summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
* MAINT,BUG: Fix mtrand for Cython 0.27.Charles Harris2017-09-261-1/+11
| | | | | | | The `import_array()` macro, that defined a C code block that included a return, was not handled correctly. The fix here is to cdef a replacement `import_array` function with a defined error return. The new function is a slight variation of the corresponding function defined by Cython.
* BUG: ensure consistent result dtype of count_nonzeroJulian Taylor2017-09-242-13/+17
| | | | | | | The slowpath using apply_along_axis for size 1 axis did not ensure that the dtype is intp like all other paths. This caused inconsistent dtypes on windows where the default integer type is int32. Closes gh-9468
* Merge pull request #9766 from charris/backport-9294Charles Harris2017-09-242-1/+9
|\ | | | | BUG: Fixes histogram monotonicity check for unsigned bin values
| * allow non-strictly increasing binsBrandon Carter2017-09-241-1/+1
| |
| * minor change to the logicBrandon Carter2017-09-241-1/+1
| |
| * BUG: fixes unsigned bins monotonicity check, see #9222Brandon Carter2017-09-241-1/+1
| |
| * TST: add test for unsigned bins monotonicity check, see #9222Brandon Carter2017-09-241-0/+8
| |
* | Merge pull request #9765 from charris/backport-9300Charles Harris2017-09-244-39/+53
|\ \ | |/ |/| BUG: PyArray_CountNonzero does not check for exceptions
| * ENH: release GIL in CountNonzero trivial loopAllan Haldane2017-09-241-6/+18
| |
| * MAINT: Simplify array_contains using PyArray_AnyAllan Haldane2017-09-242-76/+9
| |
| * MAINT: array_any_nonzero now uses new iterator APIAllan Haldane2017-09-242-21/+59
| |
| * BUG: PyArray_CountNonzero does not check for exceptionsAllan Haldane2017-09-243-2/+33
| | | | | | | | Fixes #9295
* | BUG: Ensure `_npy_scaled_cexp{,f,l}` is defined when needed.Charles Harris2017-09-241-1/+7
|/ | | | | | The `_npy_scaled_cexp{,f,l}` functions were previously only defined when the npy_cexp fallback functions were needed, but they are also called by the npy_csinh and npy_ccosh functions.
* Merge pull request #9747 from charris/backport-9724Charles Harris2017-09-222-2/+4
|\ | | | | BUG: adjust gfortran version search regex
| * BUG: adjust gfortran version search regexJonathan Helmus2017-09-222-2/+4
| | | | | | | | | | | | | | Adjust the gfortran regex to require a '.' to be present in the version number. This avoids matching a hash or text containing a '-' such as the version string from the compilers produced by crosstool-NG. For example: GNU Fortran (crosstool-NG 8a21ab48) 7.2.0
* | Merge pull request #9746 from charris/backport-9624Charles Harris2017-09-222-1/+13
|\ \ | | | | | | BUG: memory leak in np.dot of size 0
| * | TST: test for leak in np.dot of size 0 arraysMichael Lamparski2017-09-221-0/+11
| | |
| * | BUG: memory leak in np.dot of size 0Michael Lamparski2017-09-221-1/+2
| |/
* | Merge pull request #9745 from charris/backport-9682Charles Harris2017-09-226-49/+49
|\ \ | | | | | | DOC: Add whitespace after "versionadded::" directive so it actually displays in the docs
| * | DOC: Fix inline "math" directives [skip ci]Michael Seifert2017-09-226-6/+6
| | |
| * | DOC: Add whitespace after "versionadded::" directiveMichael Seifert2017-09-225-43/+43
| |/ | | | | | | | | | | so it actually displays in the docs [skip ci]
* | BUG: Check for exception in sort functions, add testsmattip2017-09-223-8/+31
|/
* Merge pull request #9732 from charris/fix-scalar-elisionCharles Harris2017-09-222-2/+13
|\ | | | | BUG: Make scalar function elision check if temp is writeable.
| * MAINT: Use PyArray_CHKFLAGS in more places.Charles Harris2017-09-221-2/+2
| | | | | | | | This is more direct than `PyArray_FLAGS(a) & AFLAG`.
| * BUG: Make scalar function elision check writeable.Charles Harris2017-09-212-0/+11
| | | | | | | | | | | | | | Add checks for writeable and updateifcopy in the can_elide_temp_unary function. Closes #9679.
* | Merge pull request #9742 from charris/backport-pad-fixesCharles Harris2017-09-222-0/+19
|\ \ | | | | | | BUG: Fix np.pad for CVE-2017-12852
| * | BUG: fix padding an empty array in reflect mode.Iryna Shcherbina2017-09-222-3/+15
| | | | | | | | | | | | Check that axes with non-zero padding are non-empty.
| * | BUG: fix infinite loop when creating np.pad on an empty arrayIryna Shcherbina2017-09-222-0/+7
| |/
* | Merge pull request #9736 from eric-wieser/gradient-fixCharles Harris2017-09-212-8/+20
|\ \ | |/ |/| BUG: various fixes to np.gradient
| * MAINT: Use clearer variableEric Wieser2017-07-131-1/+1
| |
| * BUG: Use np.ndim not asarray, to allow duck-typesEric Wieser2017-07-131-7/+6
| |
| * BUG: Only allow 1d distance arraysEric Wieser2017-07-132-1/+7
| | | | | | | | | | | | 2d arrays would work, but in unpredictable and undocumented ways. This at least makes gh-9401 give a better error message.
| * BUG: Allow 0d arrays instead of scalars in gradientEric Wieser2017-07-132-8/+15
| | | | | | | | This fixes gh-8292
* | BUG: Add hypot and cabs functions to WIN32 blacklist.Charles Harris2017-08-171-0/+13
| | | | | | | | | | MSVC 32 bit generates code for _hypot and cabs that changes the fpu precision mode, enabling ieee extended precision.
* | BLD: remove -xhost flag from IntelFCompiler. Closes gh-9042.Ralf Gommers2017-08-141-1/+1
| | | | | | | | | | | | | | Note that this was discussed extensively in gh-7287, but removing this -xhost flag was missed in the PR that closed that issue. [ci skip]
* | Merge pull request #9556 from charris/backport-9469Charles Harris2017-08-146-42/+143
|\ \ | | | | | | BUG: Fix true_divide when dtype=np.float64 specified.
| * | BUG: Fix true_divide when dtype=np.float64 specified.Charles Harris2017-08-146-42/+143
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The problem here was due to the signature matching when dtype is specified. A signature matches when 1) The output dtype matches the specified dtype 2) The input arguments can be cast to the signatures dtypes using casting="same_kind" The signature 'bb->d' was present for true_divide, consequently when dtype=np.float64 was specified, all integer inputs were cast to int8. The solution adopted here was to remove all signatures/loops for true_divide that had integer inputs. Lesser measures broke a lot of existing code that relied on the behavior for other functions Closes #3484. Again.
* | | DOC: Fix some rst markup in numpy/doc/basics.py.Charles Harris2017-08-141-3/+3
|/ / | | | | | | [ci skip]
* | Merge pull request #9555 from charris/backport-9554Charles Harris2017-08-142-4/+8
|\ \ | | | | | | BUG: fix regression in 1.13.x in distutils.mingw32ccompiler.
| * | MAINT: Suppress warning in py3compat test.Charles Harris2017-08-141-3/+7
| | |
| * | BUG: fix regression in 1.13.x in distutils.mingw32ccompiler.Ralf Gommers2017-08-141-1/+1
| | | | | | | | | | | | | | | | | | Issue was introduced in gh-8454. Thanks to @jennirinker for pointing out the issue and fix. Closes gh-9553.
* | | Merge pull request #9390 from eric-wieser/poly1d-fixes-fixes-fixes-fixesCharles Harris2017-07-112-11/+17
|\ \ \ | |/ / |/| | BUG: Return the coefficients array directly
| * | BUG: Return the coefficients array directlyEric Wieser2017-07-082-11/+17
| | | | | | | | | | | | | | | | | | | | | Turns out that this was relied upon downstream We also add a setter for coeffs, so that augmented assignment does not both change state and raise an exception suggesting state could not be changed.
* | | Merge pull request #9378 from charris/backport-9323Charles Harris2017-07-061-0/+1
|\ \ \ | | | | | | | | DOC: Add $PARAMS to the isnat docstring
| * | | DOC: Add $PARAMS to the isnat docstringAndras Deak2017-07-061-0/+1
| | | |
* | | | Merge pull request #9377 from charris/backport-9320Charles Harris2017-07-061-9/+8
|\ \ \ \ | | | | | | | | | | DOC: Use x1 and x2 in the heaviside docstring.
| * | | | DOC: Use x1 and x2 in the heaviside docstring.Warren Weckesser2017-07-061-9/+8
| |/ / / | | | | | | | | | | | | | | | | | | | | Also remove a duplicate $PARAMS. Closes gh-9314.
* | | | MAINT: Use a for loop to traverse the linked listEric Wieser2017-07-061-4/+3
| | | | | | | | | | | | | | | | This avoids bugs like gh-9351 being introduced later
* | | | BUG: Prevent hangs traversing ufunc userloop linked listsEric Wieser2017-07-062-8/+12
|/ / / | | | | | | | | | Fixes gh-9351
* | | Merge pull request #9373 from charris/backport-9364Charles Harris2017-07-061-2/+1
|\ \ \ | | | | | | | | BUG: ')' is printed at the end pointer of the buffer in numpy.f2py.
View Lorry Controller Status