| Commit message (Collapse) | Author | Age | Files | Lines | |
|---|---|---|---|---|---|
| * | pass descriptor rather than null | Gregory Lee | 2021-06-24 | 1 | -5/+2 |
| | | |||||
| * | BUG: protect against access an attribute of a NULL pointer | Gregory Lee | 2021-06-24 | 1 | -2/+7 |
| | | | | | Have PyArray_GetCastSafety return -1 if from is NULL | ||||
| * | Merge pull request #19135 from ganesh-k13/BUG_13236_remove_complex_floor_divide | Charles Harris | 2021-06-23 | 8 | -116/+35 |
| |\ | | | | | BUG: Remove complex floor divide | ||||
| | * | BUG, TST: Determine complex types using typecodes | Ganesh Kathiresan | 2021-05-30 | 1 | -2/+2 |
| | | | |||||
| | * | BUG, TST: Removed complex floor division cases in typing:reveal | Ganesh Kathiresan | 2021-05-30 | 1 | -24/+0 |
| | | | |||||
| | * | BUG, TST: Removed complex floor division cases in typing:pass | Ganesh Kathiresan | 2021-05-30 | 1 | -22/+0 |
| | | | |||||
| | * | BUG: Removed `__*floordiv__` references | Ganesh Kathiresan | 2021-05-30 | 1 | -6/+0 |
| | | | |||||
| | * | TST: Changed TC to check for `TypeError` in floor divide | Ganesh Kathiresan | 2021-05-29 | 1 | -15/+27 |
| | | | |||||
| | * | TST: Changed TC to check for `TypeError` in floor divide | Added asserts for ↵ | Ganesh Kathiresan | 2021-05-29 | 1 | -9/+7 |
| | | | | | | | | | divmod and remainder | ||||
| | * | BUG: Removed complex type for floor_divide | Ganesh Kathiresan | 2021-05-29 | 1 | -1/+1 |
| | | | |||||
| | * | BUG: Removed floor_divide definition for complex type | Ganesh Kathiresan | 2021-05-29 | 1 | -36/+0 |
| | | | |||||
| | * | BUG: Removed floor_divide declaration for complex type | Ganesh Kathiresan | 2021-05-29 | 1 | -3/+0 |
| | | | |||||
| * | | Merge pull request #19225 from asmeurer/matrix_rank-doc-fix | Charles Harris | 2021-06-23 | 2 | -23/+23 |
| |\ \ | | | | | | | DOC: Fix some inconsistencies in the docstring of matrix_rank | ||||
| | * | | Fix the matrix_rank signature in the stub file | Aaron Meurer | 2021-06-14 | 1 | -1/+1 |
| | | | | |||||
| | * | | Merge branch 'main' into matrix_rank-doc-fix | Aaron Meurer | 2021-06-14 | 510 | -15519/+44645 |
| | |\ \ | |||||
| | * | | | Fix some inconsistencies in the docstring of matrix_rank | Aaron Meurer | 2021-06-11 | 1 | -22/+22 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | - Rename the argument from M to A, since M already refers to the number of rows in the docstring. - Fix the documentation for the formula for the default value of tol. It only uses the maximum of the number of rows and columns, not the max of all dimensions. The code here was correct; only the documentation was wrong. | ||||
| * | | | | Merge pull request #19286 from BvB93/type_check | Charles Harris | 2021-06-23 | 4 | -17/+331 |
| |\ \ \ \ | | | | | | | | | | | ENH: Add annotations for `np.lib.type_check` | ||||
| | * | | | | TST: Add typing tests for `np.lib.type_check` | Bas van Beek | 2021-06-20 | 2 | -0/+86 |
| | | | | | | |||||
| | * | | | | MAINT: Fixed an issue with the return-dtype of `ndarray.real` and `imag | Bas van Beek | 2021-06-20 | 1 | -2/+14 |
| | | | | | | | | | | | | | | | | | | | | | The latter two would previously return complex arrays if the initial array was also complex | ||||
| | * | | | | ENH: Add annotations for `np.lib.type_check` | Bas van Beek | 2021-06-20 | 1 | -15/+231 |
| | | | | | | |||||
| * | | | | | Merge pull request #19298 from BvB93/c_intp_docs | Matti Picus | 2021-06-22 | 1 | -5/+4 |
| |\ \ \ \ \ | | | | | | | | | | | | | DOC: Add documentation for `np.ctypeslib.c_intp` | ||||
| | * | | | | | DOC: Add documentation for `np.ctypeslib.c_intp` | Bas van Beek | 2021-06-22 | 1 | -5/+4 |
| | | | | | | | |||||
| * | | | | | | MAINT: Add simple tuple creation helper and use it | Sebastian Berg | 2021-06-21 | 4 | -26/+35 |
| |/ / / / / | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The new dispatching code requires a few more places where a similar tuple is created, so add the helper and use it in three places only. But more will be added (especially more that require the NULL -> None logic). | ||||
| * | | | | | Merge pull request #19257 from seberg/main-ufunc-move-prep-and-resolution | Matti Picus | 2021-06-21 | 3 | -166/+116 |
| |\ \ \ \ \ | | | | | | | | | | | | | MAINT: Move array-prep and type resolution to earlier | ||||
| | * | | | | | Update numpy/core/src/umath/ufunc_object.c | Sebastian Berg | 2021-06-20 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | Co-authored-by: Matti Picus <matti.picus@gmail.com> | ||||
| | * | | | | | MAINT: Move array-prep and type resolution to earlier | Sebastian Berg | 2021-06-17 | 3 | -166/+116 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This also fixes a bug in the masked handling of array prep, that was seems to have been simply completely broken. Note that the long term goal is to unify the masked and non-masked, but that is tricky right now due to the different signatures. | ||||
| * | | | | | | Merge pull request #19276 from seberg/issue-19153 | Matti Picus | 2021-06-21 | 5 | -53/+38 |
| |\ \ \ \ \ \ | | | | | | | | | | | | | | | BUG: Fix `arr.flat.index` for large arrays and big-endian machines | ||||
| | * | | | | | | BUG: Fix `arr.flat.index` for large or big-endian machines | Sebastian Berg | 2021-06-18 | 2 | -6/+21 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The type read when exposing was previously int, but has to be intp. this would only be visible for >2**31 elements, but is also visible on big-endian machines. Closes gh-19153 | ||||
| | * | | | | | | MAINT: Introduce `PyArray_PyIntFromIntp` to remove ifdef's | Sebastian Berg | 2021-06-18 | 4 | -47/+17 |
| | | |/ / / / | |/| | | | | |||||
| * | | | | | | Merge pull request #19275 from BvB93/dtypes | Charles Harris | 2021-06-19 | 2 | -38/+50 |
| |\ \ \ \ \ \ | | | | | | | | | | | | | | | MAINT: Misc typing maintenance for `np.dtype` | ||||
| | * | | | | | | ENH: Allow `dtype.__getitem__` to take any object implementing the ↵ | Bas van Beek | 2021-06-18 | 1 | -1/+1 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | `__index__` protocol | ||||
| | * | | | | | | TST: Update the dtype typing tests | Bas van Beek | 2021-06-18 | 1 | -21/+30 |
| | | | | | | | | |||||
| | * | | | | | | MAINT: Set the return-dtype of `dtype.base` and `dtype.subdtype` to `Any` | Bas van Beek | 2021-06-18 | 1 | -2/+2 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Whether the return dtype is equivalent to the initial dtype depends on whether the initial dtype is structured or if it has fields of a fixed size | ||||
| | * | | | | | | MAINT: Fixed the signature of `dtype.__mul__` | Bas van Beek | 2021-06-18 | 1 | -6/+11 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | * Multiplication by 0 does not return `None` * Multiplication of a flexible dtype returns a flexible dtype; not just void * In principle any object implementing the `__index__` protocol is a valid multiplication value; not just integers. | ||||
| | * | | | | | | MAINT: Removed a duplicate attribute | Bas van Beek | 2021-06-18 | 1 | -2/+0 |
| | | | | | | | | |||||
| | * | | | | | | ENH: Use the concrete `MappingProxyType` class over the `Mapping` ABC | Bas van Beek | 2021-06-18 | 1 | -3/+3 |
| | | | | | | | | |||||
| | * | | | | | | STY: Use the pipe operator to represent Unions in `np.dtype` | Bas van Beek | 2021-06-18 | 1 | -8/+8 |
| | |/ / / / / | |||||
| * | | | | | | BUG: Fix refcount leak in ResultType | Sebastian Berg | 2021-06-18 | 1 | -10/+13 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | This slightly reorganizes the error path handling (duplicating the freeing for the non-error path). It just seemed a bit clearer. | ||||
| * | | | | | | BUG: Add missing DECREF in new path | Sebastian Berg | 2021-06-18 | 1 | -0/+1 |
| |/ / / / / | | | | | | | | | | | | | | | | Pretty, harmless reference count leak (the method is a singleton) | ||||
| * | | | | | Merge pull request #19258 from seberg/maint-ufunc-refactor-iterator-loop | Sebastian Berg | 2021-06-17 | 1 | -55/+61 |
| |\ \ \ \ \ | | | | | | | | | | | | | MAINT: Refactor and simplify the main ufunc iterator loop code | ||||
| | * | | | | | MAINT: Refactor and simplify the main ufunc iterator loop code | Sebastian Berg | 2021-06-17 | 1 | -55/+61 |
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | Simple cleanups to the main ufunc loop code in preparation of larger ones. | ||||
| * | | | | | | Merge pull request #19228 from ganesh-k13/BUG_19187_invalid_dtypes_comparison | Matti Picus | 2021-06-17 | 3 | -3/+21 |
| |\ \ \ \ \ \ | |_|/ / / / |/| | | | | | BUG: Invalid dtypes comparison should not raise `TypeError` | ||||
| | * | | | | | BUG: Removed typing for == and != in dtypes | Ganesh Kathiresan | 2021-06-15 | 1 | -2/+0 |
| | | | | | | | |||||
| | * | | | | | TST: Check invalid dtypes for equality | Check TypeError for comparisions | Ganesh Kathiresan | 2021-06-15 | 1 | -9/+14 |
| | | | | | | | |||||
| | * | | | | | BUG: Return NotImplemented for unrecognized dtypes | Ganesh Kathiresan | 2021-06-15 | 1 | -14/+1 |
| | | | | | | | |||||
| | * | | | | | TST: Check invalid dtypes comparison | Ganesh Kathiresan | 2021-06-12 | 1 | -0/+13 |
| | | | | | | | |||||
| | * | | | | | BUG: Remove TypeError on invalid dtypes comparison | Ganesh Kathiresan | 2021-06-12 | 1 | -1/+16 |
| | | | | | | | |||||
| * | | | | | | Merge pull request #19254 from seberg/maint-ufunc-output-wrapping | Charles Harris | 2021-06-17 | 1 | -56/+80 |
| |\ \ \ \ \ \ | | | | | | | | | | | | | | | MAINT: Refactor output ufunc wrapping logic | ||||
| | * | | | | | | MAINT: Refactor output ufunc wrapping logic | Sebastian Berg | 2021-06-15 | 1 | -56/+80 |
| | | |/ / / / | |/| | | | | | | | | | | | | | | | | | | | | | | This replaces the output wrapping logic to work towards (and simplify) the big ufunc refactor for NEP 43. | ||||
| * | | | | | | Merge pull request #19251 from pearu/gh-18431-string_from_pyobj-2 | Charles Harris | 2021-06-17 | 5 | -68/+310 |
| |\ \ \ \ \ \ | | | | | | | | | | | | | | | BUG: revise string_from_pyobj/try_pyarr_from_string with respect to malloc and copy (the second round) | ||||
 |
index : delta/python-packages/numpy.git | |
| github.com: numpy/numpy.git |
| summaryrefslogtreecommitdiff |