platforms the current flags: _FILE_OFFSET_BITS, _LARGEFILE_SOURCE, and _LARGEFILE64_SOURCE are "exported". This fix should not have any impact or side-effect based on the version of python used. History: Starting around 1997 AIX started supporting so-called "large files", i.e., length > signed 32-bit. In the period 1997-1998 the flag _LARGE_FILES was established to simplify porting 32-bit applications to 64-bit. The convention is to define _LARGE_FILES before including any "system include files" either as an argument to ${CC} (e.g., in ${CFLAGS} or as the first #define in every source file. This is to ensure that that no relevant function calls would be redefined later in the build process.
* BUG: fix _array2string for strustured array (issue #5692)Shota Kawabuchi2016-10-182-33/+81
| | | | | | | | | | | | | The cause of issue #5692 is that `_array2string` (in numpy/core/arrayprint.py) doesn’t have format function for structured arrays and it uses general purpose format function to format array elements of structured arrays. This commit adds `StructureFormat` class to format structured array elements. `_get_format_function` instantiates `StructureFormat` by instantiating format function classes for each field of structure recursively and merge them. Closes #5692.
* DOC: added example with empty indices for a scalarMattHarrigan2016-10-141-0/+3
|
* Merge pull request #8153 from ahaldane/scalar_baseCharles Harris2016-10-132-0/+22
|\ | | | | MAINT: Expose void-scalar "base" attribute in python
| * MAINT: Expose void-scalar "base" attribute in pythonAllan Haldane2016-10-132-0/+22
| |
* | MAINT: make sure __builtin_cpu_supports is not optimized awayJulian Taylor2016-10-111-1/+3
| | | | | | | | | | | | OSX 10.11 ships a clang that uses it but does not implement the necesary library function closes gh-8130
* | MAINT: check that linker can handle AVXJulian Taylor2016-10-113-6/+13
|/ | | | | | some people use new compilers with old linkers so the target attribute check for AVX is not enough closes gh-8128
* Merge pull request #8103 from mattip/pypy-fixesNathaniel J. Smith2016-10-083-57/+78
|\ | | | | Pypy fixes
| * ENH: pypy needs a basic tp_as_number at PyType_Ready in order to allow ↵Matti Picus2016-10-011-0/+21
| | | | | | | | overriding function pointers
| * STY: add spaces and endif documentationMatti Picus2016-10-012-57/+57
| |
* | Merge pull request #8125 from charris/deepcopy_objectCharles Harris2016-10-072-16/+63
|\ \ | | | | | | Rebase, BUG: Fixed deepcopy of F-order object arrays.
| * | STY: Make some C style fixups to methods.c.Charles Harris2016-10-071-11/+7
| | |
| * | BUG: Fixed deepcopy of F-order object arrays.pelson2016-10-072-15/+66
| | |
* | | BUG: Fix array printing with precision=0.wrwrwr2016-10-072-3/+16
| | | | | | | | | | | | Values rounding to zero were formatted with negative precision.
* | | BUG: core: add missing error check after PyLong_AsSsize_tPauli Virtanen2016-10-032-0/+9
|/ /
* | ENH: Allows contraction order optimization in einsum function.Daniel Smith2016-09-265-300/+1565
| |
* | Merge pull request #7980 from juliantaylor/avx-runtimeJulian Taylor2016-09-255-60/+144
|\ \ | | | | | | ENH: Add ability to runtime select ufunc loops, add AVX2 integer loops
| * | MAINT: add runtime check for AVX macrosJulian Taylor2016-09-242-0/+13
| | |
| * | ENH: add some AVX2 optimized integer ufunc loopsJulian Taylor2016-08-273-57/+90
| | | | | | | | | | | | Selected at runtime depending on CPU features.
| * | MAINT: add support for runtime selected ufunc SIMD loopsJulian Taylor2016-08-271-3/+21
| | |
| * | MAINT: add avx __builtin_cpu_supports and target attribute checksJulian Taylor2016-08-272-0/+20
| | | | | | | | | | | | NPY_CPU_SUPPORTS_AVX2 checks at runtime if AVX2 is supported
* | | Merge pull request #6054 from ahaldane/warnviewCharles Harris2016-09-247-34/+121
|\ \ \ | | | | | | | | WIP: MAINT: Add deprecation warning to views of multi-field indexes
| * | | MAINT: Add deprecation warning to multi-field views/assignmentAllan Haldane2016-09-247-34/+121
| | |/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | Behavior of multi-field indexes will change in 1.13: Multi-field indexes will return a view (not a copy) and assignment between structures with non-identical fieldnames occurs "by position" (not "by fieldname"): >>> a = zeros(10, dtype=[('x', 'i8'), ('y', 'i8'), ('z', 'i8')]) >>> a[['x', 'z']].view('i4') # Deprecation warning for multifield view >>> b = ones(10, dtype=[('y', 'i4'), ('x', 'f4')]) >>> a[:] = b # Deprecation warning for multifield assignment
* | | BUG: non-integers can end up in dtype offsetsAllan Haldane2016-09-223-3/+30
|/ / | | | | | | | | Fix is to convert offsets to python ints at dtype creation. Fixes #8059
* | Merge pull request #8051 from pslacerda/clear_snanJulian Taylor2016-09-193-2/+9
|\ \ | | | | | | Clear signaling NaN exceptions
| * | BUG: Clear signaling NaN exceptionsPedro Lacerda2016-09-153-2/+9
| | | | | | | | | | | | Closes gh-7838
* | | Merge pull request #8050 from mattip/pypy-fixesseberg2016-09-196-12/+48
|\ \ \ | |/ / |/| | ENH: a.resize(.., refcheck=True) is almost unusable on PyPy
| * | ENH: a.resize(.., refcheck=True) is almost unusable on PyPyMatti Picus2016-09-176-12/+48
| | |
* | | Merge pull request #8038 from gfyoung/rollaxis-err-msgStephan Hoyer2016-09-111-4/+4
|\ \ \ | |/ / |/| | MAINT: Update error message in rollaxis.
| * | DOC, MAINT: Update error message in rollaxis.gfyoung2016-09-111-4/+4
| | | | | | | | | | | | | | | | | | | | | Updates error message to properly reflect that the 'start' or 'axis' can be negative. Closes gh-7974.
* | | BUG: core: fix bug in NpyIter buffering with discontinuous arraysPauli Virtanen2016-09-113-1/+38
|/ / | | | | | | | | | | It is possible to skip copying an array to the buffer only if the reduction outer iterator is 1-dimensional --- the array may not be c-contiguous.
* | DOC: change version references from x.y to x.y.zPierre de Buyl2016-09-072-2/+2
| |
* | DOC: change version references from x.y to x.y.zPierre de Buyl2016-09-072-4/+4
| |
* | DOC: change Numpy to NumPy in dosctrings and commentsPierre de Buyl2016-09-0618-44/+44
| | | | | | | | The strings in error messages were left untouched
* | MAINT: Remove leftover imp module imports.Charles Harris2016-09-031-1/+0
| | | | | | | | | | | | | | There were two remaining imports of the deprecated imp module, neither of which seems to have been used beyond the import. Closes #5997.
* | ENH: Add stacklevel to all (or almost all) our function callsSebastian Berg2016-09-026-10/+13
| |
* | Merge pull request #7099 from seberg/suppressed_warningsCharles Harris2016-09-0212-119/+130
|\ \ | | | | | | Suppressed warnings
| * | TST: Silence warnings otherwise shown in release modeSebastian Berg2016-09-021-1/+2
| | | | | | | | | | | | | | | These are warnings, which when raised as an error for one reason or another are already silenced.
| * | TST: Use new warnings context manager in all testsSebastian Berg2016-09-029-85/+80
| | | | | | | | | | | | | | | | | | | | | | | | In some places, just remove aparently unnecessary filters. After this, all cases of ignore filters should be removed from the tests, making testing (even multiple runs) normally fully predictable.
| * | MAINT: Add missing teardown in deprecation test caseSebastian Berg2016-09-021-0/+2
| | |
| * | BUG: Suppress common NaT warningsSebastian Berg2016-09-022-34/+47
| | | | | | | | | | | | | | | | | | | | | | | | | | | Printing of datetime arrays used to cause warning due to comparison warnings in NaT. This is circumvented by using views to integer values. Part of this should be simplified when the deprecation is over. Also fixes a bug with non-native byteorder.
* | | TST: fix inplace case of alignment data generatorJulian Taylor2016-09-012-7/+8
| | | | | | | | | | | | | | | | | | | | | | | | Due to the lambdas used it didn't actually generate inplace cases. Fix a few tests that now break due to assuming that the cases are always out of place. Also update some numbers to make it more likely to find issues like loading from wrong array.
* | | ENH: add inplace cases to fast ufunc loop macrosJulian Taylor2016-09-012-33/+50
|/ / | | | | | | | | | | | | | | Both gcc and clang don't automatically specialize the inplace case, so add extra conditions to the loop macros to get the compilers to emit decent code. Without them inplace code ends up much slower than the out of place code.
* | Merge pull request #7823 from madphysicist/ma-bugsJulian Taylor2016-08-271-1/+1
|\ \ | |/ |/| BUG: Fixed masked array behavior for scalar inputs to np.ma.atleast_*d
| * BUG: Fixed masked array behavior for scalar inputs to np.ma.atleast_*dJoseph Fox-Rabinovitz2016-08-241-1/+1
| |
* | Merge pull request #7943 from theultimatecrouton/doc_fixCharles Harris2016-08-251-5/+2
|\ \ | |/ |/| DOC: #7927. Remove deprecated note for memmap relevant for Python <2.5
| * DOC: #7927. Remove deprecated note for memmap relevant for Python <2.5 [ci skip]Tom Bird2016-08-171-0/+2
| |
| * DOC: #7927. Remove deprecated note for memmap relevant for Python <2.5Tom Bird2016-08-171-5/+0
| |
* | BUG: cannot assign to tuple after usemattip2016-08-241-8/+6
|/
* Merge pull request #7937 from charris/fix-quicksort-for-bogus-comparisonJulian Taylor2016-08-172-4/+19
|\ | | | | BUG: Guard against buggy comparisons in generic quicksort.
View Lorry Controller Status