From noreply at github.com Thu Oct 2 11:34:28 2014 From: noreply at github.com (GitHub) Date: Thu, 02 Oct 2014 08:34:28 -0700 Subject: [Numpy-svn] [numpy/numpy] d3470b: ENH: unroll correlate loops Message-ID: <542d708428baf_45453f7ff2f5d2bc104925@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: d3470b7351358e7d34141a818a6f71e81e0172ef https://github.com/numpy/numpy/commit/d3470b7351358e7d34141a818a6f71e81e0172ef Author: Julian Taylor Date: 2014-09-29 (Mon, 29 Sep 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src M numpy/core/src/multiarray/arraytypes.h M numpy/core/src/multiarray/templ_common.h.src Log Message: ----------- ENH: unroll correlate loops Unrolling the inner loop ensures the kernel is placed into registers and reused, this almost doubles performance. Also move the function to arraytypes.c.src where also our dot product is placed. The templ_common.h might still be useful in future so its kept. Commit: 58350f4608a22f4b4b66795f51eaefc206bd02b8 https://github.com/numpy/numpy/commit/58350f4608a22f4b4b66795f51eaefc206bd02b8 Author: Charles Harris Date: 2014-10-02 (Thu, 02 Oct 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src M numpy/core/src/multiarray/arraytypes.h M numpy/core/src/multiarray/templ_common.h.src Log Message: ----------- Merge pull request #5135 from juliantaylor/correlate-improve ENH: unroll correlate loops Compare: https://github.com/numpy/numpy/compare/fa14ae0ef313...58350f4608a2 From noreply at github.com Sat Oct 4 14:36:00 2014 From: noreply at github.com (GitHub) Date: Sat, 04 Oct 2014 11:36:00 -0700 Subject: [Numpy-svn] [numpy/numpy] dc4539: BUG: Make numpy import when run with Python flag '... Message-ID: <54303e10117a3_297a3fa28dc8f2c048482@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: dc453917978e98dcdf3bbc106b080c80f0a1301e https://github.com/numpy/numpy/commit/dc453917978e98dcdf3bbc106b080c80f0a1301e Author: Charles Harris Date: 2014-10-03 (Fri, 03 Oct 2014) Changed paths: M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_utils.py M numpy/ma/extras.py Log Message: ----------- BUG: Make numpy import when run with Python flag '-OO'. This consists of checking for a docstring equal to None and skipping two tests that require docstrings. Closes #5148. Commit: 43ca904f79ced66552ff02e7e92da1c4566beadd https://github.com/numpy/numpy/commit/43ca904f79ced66552ff02e7e92da1c4566beadd Author: Charles Harris Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M .travis.yml M tools/travis-test.sh Log Message: ----------- TST: Add test with python flag -OO. Commit: ed228e68451106e0673ad050cf71942e90a831c3 https://github.com/numpy/numpy/commit/ed228e68451106e0673ad050cf71942e90a831c3 Author: Julian Taylor Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M .travis.yml M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_utils.py M numpy/ma/extras.py M tools/travis-test.sh Log Message: ----------- Merge pull request #5149 from charris/fix-ma-docs-for-OO-option BUG: Make numpy import when run with Python flag '-OO Compare: https://github.com/numpy/numpy/compare/e1e9117c4921...ed228e684511 From noreply at github.com Sat Oct 4 14:35:51 2014 From: noreply at github.com (GitHub) Date: Sat, 04 Oct 2014 11:35:51 -0700 Subject: [Numpy-svn] [numpy/numpy] dc4539: BUG: Make numpy import when run with Python flag '... Message-ID: <54303e071f478_23603ff2de4f92bc452eb@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: dc453917978e98dcdf3bbc106b080c80f0a1301e https://github.com/numpy/numpy/commit/dc453917978e98dcdf3bbc106b080c80f0a1301e Author: Charles Harris Date: 2014-10-03 (Fri, 03 Oct 2014) Changed paths: M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_utils.py M numpy/ma/extras.py Log Message: ----------- BUG: Make numpy import when run with Python flag '-OO'. This consists of checking for a docstring equal to None and skipping two tests that require docstrings. Closes #5148. Commit: 43ca904f79ced66552ff02e7e92da1c4566beadd https://github.com/numpy/numpy/commit/43ca904f79ced66552ff02e7e92da1c4566beadd Author: Charles Harris Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M .travis.yml M tools/travis-test.sh Log Message: ----------- TST: Add test with python flag -OO. Commit: cd0b8e7031fa16a60b22b74f1b0273ca60d58668 https://github.com/numpy/numpy/commit/cd0b8e7031fa16a60b22b74f1b0273ca60d58668 Author: Julian Taylor Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M .travis.yml M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_utils.py M numpy/ma/extras.py M tools/travis-test.sh Log Message: ----------- Merge pull request #5149 from charris/fix-ma-docs-for-OO-option BUG: Make numpy import when run with Python flag '-OO Compare: https://github.com/numpy/numpy/compare/58350f4608a2...cd0b8e7031fa From noreply at github.com Sat Oct 4 19:07:13 2014 From: noreply at github.com (GitHub) Date: Sat, 04 Oct 2014 16:07:13 -0700 Subject: [Numpy-svn] [numpy/numpy] 8615d5: BUG: fixed #5147 collapsingbug in einsum Message-ID: <54307da1cc41_27033f7ffbfbf29c215e6@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 8615d5e57c7f07c41f5f5c87d3b8ab6da4bfb4ec https://github.com/numpy/numpy/commit/8615d5e57c7f07c41f5f5c87d3b8ab6da4bfb4ec Author: Toon Verstraelen Date: 2014-10-05 (Sun, 05 Oct 2014) Changed paths: M numpy/core/src/multiarray/einsum.c.src M numpy/core/tests/test_einsum.py Log Message: ----------- BUG: fixed #5147 collapsingbug in einsum Commit: 7d1a3cf3478b5dde9fcbeb54a4fe826a5f22f530 https://github.com/numpy/numpy/commit/7d1a3cf3478b5dde9fcbeb54a4fe826a5f22f530 Author: Julian Taylor Date: 2014-10-05 (Sun, 05 Oct 2014) Changed paths: M numpy/core/src/multiarray/einsum.c.src M numpy/core/tests/test_einsum.py Log Message: ----------- Merge pull request #5152 from tovrstra/collapsingbug fixed #5147 collapsingbug in einsum Compare: https://github.com/numpy/numpy/compare/cd0b8e7031fa...7d1a3cf3478b From noreply at github.com Sat Oct 4 19:07:23 2014 From: noreply at github.com (GitHub) Date: Sat, 04 Oct 2014 16:07:23 -0700 Subject: [Numpy-svn] [numpy/numpy] 8615d5: BUG: fixed #5147 collapsingbug in einsum Message-ID: <54307dabdbe14_35ca3fdcdc06f2b834716@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 8615d5e57c7f07c41f5f5c87d3b8ab6da4bfb4ec https://github.com/numpy/numpy/commit/8615d5e57c7f07c41f5f5c87d3b8ab6da4bfb4ec Author: Toon Verstraelen Date: 2014-10-05 (Sun, 05 Oct 2014) Changed paths: M numpy/core/src/multiarray/einsum.c.src M numpy/core/tests/test_einsum.py Log Message: ----------- BUG: fixed #5147 collapsingbug in einsum Commit: 5a5dc6e8ebba7bc4c664e547e8f8e4144b1005b6 https://github.com/numpy/numpy/commit/5a5dc6e8ebba7bc4c664e547e8f8e4144b1005b6 Author: Julian Taylor Date: 2014-10-05 (Sun, 05 Oct 2014) Changed paths: M numpy/core/src/multiarray/einsum.c.src M numpy/core/tests/test_einsum.py Log Message: ----------- Merge pull request #5152 from tovrstra/collapsingbug fixed #5147 collapsingbug in einsum Compare: https://github.com/numpy/numpy/compare/ed228e684511...5a5dc6e8ebba From noreply at github.com Wed Oct 8 13:43:17 2014 From: noreply at github.com (GitHub) Date: Wed, 08 Oct 2014 10:43:17 -0700 Subject: [Numpy-svn] [numpy/numpy] ee7ca7: fix intention of loop in test_copyto_permut Message-ID: <543577b55c99b_38003ff4c8b5d2c07575a@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: ee7ca7d14cc073453b00d73102046515ea1b077c https://github.com/numpy/numpy/commit/ee7ca7d14cc073453b00d73102046515ea1b077c Author: Brian Kearns Date: 2014-10-08 (Wed, 08 Oct 2014) Changed paths: M numpy/core/tests/test_api.py Log Message: ----------- fix intention of loop in test_copyto_permut Commit: e6e2bb0f4a225559eff23089d84a57c0f2862221 https://github.com/numpy/numpy/commit/e6e2bb0f4a225559eff23089d84a57c0f2862221 Author: Julian Taylor Date: 2014-10-08 (Wed, 08 Oct 2014) Changed paths: M numpy/core/tests/test_api.py Log Message: ----------- Merge pull request #5160 from bdkearns/master fix intention of loop in test_copyto_permut Compare: https://github.com/numpy/numpy/compare/7d1a3cf3478b...e6e2bb0f4a22 From noreply at github.com Fri Oct 10 12:15:39 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 09:15:39 -0700 Subject: [Numpy-svn] [numpy/numpy] 017969: BUG: remove static from function to avoid gcc-4.1 ... Message-ID: <5438062b9def_4d003fcc7a52d2b8757eb@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 017969b193a135cd3bb63d5704bd11b62449f34e https://github.com/numpy/numpy/commit/017969b193a135cd3bb63d5704bd11b62449f34e Author: Julian Taylor Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src Log Message: ----------- BUG: remove static from function to avoid gcc-4.1 miscompile gcc 4.1.2 (e.g. in red hat 5) will miscompile the function when inlining. Without static it will not inline solving the issue. Closes gh-5163 Commit: 7eff5b881b7d6353b58195762fe2a677826fa84f https://github.com/numpy/numpy/commit/7eff5b881b7d6353b58195762fe2a677826fa84f Author: Charles Harris Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src Log Message: ----------- Merge pull request #5167 from juliantaylor/rh5-miscompile BUG: remove static from function Compare: https://github.com/numpy/numpy/compare/e6e2bb0f4a22...7eff5b881b7d From noreply at github.com Fri Oct 10 12:16:30 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 09:16:30 -0700 Subject: [Numpy-svn] [numpy/numpy] 0650d0: MAINT: Improve computation of scaled companion mat... Message-ID: <5438065ec831d_2fc03ff04b5a929c789c1@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 0650d04078666997c3dac3beec6fe872dca797dd https://github.com/numpy/numpy/commit/0650d04078666997c3dac3beec6fe872dca797dd Author: Charles Harris Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M numpy/polynomial/hermite.py M numpy/polynomial/hermite_e.py Log Message: ----------- MAINT: Improve computation of scaled companion matrices. The previous method used for hermite and hermite_e polynomials suffered from double overflow for polynomials of large degree. Those numbers were later scaled down by equally large numbers, but the result was NaN. The wanted values are now computed in such a way that overflow in some entries is replaced by underflow in others. The resulting zeros are a negligible perturbation of the companion matrix. Commit: dc002061a0a7bf3cb1c6b7c9e3c68553b5d22f82 https://github.com/numpy/numpy/commit/dc002061a0a7bf3cb1c6b7c9e3c68553b5d22f82 Author: Charles Harris Date: 2014-10-04 (Sat, 04 Oct 2014) Changed paths: M numpy/polynomial/hermite.py M numpy/polynomial/hermite_e.py Log Message: ----------- MAINT: Improve computation of gauss weights for hermite and hermite_e. The new computations use normalized polynomials. Those polynomials are less subject to overflow and allow the determination of the weights and roots up to about degree 350, at which point the weights underflow. Two private functions are added for computing the values of the normalized polynomials: 1. _normed_hermite_n 2. _normed_hermite_e_n Commit: d6d67957d3af1f4a4845ba3bc2d8b1fdd9e49e96 https://github.com/numpy/numpy/commit/d6d67957d3af1f4a4845ba3bc2d8b1fdd9e49e96 Author: Charles Harris Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/polynomial/hermite.py M numpy/polynomial/hermite_e.py Log Message: ----------- Merge pull request #5154 from charris/improve-gauss-hermite Improve gauss hermite Compare: https://github.com/numpy/numpy/compare/7eff5b881b7d...d6d67957d3af From noreply at github.com Fri Oct 10 12:18:04 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 09:18:04 -0700 Subject: [Numpy-svn] [numpy/numpy] 1d6ef5: BUG: Make f2py work with intent(in out). Message-ID: <543806bca3eab_3f973f97fafc92bc61968@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 1d6ef527cc9e63272cf20bdfab0accb40cb12635 https://github.com/numpy/numpy/commit/1d6ef527cc9e63272cf20bdfab0accb40cb12635 Author: Charles Harris Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: M numpy/f2py/crackfortran.py Log Message: ----------- BUG: Make f2py work with intent(in out). Note that Fortran ignores spaces in this case, so that 'in out' is treated as 'inout'. Closes #479. Commit: e98686d912ddc641f11b7531d164a5b09194b1df https://github.com/numpy/numpy/commit/e98686d912ddc641f11b7531d164a5b09194b1df Author: Charles Harris Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: A numpy/f2py/tests/src/regression/inout.f90 A numpy/f2py/tests/test_regression.py Log Message: ----------- TST: Add basic test for compiling with intent(in out). This checks that the compilation works and that the expected error is raised when non-contiguous arrays are passed as intent(in out). Commit: 6457686e8fd5df3ef5d64f3e3cff27ccec4dac6c https://github.com/numpy/numpy/commit/6457686e8fd5df3ef5d64f3e3cff27ccec4dac6c Author: Charles Harris Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/f2py/crackfortran.py A numpy/f2py/tests/src/regression/inout.f90 A numpy/f2py/tests/test_regression.py Log Message: ----------- Merge pull request #5168 from charris/f2py-space-fix BUG: Make f2py work with intent(in out). Compare: https://github.com/numpy/numpy/compare/d6d67957d3af...6457686e8fd5 From noreply at github.com Fri Oct 10 13:02:40 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 10:02:40 -0700 Subject: [Numpy-svn] [numpy/numpy] 017969: BUG: remove static from function to avoid gcc-4.1 ... Message-ID: <5438113065f61_51233f9645d492a08545@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 017969b193a135cd3bb63d5704bd11b62449f34e https://github.com/numpy/numpy/commit/017969b193a135cd3bb63d5704bd11b62449f34e Author: Julian Taylor Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src Log Message: ----------- BUG: remove static from function to avoid gcc-4.1 miscompile gcc 4.1.2 (e.g. in red hat 5) will miscompile the function when inlining. Without static it will not inline solving the issue. Closes gh-5163 Commit: 1b7712bde24a55dc457ade84bb2e214cdf9e5637 https://github.com/numpy/numpy/commit/1b7712bde24a55dc457ade84bb2e214cdf9e5637 Author: Julian Taylor Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/core/src/multiarray/arraytypes.c.src Log Message: ----------- Merge pull request #5167 from juliantaylor/rh5-miscompile BUG: remove static from function Compare: https://github.com/numpy/numpy/compare/5a5dc6e8ebba...1b7712bde24a From noreply at github.com Fri Oct 10 13:48:36 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 10:48:36 -0700 Subject: [Numpy-svn] [numpy/numpy] 6cf7ec: REL: set version number to unreleased 1.9.1 Message-ID: <54381bf4f21c3_54813faedab572a059654@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 6cf7ecefbfbc587647760b68fd7f803271fca3be https://github.com/numpy/numpy/commit/6cf7ecefbfbc587647760b68fd7f803271fca3be Author: Julian Taylor Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M setup.py Log Message: ----------- REL: set version number to unreleased 1.9.1 From noreply at github.com Fri Oct 10 14:56:24 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 11:56:24 -0700 Subject: [Numpy-svn] [numpy/numpy] d6278b: MAINT: replace bytewise copy with memcpy Message-ID: <54382bd855532_52583fa36dcc92a0520dc@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: d6278bdc47d12acc5d3edfad3ad0654055f77737 https://github.com/numpy/numpy/commit/d6278bdc47d12acc5d3edfad3ad0654055f77737 Author: Julian Taylor Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_cpu.h Log Message: ----------- MAINT: replace bytewise copy with memcpy memcpy takes care of unaligned memory and is inlined by the compiler, there is no need for own code. Commit: 8e113c9844b8760f9ce089ef774d46f06fdc7e82 https://github.com/numpy/numpy/commit/8e113c9844b8760f9ce089ef774d46f06fdc7e82 Author: Charles Harris Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_cpu.h Log Message: ----------- Merge pull request #5169 from juliantaylor/pyobjcopy-simplification MAINT: replace bytewise copy with memcpy Compare: https://github.com/numpy/numpy/compare/6457686e8fd5...8e113c9844b8 From noreply at github.com Fri Oct 10 14:59:42 2014 From: noreply at github.com (GitHub) Date: Fri, 10 Oct 2014 11:59:42 -0700 Subject: [Numpy-svn] [numpy/numpy] 1d6ef5: BUG: Make f2py work with intent(in out). Message-ID: <54382c9eee549_46ea3f83ecf3b2b8952f0@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 1d6ef527cc9e63272cf20bdfab0accb40cb12635 https://github.com/numpy/numpy/commit/1d6ef527cc9e63272cf20bdfab0accb40cb12635 Author: Charles Harris Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: M numpy/f2py/crackfortran.py Log Message: ----------- BUG: Make f2py work with intent(in out). Note that Fortran ignores spaces in this case, so that 'in out' is treated as 'inout'. Closes #479. Commit: e98686d912ddc641f11b7531d164a5b09194b1df https://github.com/numpy/numpy/commit/e98686d912ddc641f11b7531d164a5b09194b1df Author: Charles Harris Date: 2014-10-09 (Thu, 09 Oct 2014) Changed paths: A numpy/f2py/tests/src/regression/inout.f90 A numpy/f2py/tests/test_regression.py Log Message: ----------- TST: Add basic test for compiling with intent(in out). This checks that the compilation works and that the expected error is raised when non-contiguous arrays are passed as intent(in out). Commit: e2851d2a6847460c965243bf241d0c1d1b85b494 https://github.com/numpy/numpy/commit/e2851d2a6847460c965243bf241d0c1d1b85b494 Author: Julian Taylor Date: 2014-10-10 (Fri, 10 Oct 2014) Changed paths: M numpy/f2py/crackfortran.py A numpy/f2py/tests/src/regression/inout.f90 A numpy/f2py/tests/test_regression.py Log Message: ----------- Merge pull request #5168 from charris/f2py-space-fix BUG: Make f2py work with intent(in out). Compare: https://github.com/numpy/numpy/compare/6cf7ecefbfbc...e2851d2a6847 From noreply at github.com Sat Oct 11 13:18:41 2014 From: noreply at github.com (GitHub) Date: Sat, 11 Oct 2014 10:18:41 -0700 Subject: [Numpy-svn] [numpy/numpy] 2e1ff0: ENH: allow auto vectorization of integer loops Message-ID: <543966713f5b6_370e3fb2c65ef2a0733ba@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 2e1ff063751461f7c4287933b88554f4d44d7ee5 https://github.com/numpy/numpy/commit/2e1ff063751461f7c4287933b88554f4d44d7ee5 Author: Julian Taylor Date: 2014-10-11 (Sat, 11 Oct 2014) Changed paths: M numpy/core/src/umath/loops.c.src M numpy/core/tests/test_umath.py Log Message: ----------- ENH: allow auto vectorization of integer loops Generalize the loop specializations with two new macros UNARY_LOOP_FAST and BINARY_LOOP_FAST and apply it to more integer loops. Same also gh-4561 which did this for integer boolean operations. Commit: 9636420543967fce47688da5e2b96c1c5155533d https://github.com/numpy/numpy/commit/9636420543967fce47688da5e2b96c1c5155533d Author: Charles Harris Date: 2014-10-11 (Sat, 11 Oct 2014) Changed paths: M numpy/core/src/umath/loops.c.src M numpy/core/tests/test_umath.py Log Message: ----------- Merge pull request #5144 from juliantaylor/int-vectorize ENH: allow auto vectorization of integer loops Compare: https://github.com/numpy/numpy/compare/8e113c9844b8...963642054396 From noreply at github.com Sun Oct 12 10:32:44 2014 From: noreply at github.com (GitHub) Date: Sun, 12 Oct 2014 07:32:44 -0700 Subject: [Numpy-svn] [numpy/numpy] 8b17fa: MAINT: core: string constants in static buffers Message-ID: <543a910c13381_6de53ff8ecfc92a0231d@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 8b17fae1428502ade57c2aa5706add6228711713 https://github.com/numpy/numpy/commit/8b17fae1428502ade57c2aa5706add6228711713 Author: Lars Buitinck Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/core/src/multiarray/multiarraymodule.c Log Message: ----------- MAINT: core: string constants in static buffers Saves copying them into temporary buffers. Commit: 4c4595b2487002162e6ce5e6dd8a9c778e3d7c30 https://github.com/numpy/numpy/commit/4c4595b2487002162e6ce5e6dd8a9c778e3d7c30 Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/core/src/multiarray/multiarraymodule.c Log Message: ----------- Merge pull request #5175 from larsmans/c-fixes MAINT: core: string constants in static buffers Compare: https://github.com/numpy/numpy/compare/963642054396...4c4595b24870 From noreply at github.com Sun Oct 12 12:20:30 2014 From: noreply at github.com (GitHub) Date: Sun, 12 Oct 2014 09:20:30 -0700 Subject: [Numpy-svn] [numpy/numpy] ade76b: DOC: core: contents of newly allocated space in me... Message-ID: <543aaa4e1ba61_62523f9e5ad572c06347d@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: ade76b0035496ad0239b5236688fa51d4a4e592e https://github.com/numpy/numpy/commit/ade76b0035496ad0239b5236688fa51d4a4e592e Author: Lars Buitinck Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/core/memmap.py Log Message: ----------- DOC: core: contents of newly allocated space in memmap Fixes #5157. Commit: 2ba94e726c7adf5b6ea7f2e49aadb78fe8b1d7ba https://github.com/numpy/numpy/commit/2ba94e726c7adf5b6ea7f2e49aadb78fe8b1d7ba Author: Julian Taylor Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/core/memmap.py Log Message: ----------- Merge pull request #5176 from larsmans/memmap-doc DOC: core: contents of newly allocated space in memmap Compare: https://github.com/numpy/numpy/compare/4c4595b24870...2ba94e726c7a From noreply at github.com Mon Oct 13 14:12:39 2014 From: noreply at github.com (GitHub) Date: Mon, 13 Oct 2014 11:12:39 -0700 Subject: [Numpy-svn] [numpy/numpy] 8f068b: BUG: Make python2 *.npy files readable in python3. Message-ID: <543c1617384fc_670b3fcd5a4dd2a0119d2@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 8f068b7866a2959392fae2a15c1a5a19ff79bae9 https://github.com/numpy/numpy/commit/8f068b7866a2959392fae2a15c1a5a19ff79bae9 Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- BUG: Make python2 *.npy files readable in python3. The Python2 generated file had long integer literals like '1L' that broke in Python3. The fix here is to filter out the 'L' and let safe_eval take care of the integer type in converting the string. The fix here comes from Nathaniel Smith with a few added fixups. Closes #5170. Commit: 8b1f90a227b5fcf2a481c973e522693758c9f20f https://github.com/numpy/numpy/commit/8b1f90a227b5fcf2a481c973e522693758c9f20f Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M .gitattributes A numpy/lib/tests/data/python3.npy A numpy/lib/tests/data/win64python2.npy M numpy/lib/tests/test_format.py Log Message: ----------- TST: Add tests for Python2, Python3 *.npy compatibility. Commit: 2f1786306e696709481ccfc95b844d5aa2700b6d https://github.com/numpy/numpy/commit/2f1786306e696709481ccfc95b844d5aa2700b6d Author: Julian Taylor Date: 2014-10-13 (Mon, 13 Oct 2014) Changed paths: M .gitattributes M numpy/lib/format.py A numpy/lib/tests/data/python3.npy A numpy/lib/tests/data/win64python2.npy M numpy/lib/tests/test_format.py Log Message: ----------- Merge pull request #5178 from charris/fix-npz-header-incompatibility Fix npz header incompatibility Compare: https://github.com/numpy/numpy/compare/2ba94e726c7a...2f1786306e69 From noreply at github.com Mon Oct 13 14:13:50 2014 From: noreply at github.com (GitHub) Date: Mon, 13 Oct 2014 11:13:50 -0700 Subject: [Numpy-svn] [numpy/numpy] 8f068b: BUG: Make python2 *.npy files readable in python3. Message-ID: <543c165ee1aac_c423f9d6cbb32a07088b@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 8f068b7866a2959392fae2a15c1a5a19ff79bae9 https://github.com/numpy/numpy/commit/8f068b7866a2959392fae2a15c1a5a19ff79bae9 Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- BUG: Make python2 *.npy files readable in python3. The Python2 generated file had long integer literals like '1L' that broke in Python3. The fix here is to filter out the 'L' and let safe_eval take care of the integer type in converting the string. The fix here comes from Nathaniel Smith with a few added fixups. Closes #5170. Commit: 8b1f90a227b5fcf2a481c973e522693758c9f20f https://github.com/numpy/numpy/commit/8b1f90a227b5fcf2a481c973e522693758c9f20f Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M .gitattributes A numpy/lib/tests/data/python3.npy A numpy/lib/tests/data/win64python2.npy M numpy/lib/tests/test_format.py Log Message: ----------- TST: Add tests for Python2, Python3 *.npy compatibility. Commit: 90ae342909af9db1c8e53d7748ebb196d64ea69f https://github.com/numpy/numpy/commit/90ae342909af9db1c8e53d7748ebb196d64ea69f Author: Julian Taylor Date: 2014-10-13 (Mon, 13 Oct 2014) Changed paths: M .gitattributes M numpy/lib/format.py A numpy/lib/tests/data/python3.npy A numpy/lib/tests/data/win64python2.npy M numpy/lib/tests/test_format.py Log Message: ----------- Merge pull request #5178 from charris/fix-npz-header-incompatibility Fix npz header incompatibility Compare: https://github.com/numpy/numpy/compare/e2851d2a6847...90ae342909af From noreply at github.com Wed Oct 15 12:07:13 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 09:07:13 -0700 Subject: [Numpy-svn] [numpy/numpy] 657c8b: Add examples for intersect1d and union1d of more t... Message-ID: <543e9bb164129_4263fac9d8e329c31386@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 657c8b9106084ea284b3bd50f6edd6bfef880a50 https://github.com/numpy/numpy/commit/657c8b9106084ea284b3bd50f6edd6bfef880a50 Author: Christian Brueffer Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/arraysetops.py Log Message: ----------- Add examples for intersect1d and union1d of more than two arrays. The approach was suggested by Jaime Frio in issue #5179. Commit: fd06081e5cb484e9acda1db5c92574ac3d740bb4 https://github.com/numpy/numpy/commit/fd06081e5cb484e9acda1db5c92574ac3d740bb4 Author: Nathaniel J. Smith Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/arraysetops.py Log Message: ----------- Merge pull request #5189 from cbrueffer/setops Add examples for intersect1d and union1d of more than two arrays. Compare: https://github.com/numpy/numpy/compare/2f1786306e69...fd06081e5cb4 From noreply at github.com Wed Oct 15 14:54:32 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 11:54:32 -0700 Subject: [Numpy-svn] [numpy/numpy] a8e027: BUG: Fix writing of intrinsic long integers in pyt... Message-ID: <543ec2e8baac4_4d183f89765ab2b84281e@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: a8e027ff49fdfdce4d845eef7e1f8cd939ae840a https://github.com/numpy/numpy/commit/a8e027ff49fdfdce4d845eef7e1f8cd939ae840a Author: Charles Harris Date: 2014-10-13 (Mon, 13 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- BUG: Fix writing of intrinsic long integers in python2 npy files. This only happens when C long is smaller than npy_intp, in particular, on 64 bit windows. The 'L' suffix is not needed as long as safe_eval is used to recover the values, and the long intrinsic causes problems for python3. The python3 read side has been fixed, but we might as well fix the problem at the root as well. Commit: e2218a6984b756a1806041f8fb6869f63365ad10 https://github.com/numpy/numpy/commit/e2218a6984b756a1806041f8fb6869f63365ad10 Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- Merge pull request #5183 from charris/fix-npy-header-write BUG: Fix writing of intrinsic long integers in python2 npy files. Compare: https://github.com/numpy/numpy/compare/fd06081e5cb4...e2218a6984b7 From noreply at github.com Wed Oct 15 14:59:36 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 11:59:36 -0700 Subject: [Numpy-svn] [numpy/numpy] a8e027: BUG: Fix writing of intrinsic long integers in pyt... Message-ID: <543ec418d7028_a343fae8094b2b873266@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: a8e027ff49fdfdce4d845eef7e1f8cd939ae840a https://github.com/numpy/numpy/commit/a8e027ff49fdfdce4d845eef7e1f8cd939ae840a Author: Charles Harris Date: 2014-10-13 (Mon, 13 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- BUG: Fix writing of intrinsic long integers in python2 npy files. This only happens when C long is smaller than npy_intp, in particular, on 64 bit windows. The 'L' suffix is not needed as long as safe_eval is used to recover the values, and the long intrinsic causes problems for python3. The python3 read side has been fixed, but we might as well fix the problem at the root as well. Commit: fb6ff9653168841ca49946e729537eee158ffc13 https://github.com/numpy/numpy/commit/fb6ff9653168841ca49946e729537eee158ffc13 Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/format.py Log Message: ----------- Merge pull request #5183 from charris/fix-npy-header-write BUG: Fix writing of intrinsic long integers in python2 npy files. Compare: https://github.com/numpy/numpy/compare/90ae342909af...fb6ff9653168 From noreply at github.com Wed Oct 15 17:42:37 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 14:42:37 -0700 Subject: [Numpy-svn] [numpy/numpy] 89d627: BLD: change expected type sizes to 64 bit Message-ID: <543eea4d7063d_43963f852f82d2c06408d@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 89d627ee05eac6b188ad27d30b4f6a893392f9ce https://github.com/numpy/numpy/commit/89d627ee05eac6b188ad27d30b4f6a893392f9ce Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/setup.py Log Message: ----------- BLD: change expected type sizes to 64 bit 64 bit is more common these days, so switch the expected type sizes to values common on this platform. This removes a few compile tests. The expected sizes are just hints and do not have to be correct. Commit: aa8a2437b700ce9a8c3e12652b75982e96a3777c https://github.com/numpy/numpy/commit/aa8a2437b700ce9a8c3e12652b75982e96a3777c Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/setup.py Log Message: ----------- BLD: fix expected size argument for complex types The caller was passing a list of lists instead of a list with its entries doubled. Removes the binary type search for most if not all platforms. Commit: 03f0aebf1a43c672f17e89382804809601828111 https://github.com/numpy/numpy/commit/03f0aebf1a43c672f17e89382804809601828111 Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/bscript M numpy/core/setup.py M numpy/core/tests/test_dtype.py M numpy/core/tests/test_nditer.py M numpy/distutils/command/build_ext.py M numpy/f2py/setup.py M numpy/testing/utils.py Log Message: ----------- MAINT: remove obsolete version checks minimum requirement is python2.6, this allows removing a couple 2.3 and 2.4 checks. Commit: 50ff7da7799bac28fb14e3290d6c73683adf7d13 https://github.com/numpy/numpy/commit/50ff7da7799bac28fb14e3290d6c73683adf7d13 Author: Charles Harris Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/bscript M numpy/core/setup.py M numpy/core/tests/test_dtype.py M numpy/core/tests/test_nditer.py M numpy/distutils/command/build_ext.py M numpy/f2py/setup.py M numpy/testing/utils.py Log Message: ----------- Merge pull request #5191 from juliantaylor/config-maint size and version check maintenance Compare: https://github.com/numpy/numpy/compare/e2218a6984b7...50ff7da7799b From noreply at github.com Wed Oct 15 17:48:12 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 14:48:12 -0700 Subject: [Numpy-svn] [numpy/numpy] 9917a1: MAINT: Remove use of compiler module. Message-ID: <543eeb9c22e8e_4743f967bdd12a0100095@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 9917a12d397e6e1188823752aa21996dbb1444d5 https://github.com/numpy/numpy/commit/9917a12d397e6e1188823752aa21996dbb1444d5 Author: Charles Harris Date: 2014-10-12 (Sun, 12 Oct 2014) Changed paths: M numpy/lib/utils.py Log Message: ----------- MAINT: Remove use of compiler module. The ast module was added in Python 2.6 as a replacement for the compiler module. As we no longer support Python versions < 2.6, all uses of the compiler module can be removed. Commit: 35fad05fba626aeaf836f5bee45556a59a5ae198 https://github.com/numpy/numpy/commit/35fad05fba626aeaf836f5bee45556a59a5ae198 Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/utils.py Log Message: ----------- Merge pull request #5177 from charris/do-not-use-compiler-module MAINT: Remove use of compiler module. Compare: https://github.com/numpy/numpy/compare/50ff7da7799b...35fad05fba62 From noreply at github.com Wed Oct 15 19:50:25 2014 From: noreply at github.com (GitHub) Date: Wed, 15 Oct 2014 16:50:25 -0700 Subject: [Numpy-svn] [numpy/numpy] 727655: MAINT: Use ast.literal_eval in safe_eval. Message-ID: <543f08412aad2_49553fda71e652b8289e5@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 727655193aa9b4d4068650a3f8a9b49a6a2e0cb4 https://github.com/numpy/numpy/commit/727655193aa9b4d4068650a3f8a9b49a6a2e0cb4 Author: Charles Harris Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/lib/utils.py Log Message: ----------- MAINT: Use ast.literal_eval in safe_eval. This does what is needed now that the compiler module is no longer used. Commit: c2c7051d926b6ce0665e53ce8dd0f522380a57d6 https://github.com/numpy/numpy/commit/c2c7051d926b6ce0665e53ce8dd0f522380a57d6 Author: Charles Harris Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M numpy/lib/utils.py Log Message: ----------- DEP: Deprecate SafeEval class. The class is no longer used in numpy and was never exported. Commit: 51f0976c1ca101a01d09e26ee5dfea5360f73c63 https://github.com/numpy/numpy/commit/51f0976c1ca101a01d09e26ee5dfea5360f73c63 Author: Julian Taylor Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M numpy/lib/utils.py Log Message: ----------- Merge pull request #5192 from charris/deprecate-safeeval-class Deprecate SafeEval class Compare: https://github.com/numpy/numpy/compare/35fad05fba62...51f0976c1ca1 From noreply at github.com Thu Oct 16 13:17:26 2014 From: noreply at github.com (GitHub) Date: Thu, 16 Oct 2014 10:17:26 -0700 Subject: [Numpy-svn] [numpy/numpy] 8969da: BUG: Fixes #5184 gradient calculation behavior at ... Message-ID: <543ffda6ca83c_54ea3fe92312d29c95873@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 8969da59ea9803b81b0dcb2fab3f41d4b168bc68 https://github.com/numpy/numpy/commit/8969da59ea9803b81b0dcb2fab3f41d4b168bc68 Author: David M Fobes Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- BUG: Fixes #5184 gradient calculation behavior at boundaries * Previous expected behavior was that the gradient is computed using central differences in the interior and first differences at the boundaries. * gradient was updated in v1.9.0 so that second-order accurate calculations are done at the boundaries, but this breaks expected behavior with old code, so `edge_order` keyword (Default: 1) is added to specify whether first or second order calculations at the boundaries should be used. * Since the second argument is *varargs, in order to maintain compatibility with old code and compatibility with python 2.6 & 2.7, **kwargs is used, and all kwargs that are not `edge_order` raise an error, listing the offending kwargs. * Tests and documentation updated to reflect this change. * Added `.. versionadded:: 1.9.1` to the new optional kwarg `edge_order` documentation, and specified supported `edge_order` kwarg values. * Clarified documentation for `varargs`. * Made indentation in docstring consistent with other docstring styles. * Examples corrected Commit: 4c0747d5c2cb3c961f7246cd9222b89bdcf2ce42 https://github.com/numpy/numpy/commit/4c0747d5c2cb3c961f7246cd9222b89bdcf2ce42 Author: Charles Harris Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- Merge pull request #5186 from pseudocubic/gradient-boundary-order-keyword BUG: Fixes #5184 gradient calculation behavior at boundaries Compare: https://github.com/numpy/numpy/compare/51f0976c1ca1...4c0747d5c2cb From noreply at github.com Thu Oct 16 13:29:08 2014 From: noreply at github.com (GitHub) Date: Thu, 16 Oct 2014 10:29:08 -0700 Subject: [Numpy-svn] [numpy/numpy] 2f1e27: BUG: fix nanmedian on arrays containing inf Message-ID: <5440006473141_4dbb3fda1f9f52c0671e1@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 2f1e27610453f901e81cd892b2aa3b3a2ac30819 https://github.com/numpy/numpy/commit/2f1e27610453f901e81cd892b2aa3b3a2ac30819 Author: Julian Taylor Date: 2014-10-14 (Tue, 14 Oct 2014) Changed paths: M numpy/lib/tests/test_nanfunctions.py M numpy/ma/core.py M numpy/ma/extras.py M numpy/ma/tests/test_extras.py Log Message: ----------- BUG: fix nanmedian on arrays containing inf There are two issues: A masked divide of an infinite value is a masked value which means one can't use np.ma.mean to compute the median as infinity division is well defined. This behaviour seems wrong to me but it also looks intentional so changing it is not appropriate for a bugfix release. The second issue is that the ordering of the sorted masked array is undefined for equal values, the sorting considers infinity the largest floating point value which is not correct in respect to sorting where nan is considered larger. This is fixed by changing the minimum_fill_value to nan for float data in the masked sorts. Closes gh-5138 Commit: 443a49a271e6d7c4c1dda72fe721eaed318ff764 https://github.com/numpy/numpy/commit/443a49a271e6d7c4c1dda72fe721eaed318ff764 Author: Charles Harris Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M numpy/lib/tests/test_nanfunctions.py M numpy/ma/core.py M numpy/ma/extras.py M numpy/ma/tests/test_extras.py Log Message: ----------- Merge pull request #5182 from juliantaylor/nanmedian-inf BUG: fix nanmedian on arrays containing inf Compare: https://github.com/numpy/numpy/compare/4c0747d5c2cb...443a49a271e6 From noreply at github.com Thu Oct 16 15:16:31 2014 From: noreply at github.com (GitHub) Date: Thu, 16 Oct 2014 12:16:31 -0700 Subject: [Numpy-svn] [numpy/numpy] a356a1: BLD: try linking a file if compile long double rep... Message-ID: <5440198fa0c6a_67493ff0e1c2529c16949@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: a356a103460d1b97428108431f7e73e7eda9ee85 https://github.com/numpy/numpy/commit/a356a103460d1b97428108431f7e73e7eda9ee85 Author: Julian Taylor Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/setup_common.py Log Message: ----------- BLD: try linking a file if compile long double repr detection fails When compiler uses link time optimization it will not emit the binary representation of the long double during linking. This prevents building numpy with -flto. To avoid this try linking the object when the compile detection failed. With this change numpy can be built with link time optimization with 4 worker processes using following command: CC='gcc -fno-fat-lto-objects -flto=4 -fuse-linker-plugin -O3' \ LDSHARED='gcc -fno-fat-lto-objects -flto=4 -fuse-linker-plugin -shared -O3' AR=gcc-ar \ python setup.py build_ext See also: https://www.mail-archive.com/numpy-discussion at scipy.org/msg45535.html Should also work with icc's -ipo switch (probably closes gh-4992) Commit: f6c5611e941409faa36385b8d4c7d321cb7c03d4 https://github.com/numpy/numpy/commit/f6c5611e941409faa36385b8d4c7d321cb7c03d4 Author: Charles Harris Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M numpy/core/setup_common.py Log Message: ----------- Merge pull request #5190 from juliantaylor/lto-support BLD: try linking a file if compile long double repr detection fails Compare: https://github.com/numpy/numpy/compare/443a49a271e6...f6c5611e9414 From noreply at github.com Thu Oct 16 19:21:06 2014 From: noreply at github.com (GitHub) Date: Thu, 16 Oct 2014 16:21:06 -0700 Subject: [Numpy-svn] [numpy/numpy] 2f1e27: BUG: fix nanmedian on arrays containing inf Message-ID: <544052e2d701a_4f2b3fa85c44129c8407b@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 2f1e27610453f901e81cd892b2aa3b3a2ac30819 https://github.com/numpy/numpy/commit/2f1e27610453f901e81cd892b2aa3b3a2ac30819 Author: Julian Taylor Date: 2014-10-14 (Tue, 14 Oct 2014) Changed paths: M numpy/lib/tests/test_nanfunctions.py M numpy/ma/core.py M numpy/ma/extras.py M numpy/ma/tests/test_extras.py Log Message: ----------- BUG: fix nanmedian on arrays containing inf There are two issues: A masked divide of an infinite value is a masked value which means one can't use np.ma.mean to compute the median as infinity division is well defined. This behaviour seems wrong to me but it also looks intentional so changing it is not appropriate for a bugfix release. The second issue is that the ordering of the sorted masked array is undefined for equal values, the sorting considers infinity the largest floating point value which is not correct in respect to sorting where nan is considered larger. This is fixed by changing the minimum_fill_value to nan for float data in the masked sorts. Closes gh-5138 Commit: d310ff9720fe1cde7fbce94e50f6c1878a43cf4f https://github.com/numpy/numpy/commit/d310ff9720fe1cde7fbce94e50f6c1878a43cf4f Author: Julian Taylor Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M numpy/lib/tests/test_nanfunctions.py M numpy/ma/core.py M numpy/ma/extras.py M numpy/ma/tests/test_extras.py Log Message: ----------- Merge pull request #5182 from juliantaylor/nanmedian-inf BUG: fix nanmedian on arrays containing inf Compare: https://github.com/numpy/numpy/compare/fb6ff9653168...d310ff9720fe From noreply at github.com Fri Oct 17 09:51:31 2014 From: noreply at github.com (GitHub) Date: Fri, 17 Oct 2014 06:51:31 -0700 Subject: [Numpy-svn] [numpy/numpy] 48ee0b: ENH: added parameter to use periodic x-coordinates Message-ID: <54411ee390db3_7963f7f946e32c09934b@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 48ee0b25f07bd770c5fe5d328463505f535b4d1d https://github.com/numpy/numpy/commit/48ee0b25f07bd770c5fe5d328463505f535b4d1d Author: saullogiovani Date: 2014-10-02 (Thu, 02 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- ENH: added parameter to use periodic x-coordinates Commit: 673ca72af2bf312ce86e6511483c950578cb20d2 https://github.com/numpy/numpy/commit/673ca72af2bf312ce86e6511483c950578cb20d2 Author: saullogiovani Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M .gitattributes M .travis.yml M doc/release/1.10.0-notes.rst M numpy/core/bscript M numpy/core/include/numpy/npy_cpu.h M numpy/core/memmap.py M numpy/core/setup.py M numpy/core/setup_common.py M numpy/core/src/multiarray/arraytypes.c.src M numpy/core/src/multiarray/arraytypes.h M numpy/core/src/multiarray/einsum.c.src M numpy/core/src/multiarray/multiarraymodule.c M numpy/core/src/multiarray/templ_common.h.src M numpy/core/src/umath/loops.c.src M numpy/core/tests/test_api.py M numpy/core/tests/test_dtype.py M numpy/core/tests/test_einsum.py M numpy/core/tests/test_nditer.py M numpy/core/tests/test_umath.py M numpy/distutils/command/build_ext.py M numpy/f2py/crackfortran.py M numpy/f2py/setup.py A numpy/f2py/tests/src/regression/inout.f90 A numpy/f2py/tests/test_regression.py M numpy/lib/arraysetops.py M numpy/lib/format.py M numpy/lib/function_base.py A numpy/lib/tests/data/python3.npy A numpy/lib/tests/data/win64python2.npy M numpy/lib/tests/test_format.py M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_nanfunctions.py M numpy/lib/tests/test_utils.py M numpy/lib/utils.py M numpy/ma/core.py M numpy/ma/extras.py M numpy/ma/tests/test_extras.py M numpy/polynomial/hermite.py M numpy/polynomial/hermite_e.py M numpy/testing/utils.py M tools/travis-test.sh Log Message: ----------- Merge branch 'master' of https://github.com/numpy/numpy into interp_with_period Commit: 498b84a35da66bceb1b08e49ee4dcd02c870e5fe https://github.com/numpy/numpy/commit/498b84a35da66bceb1b08e49ee4dcd02c870e5fe Author: saullogiovani Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst Log Message: ----------- ENH: added parameter to use periodic x-coordinates Commit: 5505dc9cdfa54fa5a45c8e69f3f82874c5aee9a6 https://github.com/numpy/numpy/commit/5505dc9cdfa54fa5a45c8e69f3f82874c5aee9a6 Author: Jaime Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- Merge pull request #5117 from saullocastro/interp_with_period ENH: added parameter to use periodic x-coordinates Compare: https://github.com/numpy/numpy/compare/f6c5611e9414...5505dc9cdfa5 From noreply at github.com Fri Oct 17 11:52:28 2014 From: noreply at github.com (GitHub) Date: Fri, 17 Oct 2014 08:52:28 -0700 Subject: [Numpy-svn] [numpy/numpy] 9b152a: ENH: Add keepdims to linalg.norm Message-ID: <54413b3c16699_52a83ff9b0c8b2b84256c@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 9b152aaaf3abd0f98b7a88ed66df7518a6a6c85b https://github.com/numpy/numpy/commit/9b152aaaf3abd0f98b7a88ed66df7518a6a6c85b Author: Eric Moore Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M numpy/linalg/linalg.py M numpy/linalg/tests/test_linalg.py Log Message: ----------- ENH: Add keepdims to linalg.norm Commit: 6f88eea024448b913cea881efad405e5d4ece195 https://github.com/numpy/numpy/commit/6f88eea024448b913cea881efad405e5d4ece195 Author: Eric Moore Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M numpy/linalg/tests/test_linalg.py Log Message: ----------- MAINT: Stop repeatedly running duplicate tests. These three tests don't make use of the infrastructre in _TestNorm. This means they are identical tests when run as part of TestNorm(Double,Single,Int64). Once is enough. Commit: c9d61533f0a5a40ec0a38c69e19615d375af166e https://github.com/numpy/numpy/commit/c9d61533f0a5a40ec0a38c69e19615d375af166e Author: Charles Harris Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M numpy/linalg/linalg.py M numpy/linalg/tests/test_linalg.py Log Message: ----------- Merge pull request #5196 from ewmoore/norm_keepdims Add keepdims to linalg.norm Compare: https://github.com/numpy/numpy/compare/5505dc9cdfa5...c9d61533f0a5 From noreply at github.com Fri Oct 17 13:39:25 2014 From: noreply at github.com (GitHub) Date: Fri, 17 Oct 2014 10:39:25 -0700 Subject: [Numpy-svn] [numpy/numpy] aaa894: DOC: Updated from numeric doc to indicate where ax... Message-ID: <5441544d486ec_4bb13f831bbf12a0910d7@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: aaa8940913902001a4961d793e7238eafc193861 https://github.com/numpy/numpy/commit/aaa8940913902001a4961d793e7238eafc193861 Author: Cimarron Mittelsteadt Date: 2014-10-02 (Thu, 02 Oct 2014) Changed paths: M numpy/core/fromnumeric.py Log Message: ----------- DOC: Updated from numeric doc to indicate where axis can also be a tuple of ints, i.e. axes, over which to perform the given operation Commit: 39aaa43b732c68878e6207caab3cbeee5e83bf7c https://github.com/numpy/numpy/commit/39aaa43b732c68878e6207caab3cbeee5e83bf7c Author: Cimarron Mittelsteadt Date: 2014-10-16 (Thu, 16 Oct 2014) Changed paths: M numpy/core/fromnumeric.py Log Message: ----------- DOC: Update axis documentation for tuple of ints to be consistent between functions Commit: 7b3809b0061ffbb64c227d393a3771f60d1427d5 https://github.com/numpy/numpy/commit/7b3809b0061ffbb64c227d393a3771f60d1427d5 Author: Jaime Date: 2014-10-17 (Fri, 17 Oct 2014) Changed paths: M numpy/core/fromnumeric.py Log Message: ----------- Merge pull request #5146 from cimarronm/fromnumeric_axis_tuples_doc Updated function docstrings for tuples in axis Compare: https://github.com/numpy/numpy/compare/c9d61533f0a5...7b3809b0061f From noreply at github.com Sun Oct 19 01:55:21 2014 From: noreply at github.com (GitHub) Date: Sat, 18 Oct 2014 22:55:21 -0700 Subject: [Numpy-svn] [numpy/numpy] 6114f1: MAINT: Fix typo in the 'issubclass' docstring. Message-ID: <5443524997826_3c863f811c8692a0411e7@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 6114f1c963ad011d276701967864f339e8e45ae0 https://github.com/numpy/numpy/commit/6114f1c963ad011d276701967864f339e8e45ae0 Author: Warren Weckesser Date: 2014-10-19 (Sun, 19 Oct 2014) Changed paths: M numpy/core/numerictypes.py Log Message: ----------- MAINT: Fix typo in the 'issubclass' docstring. Commit: df9db6ed5f9c9e315993f4efed5dba613e6cd84c https://github.com/numpy/numpy/commit/df9db6ed5f9c9e315993f4efed5dba613e6cd84c Author: Jaime Date: 2014-10-18 (Sat, 18 Oct 2014) Changed paths: M numpy/core/numerictypes.py Log Message: ----------- Merge pull request #5206 from WarrenWeckesser/typo MAINT: Fix typo in the 'issubclass' docstring. Compare: https://github.com/numpy/numpy/compare/7b3809b0061f...df9db6ed5f9c From noreply at github.com Mon Oct 20 17:52:21 2014 From: noreply at github.com (GitHub) Date: Mon, 20 Oct 2014 14:52:21 -0700 Subject: [Numpy-svn] [numpy/numpy] 270ae2: MAINT: Give a more helpful error for bad axis spec... Message-ID: <54458415f309a_4da13f9b00d8b2bc443c0@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 270ae2c7cac71c91a52701c5b0eb7e5f6d388b14 https://github.com/numpy/numpy/commit/270ae2c7cac71c91a52701c5b0eb7e5f6d388b14 Author: Eric Moore Date: 2014-10-20 (Mon, 20 Oct 2014) Changed paths: M numpy/linalg/linalg.py Log Message: ----------- MAINT: Give a more helpful error for bad axis specifications. This specifically addresses passing things like axis=[0,1] which gave an error message that an int was required. The real error was that if the axis isn't None or a tuple, it was simply wraped in a tuple, so the ufunc that is eventually called ended up with a list. The error message is matches that in core/src/multiarray/conversion_utils.c::PyArray_ConvertMultiAxis. Commit: 4ed4aec23912346c00d5ec2850dcd457a048479e https://github.com/numpy/numpy/commit/4ed4aec23912346c00d5ec2850dcd457a048479e Author: Jaime Date: 2014-10-20 (Mon, 20 Oct 2014) Changed paths: M numpy/linalg/linalg.py Log Message: ----------- Merge pull request #5210 from ewmoore/norm_doc MAINT: more informative error message from linalg.norm Compare: https://github.com/numpy/numpy/compare/df9db6ed5f9c...4ed4aec23912 From noreply at github.com Tue Oct 21 11:24:37 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 08:24:37 -0700 Subject: [Numpy-svn] [numpy/numpy] 3205c8: BUG: copy inherited masks in MaskedArray.__array_f... Message-ID: <54467ab57a8b1_20383fee1fa5b2a0851a0@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 3205c89ad3090c26fef8fd060117ee468c1b6c97 https://github.com/numpy/numpy/commit/3205c89ad3090c26fef8fd060117ee468c1b6c97 Author: Nathaniel J. Smith Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py M numpy/ma/core.py M numpy/ma/tests/test_core.py Log Message: ----------- BUG: copy inherited masks in MaskedArray.__array_finalize__ Previously, operations which created a new masked array from an old masked array -- e.g., np.empty_like -- would tend to result in the new and old arrays sharing the same .mask attribute. This leads to horrible brokenness in which writes to one array affect the other. In particular this was responsible for part of the brokenness that @jenshnielsen reported in gh-5184 in which np.gradient on masked arrays would modify the original array's mask. This fixes the worst part of the issues addressed in gh-3404, though there's still an argument that we ought to deprecate the mask-copying behaviour entirely so that empty_like returns an array with an empty mask. That can wait until we find someone who cares though. I also applied a small speedup to np.gradient (avoiding one copy); previously this inefficiency was masking (heh) some of the problems with masked arrays, so removing it is both an optimization and makes it easier to test that things are working now. Commit: 72e9072fadc048ef9fa4f76fbf37e31058b3e24a https://github.com/numpy/numpy/commit/72e9072fadc048ef9fa4f76fbf37e31058b3e24a Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py M numpy/ma/core.py M numpy/ma/tests/test_core.py Log Message: ----------- Merge pull request #5203 from njsmith/master BUG: copy inherited masks in MaskedArray.__array_finalize__ Compare: https://github.com/numpy/numpy/compare/4ed4aec23912...72e9072fadc0 From noreply at github.com Tue Oct 21 13:07:20 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 10:07:20 -0700 Subject: [Numpy-svn] [numpy/numpy] d90c6e: BUG: fix undefined ldexpl and frexpl with msvc Message-ID: <544692c8425_56b53f8b28c4d2c01946c@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: d90c6e018318720fac72f363cd9f49c2af77b159 https://github.com/numpy/numpy/commit/d90c6e018318720fac72f363cd9f49c2af77b159 Author: Julian Taylor Date: 2014-10-19 (Sun, 19 Oct 2014) Changed paths: M numpy/core/src/private/npy_config.h Log Message: ----------- BUG: fix undefined ldexpl and frexpl with msvc The functions are macros in msvc which we undefine and redeclare in npy_math. Since we now generate the ldexpl and frexpl ufuncs in gh-4852 this was broken by the extra logic for them in npy_config.h Commit: 96ef52e560c359eb0182682b56725e05da07f064 https://github.com/numpy/numpy/commit/96ef52e560c359eb0182682b56725e05da07f064 Author: Julian Taylor Date: 2014-10-19 (Sun, 19 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math.c.src Log Message: ----------- BUG: don't set invalid flag in cbrt fallback cbrt does not set the invalid flag on nan input as all inputs are valid. Fix the fallback to behave the same. Commit: a435de365a603f29bf243882644f1e26edda6e9a https://github.com/numpy/numpy/commit/a435de365a603f29bf243882644f1e26edda6e9a Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math.c.src M numpy/core/src/private/npy_config.h Log Message: ----------- Merge pull request #5207 from juliantaylor/msvc-fix MSVC fixes Compare: https://github.com/numpy/numpy/compare/72e9072fadc0...a435de365a60 From noreply at github.com Tue Oct 21 16:04:21 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 13:04:21 -0700 Subject: [Numpy-svn] [numpy/numpy] 0ddb8e: BUG: Use 'll' as the default length specifier for ... Message-ID: <5446bc45dbe8c_20553fa75ff452b8286e1@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 0ddb8e35ea464da08ca3327482d96c2ec4640583 https://github.com/numpy/numpy/commit/0ddb8e35ea464da08ca3327482d96c2ec4640583 Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- BUG: Use 'll' as the default length specifier for long long. The previous default was 'L', which seems gcc specific, whereas 'll' is the now recognized standard and is also accepted by gcc. Closes #5027. Commit: 0e091af94b03d5f3d5f5ddf45d54cf5ed2d7e612 https://github.com/numpy/numpy/commit/0e091af94b03d5f3d5f5ddf45d54cf5ed2d7e612 Author: Julian Taylor Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- Merge pull request #5212 from charris/fix-long-long-fmt BUG: Use 'll' as the default length specifier for long long. Compare: https://github.com/numpy/numpy/compare/a435de365a60...0e091af94b03 From noreply at github.com Tue Oct 21 16:05:16 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 13:05:16 -0700 Subject: [Numpy-svn] [numpy/numpy] 0ddb8e: BUG: Use 'll' as the default length specifier for ... Message-ID: <5446bc7c69639_24533fb0dd13b2bc64572@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 0ddb8e35ea464da08ca3327482d96c2ec4640583 https://github.com/numpy/numpy/commit/0ddb8e35ea464da08ca3327482d96c2ec4640583 Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- BUG: Use 'll' as the default length specifier for long long. The previous default was 'L', which seems gcc specific, whereas 'll' is the now recognized standard and is also accepted by gcc. Closes #5027. Commit: ddb5e790dc413e57cc9526eac1edab3651740f43 https://github.com/numpy/numpy/commit/ddb5e790dc413e57cc9526eac1edab3651740f43 Author: Julian Taylor Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- Merge pull request #5212 from charris/fix-long-long-fmt BUG: Use 'll' as the default length specifier for long long. Compare: https://github.com/numpy/numpy/compare/d310ff9720fe...ddb5e790dc41 From noreply at github.com Tue Oct 21 16:12:43 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 13:12:43 -0700 Subject: [Numpy-svn] [numpy/numpy] a4e68f: BUG: fix build error with MSVC 2013 caused by C99 ... Message-ID: <5446be3b92eeb_4fd73fc654dd52c088070@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: a4e68f411e4ec390b9250d9a8c24a8849a95ba9a https://github.com/numpy/numpy/commit/a4e68f411e4ec390b9250d9a8c24a8849a95ba9a Author: RobberPhex Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math_complex.c.src M numpy/core/src/npymath/npy_math_private.h Log Message: ----------- BUG: fix build error with MSVC 2013 caused by C99 complex support closes gh-4896 Commit: 555d803792a9f78db64961d6236eb0b01cd7af45 https://github.com/numpy/numpy/commit/555d803792a9f78db64961d6236eb0b01cd7af45 Author: Julian Taylor Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math_complex.c.src M numpy/core/src/npymath/npy_math_private.h Log Message: ----------- Merge pull request #5204 from RobberPhex/master fixed #4896 (fails to build with MSVC 2013) Compare: https://github.com/numpy/numpy/compare/ddb5e790dc41...555d803792a9 From noreply at github.com Tue Oct 21 16:12:33 2014 From: noreply at github.com (GitHub) Date: Tue, 21 Oct 2014 13:12:33 -0700 Subject: [Numpy-svn] [numpy/numpy] a4e68f: BUG: fix build error with MSVC 2013 caused by C99 ... Message-ID: <5446be31cb4d8_184b3fe4895ff2a064270@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: a4e68f411e4ec390b9250d9a8c24a8849a95ba9a https://github.com/numpy/numpy/commit/a4e68f411e4ec390b9250d9a8c24a8849a95ba9a Author: RobberPhex Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math_complex.c.src M numpy/core/src/npymath/npy_math_private.h Log Message: ----------- BUG: fix build error with MSVC 2013 caused by C99 complex support closes gh-4896 Commit: 06256714cb6209db5a3fac3681ebfad1ffe715de https://github.com/numpy/numpy/commit/06256714cb6209db5a3fac3681ebfad1ffe715de Author: Julian Taylor Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/npymath/npy_math_complex.c.src M numpy/core/src/npymath/npy_math_private.h Log Message: ----------- Merge pull request #5204 from RobberPhex/master fixed #4896 (fails to build with MSVC 2013) Compare: https://github.com/numpy/numpy/compare/0e091af94b03...06256714cb62 From noreply at github.com Fri Oct 24 15:21:43 2014 From: noreply at github.com (GitHub) Date: Fri, 24 Oct 2014 12:21:43 -0700 Subject: [Numpy-svn] [numpy/numpy] 0332a2: BUG: fix crash when using arange on datetime witho... Message-ID: <544aa6c73ef56_6ee93f9ae780f2c04559f@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 0332a2d8db6bd55b1d7e96c33ba4dd4f858d273d https://github.com/numpy/numpy/commit/0332a2d8db6bd55b1d7e96c33ba4dd4f858d273d Author: Julian Taylor Date: 2014-10-23 (Thu, 23 Oct 2014) Changed paths: M numpy/core/src/multiarray/datetime.c M numpy/core/tests/test_datetime.py Log Message: ----------- BUG: fix crash when using arange on datetime without dtype set Commit: 280f6050d2291e50aeb0716a66d1258ab3276553 https://github.com/numpy/numpy/commit/280f6050d2291e50aeb0716a66d1258ab3276553 Author: Charles Harris Date: 2014-10-24 (Fri, 24 Oct 2014) Changed paths: M numpy/core/src/multiarray/datetime.c M numpy/core/tests/test_datetime.py Log Message: ----------- Merge pull request #5225 from juliantaylor/datetime-arange-crash BUG: fix crash when using arange on datetime without dtype set Compare: https://github.com/numpy/numpy/compare/06256714cb62...280f6050d229 From noreply at github.com Fri Oct 24 16:08:56 2014 From: noreply at github.com (GitHub) Date: Fri, 24 Oct 2014 13:08:56 -0700 Subject: [Numpy-svn] [numpy/numpy] 0332a2: BUG: fix crash when using arange on datetime witho... Message-ID: <544ab1d8c15e2_13203ff484d1929c10231d@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 0332a2d8db6bd55b1d7e96c33ba4dd4f858d273d https://github.com/numpy/numpy/commit/0332a2d8db6bd55b1d7e96c33ba4dd4f858d273d Author: Julian Taylor Date: 2014-10-23 (Thu, 23 Oct 2014) Changed paths: M numpy/core/src/multiarray/datetime.c M numpy/core/tests/test_datetime.py Log Message: ----------- BUG: fix crash when using arange on datetime without dtype set Commit: ce0c6e04771f89a5baf7bd73f008dbe35f155ecc https://github.com/numpy/numpy/commit/ce0c6e04771f89a5baf7bd73f008dbe35f155ecc Author: Julian Taylor Date: 2014-10-24 (Fri, 24 Oct 2014) Changed paths: M numpy/core/src/multiarray/datetime.c M numpy/core/tests/test_datetime.py Log Message: ----------- Merge pull request #5225 from juliantaylor/datetime-arange-crash BUG: fix crash when using arange on datetime without dtype set Compare: https://github.com/numpy/numpy/compare/555d803792a9...ce0c6e04771f From noreply at github.com Fri Oct 24 17:45:06 2014 From: noreply at github.com (GitHub) Date: Fri, 24 Oct 2014 14:45:06 -0700 Subject: [Numpy-svn] [numpy/numpy] 892c46: BUG: Make PyArray_PutTo respect writeable flag. Message-ID: <544ac862290ec_39013ff75d3492b8102335@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 892c46910c7f9f0211f01b42f9aa81d0db892c8e https://github.com/numpy/numpy/commit/892c46910c7f9f0211f01b42f9aa81d0db892c8e Author: Jarl Haggerty Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/multiarray/item_selection.c Log Message: ----------- BUG: Make PyArray_PutTo respect writeable flag. Closes #4465. Commit: b1f8bcf451ef75344439d56c9953f6652af899d7 https://github.com/numpy/numpy/commit/b1f8bcf451ef75344439d56c9953f6652af899d7 Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/tests/test_multiarray.py Log Message: ----------- TST: Add some tests for ndarray.put. Commit: fe7816d448224633b0f8056ffb8c27478ab934e2 https://github.com/numpy/numpy/commit/fe7816d448224633b0f8056ffb8c27478ab934e2 Author: Julian Taylor Date: 2014-10-24 (Fri, 24 Oct 2014) Changed paths: M numpy/core/src/multiarray/item_selection.c M numpy/core/tests/test_multiarray.py Log Message: ----------- Merge pull request #5216 from charris/cleanup-gh-5132 BUG: Make PyArray_PutTo respect writeable flag. Compare: https://github.com/numpy/numpy/compare/280f6050d229...fe7816d44822 From noreply at github.com Fri Oct 24 17:46:41 2014 From: noreply at github.com (GitHub) Date: Fri, 24 Oct 2014 14:46:41 -0700 Subject: [Numpy-svn] [numpy/numpy] 892c46: BUG: Make PyArray_PutTo respect writeable flag. Message-ID: <544ac8c13977e_78e53f913027d2a011559e@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 892c46910c7f9f0211f01b42f9aa81d0db892c8e https://github.com/numpy/numpy/commit/892c46910c7f9f0211f01b42f9aa81d0db892c8e Author: Jarl Haggerty Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/src/multiarray/item_selection.c Log Message: ----------- BUG: Make PyArray_PutTo respect writeable flag. Closes #4465. Commit: b1f8bcf451ef75344439d56c9953f6652af899d7 https://github.com/numpy/numpy/commit/b1f8bcf451ef75344439d56c9953f6652af899d7 Author: Charles Harris Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/core/tests/test_multiarray.py Log Message: ----------- TST: Add some tests for ndarray.put. Commit: 42d890f549c6a7c8e70ac33c06f515ad9be0874a https://github.com/numpy/numpy/commit/42d890f549c6a7c8e70ac33c06f515ad9be0874a Author: Julian Taylor Date: 2014-10-24 (Fri, 24 Oct 2014) Changed paths: M numpy/core/src/multiarray/item_selection.c M numpy/core/tests/test_multiarray.py Log Message: ----------- Merge pull request #5216 from charris/cleanup-gh-5132 BUG: Make PyArray_PutTo respect writeable flag. Compare: https://github.com/numpy/numpy/compare/ce0c6e04771f...42d890f549c6 From noreply at github.com Sun Oct 26 08:02:27 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 05:02:27 -0700 Subject: [Numpy-svn] [numpy/numpy] 76fc3e: BUG: fix build in c99 mode Message-ID: <544ce2d378efa_281b3f8fc65a92a019de@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 76fc3e7df95779b2a67cf40210585cc1bb776591 https://github.com/numpy/numpy/commit/76fc3e7df95779b2a67cf40210585cc1bb776591 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- BUG: fix build in c99 mode The definition of off_t is located in sys/types.h and is not pulled by other files when compiling in c99 mode instead of the default gnu89. Closes gh-5231 Commit: fb40fe8618ff472d09ae819fd9734bd94cb9bdc9 https://github.com/numpy/numpy/commit/fb40fe8618ff472d09ae819fd9734bd94cb9bdc9 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- Merge pull request #5233 from juliantaylor/off_t-fix BUG: fix build in c99 mode Compare: https://github.com/numpy/numpy/compare/fe7816d44822...fb40fe8618ff From noreply at github.com Sun Oct 26 08:03:18 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 05:03:18 -0700 Subject: [Numpy-svn] [numpy/numpy] 76fc3e: BUG: fix build in c99 mode Message-ID: <544ce3068a9ff_7e2b3f9e7bf072c098460@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 76fc3e7df95779b2a67cf40210585cc1bb776591 https://github.com/numpy/numpy/commit/76fc3e7df95779b2a67cf40210585cc1bb776591 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- BUG: fix build in c99 mode The definition of off_t is located in sys/types.h and is not pulled by other files when compiling in c99 mode instead of the default gnu89. Closes gh-5231 Commit: 9b4b7b37efc740756ddfb36efca2f0c17642f001 https://github.com/numpy/numpy/commit/9b4b7b37efc740756ddfb36efca2f0c17642f001 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/core/include/numpy/npy_common.h Log Message: ----------- Merge pull request #5233 from juliantaylor/off_t-fix BUG: fix build in c99 mode Compare: https://github.com/numpy/numpy/compare/42d890f549c6...9b4b7b37efc7 From noreply at github.com Sun Oct 26 10:15:08 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 07:15:08 -0700 Subject: [Numpy-svn] [numpy/numpy] 8b58d9: BUG: io: genfromtxt did not handle filling_values=... Message-ID: <544d01ec41b4e_6eed3fcc732d32bc7949c@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 8b58d982d481fc362c8dbe81daf47bd7719095e3 https://github.com/numpy/numpy/commit/8b58d982d481fc362c8dbe81daf47bd7719095e3 Author: Warren Weckesser Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/lib/npyio.py M numpy/lib/tests/test_io.py Log Message: ----------- BUG: io: genfromtxt did not handle filling_values=0 correctly. Closes gh-2317. Commit: 079fcfc7b3f997a88d1f36f37adb4ccfc61f9526 https://github.com/numpy/numpy/commit/079fcfc7b3f997a88d1f36f37adb4ccfc61f9526 Author: David M Fobes Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- BUG: Fixes #5184 gradient calculation behavior at boundaries * Previous expected behavior was that the gradient is computed using central differences in the interior and first differences at the boundaries. * gradient was updated in v1.9.0 so that second-order accurate calculations are done at the boundaries, but this breaks expected behavior with old code, so `edge_order` keyword (Default: 1) is added to specify whether first or second order calculations at the boundaries should be used. * Since the second argument is *varargs, in order to maintain compatibility with old code and compatibility with python 2.6 & 2.7, **kwargs is used, and all kwargs that are not `edge_order` raise an error, listing the offending kwargs. * Tests and documentation updated to reflect this change. * Added `.. versionadded:: 1.9.1` to the new optional kwarg `edge_order` documentation, and specified supported `edge_order` kwarg values. * Clarified documentation for `varargs`. * Made indentation in docstring consistent with other docstring styles. * Examples corrected Commit: c6900b2a0dec4b50f35eed95d969bcd4cc2eb0bc https://github.com/numpy/numpy/commit/c6900b2a0dec4b50f35eed95d969bcd4cc2eb0bc Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/npyio.py M numpy/lib/tests/test_function_base.py M numpy/lib/tests/test_io.py Log Message: ----------- Merge pull request #5234 from juliantaylor/backports-1.9.1 Backports for 1.9.1 Compare: https://github.com/numpy/numpy/compare/9b4b7b37efc7...c6900b2a0dec From noreply at github.com Sun Oct 26 10:17:59 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 07:17:59 -0700 Subject: [Numpy-svn] [numpy/numpy] 3205c8: BUG: copy inherited masks in MaskedArray.__array_f... Message-ID: <544d029710ddd_4e863fe14353b2b88546c@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 3205c89ad3090c26fef8fd060117ee468c1b6c97 https://github.com/numpy/numpy/commit/3205c89ad3090c26fef8fd060117ee468c1b6c97 Author: Nathaniel J. Smith Date: 2014-10-21 (Tue, 21 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py M numpy/ma/core.py M numpy/ma/tests/test_core.py Log Message: ----------- BUG: copy inherited masks in MaskedArray.__array_finalize__ Previously, operations which created a new masked array from an old masked array -- e.g., np.empty_like -- would tend to result in the new and old arrays sharing the same .mask attribute. This leads to horrible brokenness in which writes to one array affect the other. In particular this was responsible for part of the brokenness that @jenshnielsen reported in gh-5184 in which np.gradient on masked arrays would modify the original array's mask. This fixes the worst part of the issues addressed in gh-3404, though there's still an argument that we ought to deprecate the mask-copying behaviour entirely so that empty_like returns an array with an empty mask. That can wait until we find someone who cares though. I also applied a small speedup to np.gradient (avoiding one copy); previously this inefficiency was masking (heh) some of the problems with masked arrays, so removing it is both an optimization and makes it easier to test that things are working now. Commit: 69707322db207b503fb6be8227071b2231cc7a25 https://github.com/numpy/numpy/commit/69707322db207b503fb6be8227071b2231cc7a25 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py M numpy/ma/core.py M numpy/ma/tests/test_core.py Log Message: ----------- Merge pull request #5203 from njsmith/master BUG: copy inherited masks in MaskedArray.__array_finalize__ Compare: https://github.com/numpy/numpy/compare/c6900b2a0dec...69707322db20 From noreply at github.com Sun Oct 26 10:41:39 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 07:41:39 -0700 Subject: [Numpy-svn] [numpy/numpy] 473a38: DOC: add release notes for 1.9.1 Message-ID: <544d0823a3689_283b3fe4c4f312c06783c@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 473a386db6b738ed89817542900656c8f79db9d2 https://github.com/numpy/numpy/commit/473a386db6b738ed89817542900656c8f79db9d2 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: A doc/release/1.9.1-notes.rst Log Message: ----------- DOC: add release notes for 1.9.1 [ci skip] Commit: b7cb545b2dafdd9c80baccbc4f172dd0f51e43ee https://github.com/numpy/numpy/commit/b7cb545b2dafdd9c80baccbc4f172dd0f51e43ee Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: A doc/release/1.9.1-notes.rst Log Message: ----------- Merge pull request #5235 from juliantaylor/release-notes-1.9.1 DOC: add release notes for 1.9.1 Compare: https://github.com/numpy/numpy/compare/69707322db20...b7cb545b2daf From noreply at github.com Sun Oct 26 10:54:46 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 07:54:46 -0700 Subject: [Numpy-svn] [numpy/numpy] 15a209: DOC: add 1.9.1 release notes to index and fix form... Message-ID: <544d0b36b0e86_30643f9fda4552b840410@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: 15a20970152d9c871a11a770a61bb0727002c2f9 https://github.com/numpy/numpy/commit/15a20970152d9c871a11a770a61bb0727002c2f9 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M doc/release/1.9.1-notes.rst M doc/source/release.rst Log Message: ----------- DOC: add 1.9.1 release notes to index and fix formatting [ci skip] From noreply at github.com Sun Oct 26 12:09:32 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 09:09:32 -0700 Subject: [Numpy-svn] [numpy/numpy] 0c6fd4: Fix strides when creating Python buffer Message-ID: <544d1cbc8ae35_7c8c3fddd237329c204cd@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 0c6fd4875acc733165ca71301c661535aaebed93 https://github.com/numpy/numpy/commit/0c6fd4875acc733165ca71301c661535aaebed93 Author: Michael Droettboom Date: 2014-09-19 (Fri, 19 Sep 2014) Changed paths: M numpy/core/src/multiarray/buffer.c M numpy/core/tests/test_multiarray.py Log Message: ----------- Fix strides when creating Python buffer Commit: ffacca0b2384fecbf383df137546db02f101e993 https://github.com/numpy/numpy/commit/ffacca0b2384fecbf383df137546db02f101e993 Author: Michael Droettboom Date: 2014-09-19 (Fri, 19 Sep 2014) Changed paths: M numpy/core/tests/test_multiarray.py Log Message: ----------- Explicitly set strides with negative number in test Commit: a7802ce227ac8e99286694afb8f0270281d0d32d https://github.com/numpy/numpy/commit/a7802ce227ac8e99286694afb8f0270281d0d32d Author: Michael Droettboom Date: 2014-09-26 (Fri, 26 Sep 2014) Changed paths: M numpy/core/src/multiarray/buffer.c Log Message: ----------- When requesting Fortan order, don't return C order. Commit: 7d1cb081c8677c5939995b700ab0ee34d5d0e154 https://github.com/numpy/numpy/commit/7d1cb081c8677c5939995b700ab0ee34d5d0e154 Author: Michael Droettboom Date: 2014-10-20 (Mon, 20 Oct 2014) Changed paths: M numpy/core/tests/test_multiarray.py Log Message: ----------- Add additional testing Commit: 256d256b865d42eace7d3b8a7e4cae7745fb9a2e https://github.com/numpy/numpy/commit/256d256b865d42eace7d3b8a7e4cae7745fb9a2e Author: Charles Harris Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M numpy/core/src/multiarray/buffer.c M numpy/core/tests/test_multiarray.py Log Message: ----------- Merge pull request #5086 from mdboom/stride-fix Fix how strides is set on Py_buffer object when the array is C-contiguous Compare: https://github.com/numpy/numpy/compare/fb40fe8618ff...256d256b865d From noreply at github.com Sun Oct 26 12:18:18 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 09:18:18 -0700 Subject: [Numpy-svn] [numpy/numpy] e2eaf9: BUG: Ugly fix for Apple's cblas_sgemv segfault Message-ID: <544d1ecab547c_524d3f816f6132a066743@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: e2eaf961dff8d57e3c12c19c39956a40754a1d0e https://github.com/numpy/numpy/commit/e2eaf961dff8d57e3c12c19c39956a40754a1d0e Author: Sturla Molden Date: 2014-10-23 (Thu, 23 Oct 2014) Changed paths: A numpy/core/blasdot/apple_sgemv_patch.c M numpy/core/setup.py M numpy/core/tests/test_blasdot.py M numpy/distutils/system_info.py Log Message: ----------- BUG: Ugly fix for Apple's cblas_sgemv segfault SGEMV in Accelerate framework will segfault on MacOS X version 10.9 (aka Mavericks) if arrays are not aligned to 32 byte boundaries and the CPU supports AVX instructions. This can produce segfaults in numpy.dot if we use numpy.float32 as dtype. This patch overshadows the symbols cblas_sgemv, sgemv_ and sgemv exported by Accelerate to produce the correct behavior. The MacOS X version and CPU specs are checked on module import. If Mavericks and AVX are detected the call to SGEMV is emulated with a call to SGEMM if the arrays are not 32 byte aligned. If the exported symbols cannot be overshadowed on module import, a fatal error is produced and the process aborts. All the fixes are in a self-contained C file and do not alter the _dotblas C code. The patch is not applied unless NumPy is configured to link with Apple's Accelerate framework. Commit: a373392d64eadca4e253eb5707a56e9db93e4e76 https://github.com/numpy/numpy/commit/a373392d64eadca4e253eb5707a56e9db93e4e76 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: A numpy/core/blasdot/apple_sgemv_patch.c M numpy/core/setup.py M numpy/core/tests/test_blasdot.py M numpy/distutils/system_info.py Log Message: ----------- Merge pull request #5223 from sturlamolden/accelerate-sgemv-fix BUG: Ugly fix for Apple's cblas_sgemv segfault Compare: https://github.com/numpy/numpy/compare/15a20970152d...a373392d64ea From noreply at github.com Sun Oct 26 12:21:09 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 09:21:09 -0700 Subject: [Numpy-svn] [numpy/numpy] a4edda: DOC: add Accelerate sgemv crash workaround to rele... Message-ID: <544d1f75c1d66_250f3fe21e5052b81979e@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: a4eddade30a84b23c9ddce7daac7f542c29c3eff https://github.com/numpy/numpy/commit/a4eddade30a84b23c9ddce7daac7f542c29c3eff Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M doc/release/1.9.1-notes.rst Log Message: ----------- DOC: add Accelerate sgemv crash workaround to release notes From noreply at github.com Sun Oct 26 13:08:20 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 10:08:20 -0700 Subject: [Numpy-svn] [numpy/numpy] c88fd9: REL: Release 1.9.1rc1 Message-ID: <544d2a84a1306_2faa3fd84cc472a0133dc@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: c88fd91482782f35a78d6144331471085b8695b0 https://github.com/numpy/numpy/commit/c88fd91482782f35a78d6144331471085b8695b0 Author: Julian Taylor Date: 2014-10-26 (Sun, 26 Oct 2014) Changed paths: M pavement.py M setup.py Log Message: ----------- REL: Release 1.9.1rc1 From noreply at github.com Sun Oct 26 13:08:44 2014 From: noreply at github.com (GitHub) Date: Sun, 26 Oct 2014 10:08:44 -0700 Subject: [Numpy-svn] [numpy/numpy] Message-ID: <544d2a9cdcb9_13f3ffaeeb632c0700a2@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/tags/v1.9.1rc1 Home: https://github.com/numpy/numpy From noreply at github.com Mon Oct 27 12:13:22 2014 From: noreply at github.com (GitHub) Date: Mon, 27 Oct 2014 09:13:22 -0700 Subject: [Numpy-svn] [numpy/numpy] 304d9b: Replacing list with generator compherension Message-ID: <544e6f22eac64_5db03fb3336f52b8550ed@hookshot-fe4-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 304d9b624745897bf3df13d7d368699066f3e70d https://github.com/numpy/numpy/commit/304d9b624745897bf3df13d7d368699066f3e70d Author: Yuval Langer Date: 2014-10-27 (Mon, 27 Oct 2014) Changed paths: M numpy/core/records.py Log Message: ----------- Replacing list with generator compherension Commit: 23ee379e86434518bc33ccd9e711a86188914de0 https://github.com/numpy/numpy/commit/23ee379e86434518bc33ccd9e711a86188914de0 Author: Charles Harris Date: 2014-10-27 (Mon, 27 Oct 2014) Changed paths: M numpy/core/records.py Log Message: ----------- Merge pull request #5238 from yuvallanger/patch-1 Replacing list with generator compherension Compare: https://github.com/numpy/numpy/compare/256d256b865d...23ee379e8643 From noreply at github.com Mon Oct 27 15:07:49 2014 From: noreply at github.com (GitHub) Date: Mon, 27 Oct 2014 12:07:49 -0700 Subject: [Numpy-svn] [numpy/numpy] 3a8276: BUG: Stricter checks for gufunc signatures Message-ID: <544e9805963b5_463b3fda148612c083794@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 3a827637c9eca604315491e8e017b72f2d1f8837 https://github.com/numpy/numpy/commit/3a827637c9eca604315491e8e017b72f2d1f8837 Author: jaimefrio Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/src/umath/ufunc_object.c Log Message: ----------- BUG: Stricter checks for gufunc signatures * Arguments, both input and output, must have at least as many dimensions as the corresponding number of core dimensions. In previous versions, 1's were prepended to the shape as needed. * Core dimensions with same labels must have exactly matching sizes. In previous versions, core dimensions of size 1 would broadcast against other core dimensions with the same label. * All core dimensions must have their size specified by a passed in input or output argument. In previous versions, core dimensions in an output argument that were not specified in an input argument, and whose size could not be inferred from a passed in output argument, would have their size set to 1. Commit: 140c50537087dccc764cd7540b46b039cb934530 https://github.com/numpy/numpy/commit/140c50537087dccc764cd7540b46b039cb934530 Author: jaimefrio Date: 2014-10-15 (Wed, 15 Oct 2014) Changed paths: M numpy/core/src/umath/umath_tests.c.src M numpy/core/tests/test_ufunc.py Log Message: ----------- TST: Stricter checks for gufunc signatures Added euclidean_pdist to umath/umath_tests.c.src. Modified tests to reflect new, stricter gufunc signature behavior. Commit: 528bac1380c782772b9de207bb8466b03117b96d https://github.com/numpy/numpy/commit/528bac1380c782772b9de207bb8466b03117b96d Author: jaimefrio Date: 2014-10-20 (Mon, 20 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M doc/source/reference/c-api.generalized-ufuncs.rst M doc/source/reference/c-api.ufunc.rst M numpy/core/src/umath/ufunc_object.c M numpy/core/src/umath/umath_tests.c.src M numpy/core/tests/test_ufunc.py Log Message: ----------- DOC: Stricter checks for gufunc signatures Documented the the new behavior in c-api.generalized-ufuncs.rst. Added PyUFunc_FromFuncAndDataAndSignature to c-api.ufunc.rst. Commit: 16575443239fa84615fc795692a79ef27f25c216 https://github.com/numpy/numpy/commit/16575443239fa84615fc795692a79ef27f25c216 Author: Charles Harris Date: 2014-10-27 (Mon, 27 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M doc/source/reference/c-api.generalized-ufuncs.rst M doc/source/reference/c-api.ufunc.rst M numpy/core/src/umath/ufunc_object.c M numpy/core/src/umath/umath_tests.c.src M numpy/core/tests/test_ufunc.py Log Message: ----------- Merge pull request #5077 from jaimefrio/gufuncs_core_dim_no_broadcast WIP: gufunc core dimensions should not broadcast Compare: https://github.com/numpy/numpy/compare/23ee379e8643...16575443239f From noreply at github.com Mon Oct 27 15:33:09 2014 From: noreply at github.com (GitHub) Date: Mon, 27 Oct 2014 12:33:09 -0700 Subject: [Numpy-svn] [numpy/numpy] 3d5f49: BUG: upcast weights to average result type to avoi... Message-ID: <544e9df5d361a_51453fee151f12b854294@hookshot-fe3-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 3d5f499fd6807186baf669eec491034766caab34 https://github.com/numpy/numpy/commit/3d5f499fd6807186baf669eec491034766caab34 Author: Malik Woods Date: 2014-10-27 (Mon, 27 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- BUG: upcast weights to average result type to avoid inaccuracies closes gh-5202 Commit: 7be18edfaae4d0ba6a888160d1cc57c453b568f2 https://github.com/numpy/numpy/commit/7be18edfaae4d0ba6a888160d1cc57c453b568f2 Author: Julian Taylor Date: 2014-10-27 (Mon, 27 Oct 2014) Changed paths: M numpy/lib/function_base.py M numpy/lib/tests/test_function_base.py Log Message: ----------- Merge pull request #5214 from mkowoods/np.average Update to average calculation Compare: https://github.com/numpy/numpy/compare/16575443239f...7be18edfaae4 From noreply at github.com Tue Oct 28 19:11:36 2014 From: noreply at github.com (GitHub) Date: Tue, 28 Oct 2014 16:11:36 -0700 Subject: [Numpy-svn] [numpy/numpy] b40e68: BUG: fix not returning out array from ufuncs with ... Message-ID: <545022a842024_39003fa103eb12a01030ae@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: b40e686f10421099400a533b343d6d1212f786e9 https://github.com/numpy/numpy/commit/b40e686f10421099400a533b343d6d1212f786e9 Author: Julian Taylor Date: 2014-10-28 (Tue, 28 Oct 2014) Changed paths: M numpy/core/src/umath/ufunc_object.c M numpy/core/tests/test_numeric.py M numpy/core/tests/test_umath.py M numpy/ma/core.py Log Message: ----------- BUG: fix not returning out array from ufuncs with subok=False set closes gh-5240 Commit: 3b22d87050ab63db0dcd2d763644d924a69c5254 https://github.com/numpy/numpy/commit/3b22d87050ab63db0dcd2d763644d924a69c5254 Author: Charles Harris Date: 2014-10-28 (Tue, 28 Oct 2014) Changed paths: M numpy/core/src/umath/ufunc_object.c M numpy/core/tests/test_numeric.py M numpy/core/tests/test_umath.py M numpy/ma/core.py Log Message: ----------- Merge pull request #5242 from juliantaylor/fix-ufunc-subok-out BUG: fix not returning out array from ufuncs with subok=False set Compare: https://github.com/numpy/numpy/compare/7be18edfaae4...3b22d87050ab From noreply at github.com Tue Oct 28 19:14:43 2014 From: noreply at github.com (GitHub) Date: Tue, 28 Oct 2014 16:14:43 -0700 Subject: [Numpy-svn] [numpy/numpy] b40e68: BUG: fix not returning out array from ufuncs with ... Message-ID: <54502363ae172_13da3ff1a78fb2bc867a7@hookshot-fe1-cp1-prd.iad.github.net.mail> Branch: refs/heads/maintenance/1.9.x Home: https://github.com/numpy/numpy Commit: b40e686f10421099400a533b343d6d1212f786e9 https://github.com/numpy/numpy/commit/b40e686f10421099400a533b343d6d1212f786e9 Author: Julian Taylor Date: 2014-10-28 (Tue, 28 Oct 2014) Changed paths: M numpy/core/src/umath/ufunc_object.c M numpy/core/tests/test_numeric.py M numpy/core/tests/test_umath.py M numpy/ma/core.py Log Message: ----------- BUG: fix not returning out array from ufuncs with subok=False set closes gh-5240 Commit: 994e98c598c19eb603ddd44eb8753b064adbd8cd https://github.com/numpy/numpy/commit/994e98c598c19eb603ddd44eb8753b064adbd8cd Author: Julian Taylor Date: 2014-10-29 (Wed, 29 Oct 2014) Changed paths: M numpy/core/src/umath/ufunc_object.c M numpy/core/tests/test_numeric.py M numpy/core/tests/test_umath.py M numpy/ma/core.py Log Message: ----------- Merge pull request #5242 from juliantaylor/fix-ufunc-subok-out BUG: fix not returning out array from ufuncs with subok=False set Compare: https://github.com/numpy/numpy/compare/c88fd9148278...994e98c598c1 From noreply at github.com Wed Oct 29 12:52:36 2014 From: noreply at github.com (GitHub) Date: Wed, 29 Oct 2014 09:52:36 -0700 Subject: [Numpy-svn] [numpy/numpy] 741990: DEP: deprecate pkgload Message-ID: <54511b547f8bb_7a1b3fbc2bdc72c037116@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: 7419906420307fc577a058a0c1a655869590eb01 https://github.com/numpy/numpy/commit/7419906420307fc577a058a0c1a655869590eb01 Author: alex Date: 2014-10-28 (Tue, 28 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M numpy/_import_tools.py Log Message: ----------- DEP: deprecate pkgload Commit: 4638475311228b099c278d5afc166c98c85bbc80 https://github.com/numpy/numpy/commit/4638475311228b099c278d5afc166c98c85bbc80 Author: alex Date: 2014-10-29 (Wed, 29 Oct 2014) Changed paths: M numpy/_import_tools.py Log Message: ----------- MAINT: warn instead of raise Commit: c63e1f470cf84a928499b2570a091f801cb67ba2 https://github.com/numpy/numpy/commit/c63e1f470cf84a928499b2570a091f801cb67ba2 Author: Charles Harris Date: 2014-10-29 (Wed, 29 Oct 2014) Changed paths: M doc/release/1.10.0-notes.rst M numpy/_import_tools.py Log Message: ----------- Merge pull request #5245 from argriffing/deprecate-pkgload DEP: deprecate pkgload Compare: https://github.com/numpy/numpy/compare/3b22d87050ab...c63e1f470cf8 From noreply at github.com Fri Oct 31 17:19:28 2014 From: noreply at github.com (GitHub) Date: Fri, 31 Oct 2014 14:19:28 -0700 Subject: [Numpy-svn] [numpy/numpy] b89849: Fix issue with fromarrays not correctly determinin... Message-ID: <5453fce0c2010_6f863f8d0c4372bc3372d@hookshot-fe2-cp1-prd.iad.github.net.mail> Branch: refs/heads/master Home: https://github.com/numpy/numpy Commit: b89849b1213e3384dfd72838672f5869321768db https://github.com/numpy/numpy/commit/b89849b1213e3384dfd72838672f5869321768db Author: Erik M. Bray Date: 2014-10-31 (Fri, 31 Oct 2014) Changed paths: M numpy/core/records.py M numpy/core/tests/test_records.py Log Message: ----------- Fix issue with fromarrays not correctly determining a format string for unicode data (in the process eliminate some unnecessary cruft) Commit: 0d30aef9065aed867e9276942bf0a51df6e48bf6 https://github.com/numpy/numpy/commit/0d30aef9065aed867e9276942bf0a51df6e48bf6 Author: Charles Harris Date: 2014-10-31 (Fri, 31 Oct 2014) Changed paths: M numpy/core/records.py M numpy/core/tests/test_records.py Log Message: ----------- Merge pull request #5251 from embray/astropy-3052 Issue with fromarrays not using correct format for unicode arrays Compare: https://github.com/numpy/numpy/compare/c63e1f470cf8...0d30aef9065a