summaryrefslogtreecommitdiff
path: root/numpy
Commit message (Collapse)AuthorAgeFilesLines
* BUG: Keep ignoring most errors during array-protocol lookupSebastian Berg2021-01-202-7/+24
| | | | | | | | | | | | | | | | | | Closes (the later point) in gh-17965 and reverts parts of gh-17817. Shapely did rely on being able to raise a NotImplementedError which then got ignored in the attribute lookup. Arguably, this should probably just raise an AttributeError to achieve that behaviour, but it means we can't just rip the band-aid off here. Since 1.20 is practically released, just reverting most of the change (leaving only recursion and memory error which are both arguably pretty fatal). Ignoring most errors should be deprecated (and I am happy to do so), but it is not important enough for 1.20 or very important in itself. Closes gh-17965
* Merge pull request #18184 from pearu/17797Charles Harris2021-01-196-11/+58
|\ | | | | BUG: Fix f2py bugs when wrapping F90 subroutines.
| * Applied reviewers comments: add identPearu Peterson2021-01-191-9/+10
| |
| * Apply reviewer comments.Pearu Peterson2021-01-192-6/+4
| |
| * Apply reviewers comments.Pearu Peterson2021-01-192-12/+12
| |
| * Add test for gh17797.Pearu Peterson2021-01-184-6/+47
| |
| * BUG: f2py specific __user__ modules should not be used in Fortran sources to ↵Pearu Peterson2021-01-181-3/+6
| | | | | | | | be compiled. Closes #17797.
| * BUG: subroutines using assumed shape arrays must be treated as F90 codes. ↵Pearu Peterson2021-01-181-2/+6
| | | | | | | | Don't split inline comments.
| * BUG: Fix detecting Fortran 90+ source format.Pearu Peterson2021-01-181-1/+1
| |
* | Merge pull request #18174 from BvB93/covariantCharles Harris2021-01-195-160/+169
|\ \ | | | | | | MAINT: Changed the `NBitBase` variancy in `number` from co- to invariant
| * | DOC: Added a missing `Union` importBas van Beek2021-01-191-1/+1
| | |
| * | TST: Added a tests for `number` invariancyBas van Beek2021-01-191-0/+6
| | |
| * | MAINT: Adjusted the return-type of `number` ops to `Any`Bas van Beek2021-01-191-1/+1
| | |
| * | DOC,TST: Update the `NBitBase` exampleBas van Beek2021-01-192-4/+7
| | |
| * | MAINT: Changed the `NBitBase` variancy in `number` from co- to invariantBas van Beek2021-01-192-154/+154
| | |
* | | Merge pull request #18193 from jeertmans/docs/random/typoCharles Harris2021-01-191-1/+1
|\ \ \ | | | | | | | | MAINT: Fix typo in docstring example
| * | | Fixing typo in docstringEertmans2021-01-191-1/+1
| |/ /
* | | BUG: threads.h existence test requires GLIBC > 2.12. (#18180)Pearu Peterson2021-01-191-1/+6
|/ / | | | | | | | | * defined(__STDC_NO_THREADS__) can be trusted only when using glibc > 2.12 * Add __GLIBC__ defined check.
* | Merge pull request #18191 from pmav99/panos/fix_18190Charles Harris2021-01-191-2/+2
|\ \ | | | | | | STY: Use explicit reexports for numpy.typing objects
| * | STY: Use explicit reexports for numpy.typing objectspmav992021-01-191-2/+2
| |/ | | | | | | | | | | | | | | | | `mypy --strict` is disabling `implicit_reexport`. Consequently, when we try to import `ArrayLike` and `DTypeLike` from `numpy.typing`, mypy throws an error. With this commit we add explicit "reexports" for these two objects. Fixes #18190
* | MAINT: Simplify a unionBas van Beek2021-01-183-22/+21
| | | | | | | | `_BoolLike_co` is already a subtype of `_IntLike_co`, no nead the use an explicit union here
* | STY: Cleaned up the `numpy.typing` importsBas van Beek2021-01-181-4/+13
| |
* | MAINT: Renamed `_ArrayLike<X>` to `_ArrayLike<X>_co`Bas van Beek2021-01-182-23/+23
| |
* | MAINT: Renamed `<X>Like` to `<X>Like_co`Bas van Beek2021-01-186-115/+115
|/
* Merge pull request #18181 from pearu/pearu/17859Matti Picus2021-01-182-0/+33
|\ | | | | ENH: [f2py] Add external attribute support.
| * ENH: [f2py] Add external attribute support. Closes #17859Pearu Peterson2021-01-172-0/+33
| |
* | Merge pull request #18169 from devnexen/arm_fbsd_updMatti Picus2021-01-181-1/+24
|\ \ | |/ |/| ENH: cpu features detection implementation on FreeBSD ARM
| * ENH: cpu features detection implementation on FreeBSD ARMDavid Carlier2021-01-171-1/+24
| |
* | TST: Clear the mypy cache before running any typing testsBas van Beek2021-01-151-0/+10
|/
* Merge pull request #18157 from alexhenrie/no_castable_outputSebastian Berg2021-01-141-1/+1
|\ | | | | BUG: Initialize value of no_castable_output used in ufunc_loop_matches
| * BUG: Initialize value of no_castable_output used in ufunc_loop_matchesAlex Henrie2021-01-121-1/+1
| |
* | Use sinus based formula for `chebpts1`Weh Andreas2021-01-141-2/+2
| |
* | MAINT: Remove redundant null check before freeAlex Henrie2021-01-121-3/+1
|/
* Merge pull request #18152 from seberg/fix-string-object-promotionCharles Harris2021-01-112-35/+38
|\ | | | | BUG: Promotion between strings and objects was assymetric
| * BUG: Promotion between strings and objects was assymetricSebastian Berg2021-01-112-35/+38
| | | | | | | | | | | | | | | | | | | | After an unrelated fix, the logic for string and object promotion was incorrect briefly, this fixes it to be correct (symmetric). Before, string and unicode would return that `string.__common_dtype(object)` is in fact `string`, which is of course incorrect. (since `object.__common_dtype__(other)` always returns `object` this depended on the order, and the NumPy tests apparently did only test the opposite direction (or nothing).
* | Merge pull request #18136 from ↵Sebastian Berg2021-01-111-0/+5
|\ \ | |/ |/| | | | | Iamsoto/add_keyboard_interrupts_for_array_creation ENH: Adding keyboard interrupt support for array creation
| * ENH: keyboard interrupt for large array creationiamsoto2021-01-081-0/+5
| |
* | Merge pull request #18104 from ↵Charles Harris2021-01-102-43/+42
|\ \ | | | | | | | | | | | | DWesl/change-tests-with-loops-to-parametrized-tests TST: Turn some tests with loops into parametrized tests.
| * | TST: Change big parametrized loop to nested small parametrized loops.DWesl2021-01-032-28/+9
| | | | | | | | | | | | | | | | | | | | | Suggested by charris for clarity. Since I'm not marking some tests xfail based on the values of all parameters, this looks nicer.
| * | TST: Turn some tests with loos into parametrized tests.DWesl2021-01-012-43/+61
| | | | | | | | | | | | | | | | | | | | | I wanted to mark only some parts of the loops as xfail for another PR. That part of the PR probably won't make it into numpy, but I think parametrized tests give better information on failure than tests with loops do, so I'm submitting these here.
* | | Merge pull request #18118 from seiko2plus/issue_18106Matti Picus2021-01-094-48/+72
|\ \ \ | | | | | | | | BUG, MAINT: improve avx512 mask logical operations
| * | | BUG, MAINT: improve avx512 mask logical operationsSayed Adel2021-01-052-47/+55
| | | | | | | | | | | | | | | | | | | | It also fixes conversion warning between `__mmask16` and `__mmask8` on msvc2019 when logical intrinsics of AVX512DQ are available.
| * | | BLD: Add extra check for AVX512DQ mask conversionsSayed Adel2021-01-052-1/+17
| | | |
* | | | Merge pull request #17863 from seberg/restructure-cast-loops-masterMatti Picus2021-01-0915-70/+2052
|\ \ \ \ | | | | | | | | | | MAINT: Implement new casting loops based on NEP 42 and 43
| * | | | MAINT: Implement new casting loops based on NEP 42 and 43Sebastian Berg2021-01-0814-60/+2046
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | switch to using new casting as default to run full CI Initialize casting to silence compile warnings only... BUG: As my comment even said, out_needs_api is caller-initialized so I shouldn't init it to 0, it might disable api again. Simplify tests again, it seems unnecessar to go this complex Simplify `#if` as the variable should always be defined TST: Add more tests (and some related fixes) This fixes a few small things (i.e. the decref loop can be NULL and checking for writeable was incorrect in the private API). Mostly adds new tests for some internal API as well as for the copy of the auxiliar data. The nditer test however also is a heavyweight tests for the general multistep casting
| * | | | MAINT: Move modref definition into setup.pySebastian Berg2021-01-072-10/+6
| | |_|/ | |/| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | It seems the only reason this did not cause problems before was that the test checking for C warnings only ran on a run where these were always 0 and thus undefined (and so were not redefined later). The `#ifndef` would have to happen at a later time apparently, so just avoid it.
* | | | Merge pull request #18109 from PhanatosZou/chain-exceptionsSebastian Berg2021-01-081-1/+1
|\ \ \ \ | |/ / / |/| | | MAINT: Fix exception cause in mingw32ccompiler.py
| * | | fix exception cause in mingw32ccompiler.pyPhanatosZou2021-01-031-1/+1
| |/ /
* | | ENH, SIMD: Ditching the old CPU dispatcher(Exp & Log) (#18101)Sayed Adel2021-01-078-1247/+1347
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | The second patch in a series of pull-requests aims to facilitate the migration process to our new SIMD interface(NPYV). It is basically a process that focuses on getting rid of the main umath SIMD source `simd.inc`, which contains almost all SIMD kernels, by splitting it into several dispatch-able sources without changing the base code, which facilitates the review process during the move to NPYV(universal intrinsics). In this patch, we have moved the the following raw SIMD loops to the new dispatcher: - FLOAT_exp, DOUBLE_exp - FLOAT_log, DOUBLE_log - FLOAT_frexp, DOUBLE_frexp - FLOAT_ldexp, DOUBLE_ldexp
* | | DEP: Deprecate `data_type.dtype` if attribute is not already a dtype (#13578)Sebastian Berg2021-01-065-49/+80
| | | | | | | | | | | | | | | | | | * DEP: Deprecate `data_type.dtype` if attribute is not already a dtype After the deprecation, a recursive lookup for `.dtype` will not be possible, since `.dtype` has to be a dtype instance.
View Lorry Controller Status