summaryrefslogtreecommitdiff
Commit message (Collapse)AuthorAgeFilesLines
* REL: Release.v1.8.0rc1Charles Harris2013-09-291-2/+2
|
* Merge pull request #3835 from charris/backport-1.8-3832Charles Harris2013-09-294-6/+6
|\ | | | | Backport 1.8 3832
| * BUG: fix small issues found with cppcheckJulian Taylor2013-09-294-6/+6
|/ | | | | | uninitialized use in _strided_to_strided_one_to_n_with_finish wrong error NULL checks in PyArray_MapIterNew and einsum_sub_op_from_str double free and memleak in umath_linalg.c.src init_@lapack_func@
* Merge pull request #3828 from charris/backport-1.8-3827Charles Harris2013-09-282-5/+35
|\ | | | | Backport 1.8 3827
| * BUG: core: consider both C and F order buffers as contiguousPauli Virtanen2013-09-282-5/+35
|/
* Merge pull request #3824 from charris/backport-1.8-3821Charles Harris2013-09-281-1/+3
|\ | | | | Backport 1.8 3821
| * Do not fail with NameError if SandboxViolation is raisedJason Madden2013-09-281-1/+3
| |
* | Merge pull request #3819 from charris/backport-1.8-3818Ralf Gommers2013-09-281-0/+2
|\ \ | |/ |/| Backport 1.8 3818
| * Fix platform detection for intel platform where get_platform() returns a ↵Alex Barth2013-09-281-0/+2
|/ | | | string ending in x86_64, for instance 'macosx-10.8-x86_64'.
* Merge pull request #3810 from charris/backport-1.8-3788Charles Harris2013-09-273-10/+21
|\ | | | | Backport 1.8 3788
| * BUG: INCREF before DECREF when copying object arrays.Sebastian Berg2013-09-273-10/+21
| | | | | | | | | | | | | | | | | | | | | | Only the change in dtype_transfer actually seems to be related to the simple np.copyto(a, a) bug, when a is a scalar object array. It seemed however safer to me, to generally turn around the logic, even if it may not be necessary and in one place requires an extra temporary variable. Closes gh-3787
* | Merge pull request #3814 from charris/backport-1.8-3806Charles Harris2013-09-272-3/+37
|\ \ | | | | | | Backport 1.8 3806
| * | [fix] If we fail to raise an error, at least dump something to the consoleNathaniel J. Smith2013-09-272-8/+25
| | |
| * | [fix] Check for errors in PyArray_CanCastTypeto DEPRECATENathaniel J. Smith2013-09-272-3/+20
| |/ | | | | | | | | | | | | | | | | If a user had set warnings to raise errors, then this DEPRECATE would leave us with an unpropagated exception and cause havoc downstream. Unfortunately there is no way to propagate an exception from here, so we just have to throw it away :-(. But this is still better than the alternative...
* | Merge pull request #3809 from charris/backport-1.8-3800Charles Harris2013-09-272-37/+48
|\ \ | | | | | | Backport 1.8 3800
| * | BUG: fix hashing of long integers under python3Julian Taylor2013-09-272-37/+48
| |/ | | | | | | | | | | | | | | | | python3 long_hash is more complex than int_hash so instead of copying into numpy call it via the Python capi. Same for long long for wich the numpy hash function is not correct with python 2.7 on i386. Will be slower but doesn't need adapting each timy python changes. closes #3793
* | Merge pull request #3811 from charris/backport-1.8-3802Charles Harris2013-09-272-1/+12
|\ \ | |/ |/| Backport 1.8 3802
| * BUG: do a boolean check on casting bool to other typesJulian Taylor2013-09-272-1/+12
|/ | | | | | avoids some issues of booleans coming from external buffers that may not only contain 0 or 1. closes #3801.
* Merge pull request #3805 from charris/fix-mlab-importCharles Harris2013-09-272-22/+27
|\ | | | | Fix mlab import
| * BUG: Fix failed import of mlab.Charles Harris2013-09-272-22/+27
| | | | | | | | | | | | | | | | | | | | | | | | | | Fix imports in the numpy/oldnumeric/mlab.py module. The problem was that the module used imports of the form from . import compat which do not work as '.' is apparently undefined when the module import is attempted. The fix is ugly, but it works and the oldnumeric module is going away in 1.9. Closes #3803.
* | Merge pull request #3791 from juliantaylor/simd-gen-privateCharles Harris2013-09-233-14/+14
|\ \ | |/ |/| TST: mark gen_alignment_data private
| * TST: mark gen_alignment_data privateJulian Taylor2013-09-243-14/+14
| | | | | | | | | | it not very useful outside numpy and there might be changes required in future e.g. to add min_size.
* | Merge pull request #3784 from charris/backport-1.8-3766Charles Harris2013-09-232-2/+15
|\ \ | | | | | | Backport 1.8 3766
| * | BLD: Backport Ralf's gh-3766 fixes to the bento build scripts.Charles Harris2013-09-232-2/+15
| | | | | | | | | | | | | | | | | | The original PR included fixes for ufunc_override that are not needed for 1.8, so I just copied the files numpy/core/bento.info and numpy/core/bscript from master.
* | | Merge pull request #3785 from charris/backport-1.8-3781Charles Harris2013-09-231-5/+5
|\ \ \ | | | | | | | | Backport 1.8 3781
| * | | TST: wrong indent in multiarray testsSebastian Berg2013-09-231-5/+5
| |/ /
* | | Merge pull request #3786 from charris/backport-1.8-3773Charles Harris2013-09-231-0/+1
|\ \ \ | |/ / |/| | Backport 1.8 3773
| * | ENH: Add gfortran to travis.yml configuration to enable f2py.Charles Harris2013-09-231-0/+1
|/ / | | | | | | The f2py tests would not run without gfortran.
* | Merge pull request #3780 from juliantaylor/sse-configure-check-1.8.xCharles Harris2013-09-226-20/+38
|\ \ | | | | | | Backport PR 3772
| * | BUG: make checking for sse intrinsics more robustJulian Taylor2013-09-226-20/+38
| |/ | | | | | | | | | | | | check for two intrinsics from [ex]mmintrin.h instead of only checking the existance of the headers. E.g. mingw 4.2 emmintrin.h can be included even if SSE2 is disabled. closes #3760
* | Merge pull request #3761 from ContinuumIO/2890_fixCharles Harris2013-09-192-7/+21
|\ \ | | | | | | Fix for PR #3491
| * | BUG: Fix issue with PR #3491Jay Bourque2013-09-182-7/+21
|/ / | | | | | | Don't convert unicode string to ascii string before getting byte length. Also, get length of non string type as ascii or unicode depending on output string type.
* | MAINT: update release scripts for building OS X binaries.Ralf Gommers2013-09-163-12/+4
| | | | | | | | Works with Python 3.x now that bdist_mpkg is ported (thanks @matthew-brett).
* | Merge pull request #3749 from charris/backport-1.8-3747Charles Harris2013-09-144-34/+47
|\ \ | | | | | | Backport 1.8 3747
| * | BUG: use correct type for hash valuesChristoph Gohlke2013-09-144-34/+47
| | | | | | | | | | | | As of Python 3.2 hash values are of type Py_hash_t. Previous versions use C long.
* | | Merge pull request #3745 from charris/move-nansum-warning-in-1.8Charles Harris2013-09-142-4/+5
|\ \ \ | | | | | | | | Move nansum warning in 1.8
| * | | TST: Test that FutureWarning is not raised by nansum if not all-nan.Charles Harris2013-09-141-0/+3
| | | |
| * | | MAINT: Move FutureWarning in nansum.Charles Harris2013-09-141-4/+2
| |/ / | | | | | | | | | | | | | | | Currently a FutureWarning about the sum of empty slices changing to zero is always raised. Instead make it depend on the actual occurrence of an empty slice.
* | | Merge pull request #3744 from juliantaylor/doc-update-1.8.xCharles Harris2013-09-145-1/+10
|\ \ \ | |/ / |/| | Backport PR 3741
| * | DOC: add isclose and partition to reference and link some docsJulian Taylor2013-09-145-1/+10
| |/ | | | | | | also drop sometrue/alltrue link, its equivalent to any/all.
* | Merge pull request #3738 from charris/backport-1.8-3736Charles Harris2013-09-131-0/+1
|\ \ | | | | | | Backport 1.8 3736
| * | DOC: Make savez_compressed show up in the documentation.Charles Harris2013-09-131-0/+1
|/ / | | | | | | Closes #3708.
* | Merge pull request #3735 from juliantaylor/fix-subarray-zeros-1.8.xCharles Harris2013-09-132-27/+42
|\ \ | | | | | | Backport PR 3725
| * | BUG: fix zero initialization with subarray dtypesJulian Taylor2013-09-132-27/+42
| |/ | | | | | | closes #3724
* | Merge pull request #3734 from charris/backport-1.8-3723Charles Harris2013-09-131-38/+22
|\ \ | |/ |/| Backport 1.8 3723
| * MAINT: Make the qr raw mode test independent of the LAPACK library.Charles Harris2013-09-131-38/+22
|/ | | | | | | | The qr factorization is not unique and the values returned by the raw mode may differ between LAPACK versions. Consequently, the results cannot be checked against known values. Closes #3703.
* Merge pull request #3729 from juliantaylor/f2py-negative-refcount-1.8.xCharles Harris2013-09-121-2/+3
|\ | | | | Backport PR 3728
| * BUG: f2py, fix decref on wrong objectJulian Taylor2013-09-121-2/+3
| | | | | | | | | | | | missing brackets causes decref on an wrong object. shows itself as abort with negative refcount in test_callback using python-dbg.
* | Merge pull request #3727 from juliantaylor/dbg-fixes-1.8.xCharles Harris2013-09-123-10/+12
|\ \ | | | | | | Backport PR 3726
| * | BUG: fix bad alignment assertsJulian Taylor2013-09-121-3/+3
| | | | | | | | | | | | if the arrays are smaller than a vector they don't need to be aligned.
View Lorry Controller Status