[Numpy-svn] [numpy/numpy] 74b9f5: BUG: ticket #2063, make unique return consistent i...

GitHub noreply at github.com
Tue May 1 18:10:41 EDT 2012


  Branch: refs/heads/backport-core
  Home:   https://github.com/numpy/numpy
  Commit: 74b9f5eef8fac643bf9012dbb2ac6b4b19f46892
      https://github.com/numpy/numpy/commit/74b9f5eef8fac643bf9012dbb2ac6b4b19f46892
  Author: Bryan Van de Ven <bryan at laptop.local>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/tests/test_regression.py
    M numpy/lib/arraysetops.py

  Log Message:
  -----------
  BUG: ticket #2063, make unique return consistent index.

Make unique use  mergesort when return_index is true. This quarantees that
the returned indices are of the first occurrence of the unique elements and
makes the behavior better defined and consistent.


  Commit: ff72395969c2c2a6abe39de8d78bfd03c4b0705d
      https://github.com/numpy/numpy/commit/ff72395969c2c2a6abe39de8d78bfd03c4b0705d
  Author: Ben Root <ben.v.root at gmail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/include/numpy/ndarraytypes.h
    M numpy/core/src/multiarray/arraytypes.c.src
    M numpy/core/src/multiarray/calculation.c
    M numpy/core/src/multiarray/usertypes.c
    M numpy/core/tests/test_multiarray.py

  Log Message:
  -----------
  BUG: ticket #1149, make argmin work for datetime.

Backport of 6fc0737 'ENH: Explicitly coded argmin for timedelta'


  Commit: 5dff2f91124a635292f315c4db40fc6365b4a6ba
      https://github.com/numpy/numpy/commit/5dff2f91124a635292f315c4db40fc6365b4a6ba
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/src/umath/funcs.inc.src
    M numpy/core/tests/test_multiarray.py
    M numpy/core/tests/test_umath.py

  Log Message:
  -----------
  BUG: Fix maximum/minimum for object arrays in Python 3.

Use PyObject_RichCompareBool instead of PyObject_Cmp.


  Commit: 961a1dd3e4ae800cb54aada46df0deb797083c09
      https://github.com/numpy/numpy/commit/961a1dd3e4ae800cb54aada46df0deb797083c09
  Author: Hrvoje Niksic <hrvoje.niksic at avl.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/ctors.c
    M numpy/core/tests/test_multiarray.py

  Log Message:
  -----------
  BUG: ticket #1138

This is a backport of 4daf949

PyArray_FromBuffer: Allow creating arrays from empty buffers
or empty slices.

That finally closed the ticket.


  Commit: febfb6ea928b9d95b251cf0720ac77a45263b65b
      https://github.com/numpy/numpy/commit/febfb6ea928b9d95b251cf0720ac77a45263b65b
  Author: Ralf Gommers <ralf.gommers at googlemail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/shape_base.py

  Log Message:
  -----------
  DOC: ticket #1446, correct note about correspondence vstack and
concatenate.


  Commit: a292a7223072e07757542c0dfea32e866f3b5332
      https://github.com/numpy/numpy/commit/a292a7223072e07757542c0dfea32e866f3b5332
  Author: Ralf Gommers <ralf.gommers at googlemail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/tests/test_numeric.py

  Log Message:
  -----------
  BUG: Ticket #1672 fix allclose to work for scalar inf.

Backport of 7497d60.

Also actually runs tests. Tests were not run before because the allclose
test class was a subclass of TestCase and used generators. This is not
supported by nose, tests will be skipped.

Also changes plain asserts to assert_().

Thanks to Justin Peel for the allclose inf fix.


  Commit: 5fe46bd36296a7d214cf8e7b6cceac3f5c43cc68
      https://github.com/numpy/numpy/commit/5fe46bd36296a7d214cf8e7b6cceac3f5c43cc68
  Author: Mark Wiebe <mwwiebe at gmail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/scalartypes.c.src
    M numpy/core/tests/test_print.py

  Log Message:
  -----------
  ENH: ticket #1675, Add scalar support for the format() function.

Backport of 88e8c15.


  Commit: 5f940ca3d65ec6e568dc97380a2bc3957a1add8e
      https://github.com/numpy/numpy/commit/5f940ca3d65ec6e568dc97380a2bc3957a1add8e
  Author: Ralf Gommers <ralf.gommers at googlemail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/lib/function_base.py
    M numpy/lib/tests/test_function_base.py

  Log Message:
  -----------
  BUG: ticket #1747, make np.median() work for 0-D arrays.


  Commit: 88e7e52de62e54f899348784c607b6173c172efd
      https://github.com/numpy/numpy/commit/88e7e52de62e54f899348784c607b6173c172efd
  Author: Pauli Virtanen <pav at iki.fi>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/src/scalarmathmodule.c.src
    M numpy/core/src/umath/loops.c.src
    M numpy/core/tests/test_scalarmath.py
    M numpy/core/tests/test_umath.py

  Log Message:
  -----------
  BUG: ticket #1776, make complex division by zero to yield inf properly.


  Commit: b397dde2e8a009fec5e8ec8951e564a20cb70094
      https://github.com/numpy/numpy/commit/b397dde2e8a009fec5e8ec8951e564a20cb70094
  Author: Ralf Gommers <ralf.gommers at googlemail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/numeric.py

  Log Message:
  -----------
  BUG: ticket #1905, explicitly check for NaNs in allclose().

Also clean up the logic behind handling infs.
Squashed backport of 3503c5f..aea76b8.


  Commit: 4468440463291f1fd6a4938389380df9316a0185
      https://github.com/numpy/numpy/commit/4468440463291f1fd6a4938389380df9316a0185
  Author: Ralf Gommers <ralf.gommers at googlemail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/defchararray.py
    M numpy/core/tests/test_defchararray.py

  Log Message:
  -----------
  BUG: ticket #1948, fix regression for indexing chararrays with empty list.


  Commit: 4762f4a349e8d5d7d2d728710f9e11fccac6a50b
      https://github.com/numpy/numpy/commit/4762f4a349e8d5d7d2d728710f9e11fccac6a50b
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-26 (Thu, 26 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/number.c

  Log Message:
  -----------
  BUG: array_divmod did not check for Py_NotImplemented.

The consequence of this bug was the __rdivmod__ wasn't called for user
classes that defined __array_priority__.


  Commit: 89d6d411d1bcd22524d55a042c6cc2a64c36f680
      https://github.com/numpy/numpy/commit/89d6d411d1bcd22524d55a042c6cc2a64c36f680
  Author: David Cournapeau <cournape at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/descriptor.c
    M numpy/core/src/multiarray/hashdescr.c
    M numpy/core/tests/test_dtype.py

  Log Message:
  -----------
  BUG: ticket #2017, fix type hash.

This is a squashed backport of 4 commits.
STY: cleanup hashdesc.c to follow our C conventions.

BUG: fix inconsistencies in dtype flag type at the C level.

Conflicts:

	numpy/core/src/multiarray/descriptor.c
	numpy/core/src/multiarray/hashdescr.c

BUG: fix #2017 by ignoring type_num in the hash input.

Since type_num is not considered in PyArray_EquivTypes (for dtype
equality comparison), seems reasonable to ignore it for hashing as
well.

Conflicts:

	numpy/core/tests/test_dtype.py

BUG: fix flags type when exposed to python.


  Commit: ec753165aadde35abc15d7a3d4641d566cad44b4
      https://github.com/numpy/numpy/commit/ec753165aadde35abc15d7a3d4641d566cad44b4
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/calculation.c
    M numpy/core/src/multiarray/calculation.h
    M numpy/core/src/multiarray/methods.c

  Log Message:
  -----------
  BUG: ticket #1952, allow floating ddof in std and var.

Also return nan for resulting non-positive degrees of freedom.


  Commit: 666c02d8628eb354e5efb3321f0b568b1fa81feb
      https://github.com/numpy/numpy/commit/666c02d8628eb354e5efb3321f0b568b1fa81feb
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/number.c
    M numpy/core/tests/test_umath.py

  Log Message:
  -----------
  BUG: ticket #2033, a**2.0 has incorrect type.

Backport of PR #207.


  Commit: 9a34be0d317d96687742a43d0a41b84d2c1bf8d4
      https://github.com/numpy/numpy/commit/9a34be0d317d96687742a43d0a41b84d2c1bf8d4
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/getset.c
    M numpy/core/tests/test_multiarray.py

  Log Message:
  -----------
  BUG: ticket #2046, deleting array attributes causes segfault.

Add check for NULL in setter functions and return Attribute error
if found.


  Commit: 7fcf24ce7b5a63c736a421fc00627d8e405e0eee
      https://github.com/numpy/numpy/commit/7fcf24ce7b5a63c736a421fc00627d8e405e0eee
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/descriptor.c
    M numpy/core/src/multiarray/flagsobject.c
    M numpy/core/src/multiarray/getset.c
    M numpy/core/src/multiarray/iterators.c
    M numpy/core/src/multiarray/nditer_pywrap.c

  Log Message:
  -----------
  BUG: ticket #2045, make attribute/iterator_element deletions not segfault.

Attempted attribute deletion raises an Attribute error. Deleting
an iterator index or slice raises a TypeError. This is a change for
nditer, which used to raise ValueError when deletes were attempted.


  Commit: fc2d6c8d929d4dd9ef7e0b39a61e234027608fba
      https://github.com/numpy/numpy/commit/fc2d6c8d929d4dd9ef7e0b39a61e234027608fba
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/tests/test_scalarmath.py
    M numpy/core/tests/test_umath.py

  Log Message:
  -----------
  TST: Silence some test warnings.


  Commit: 6be9dcf443551b8f28ea08a1981f9cb2806f4962
      https://github.com/numpy/numpy/commit/6be9dcf443551b8f28ea08a1981f9cb2806f4962
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/tests/test_multiarray.py
    M numpy/core/tests/test_numeric.py

  Log Message:
  -----------
  PY3: Fix for python3.2.


  Commit: 75086316357286753f9b3dcafd34d2f5fa9b83cd
      https://github.com/numpy/numpy/commit/75086316357286753f9b3dcafd34d2f5fa9b83cd
  Author: Charles Harris <charlesr.harris at gmail.com>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/convert_datatype.c
    M numpy/core/src/multiarray/ctors.c
    M numpy/core/src/multiarray/item_selection.c
    M numpy/core/tests/test_regression.py

  Log Message:
  -----------
  BUG: ticket #2021. Fix segfault.

The return from PyArray_DescrFromObject PyArray_SearchSorted needed
to be checked for NULL. Also added checks in other spots where
PyArray_DescrFromObject was called. Added regression test for same.
Made a few small code cleanups along the way.

All the checks may not be necessary, but
1) No one should have to waste brain cycles figuring that out.
2) Bugs happen.
3) Things change.


  Commit: 25b021de0247df018566112eec563a73ad7f63f1
      https://github.com/numpy/numpy/commit/25b021de0247df018566112eec563a73ad7f63f1
  Author: Nicolas Pinto <pinto at alum.mit.edu>
  Date:   2012-04-29 (Sun, 29 Apr 2012)

  Changed paths:
    M numpy/core/src/multiarray/common.c
    M numpy/core/tests/test_half.py

  Log Message:
  -----------
  BUG: ticket #2073, fix float16 __array_interface__ bug.

Squashed backport also containing two following commits.

TST: Test float16 __array_interface__.
STY: Cleanup test_half.py a bit.


  Commit: 34f27c1e74871419f307591fe89726fa9588174f
      https://github.com/numpy/numpy/commit/34f27c1e74871419f307591fe89726fa9588174f
  Author: David Warde-Farley <wardefar at iro.umontreal.ca>
  Date:   2012-05-01 (Tue, 01 May 2012)

  Changed paths:
    M numpy/core/src/multiarray/mapping.c

  Log Message:
  -----------
  BUG: PR #188, PyArray_{Get,Set}Map using low-precision counter

Should be using npy_intp rather than int.


Compare: https://github.com/numpy/numpy/compare/74b9f5e^...34f27c1


More information about the Numpy-svn mailing list