Skip to content
Commits on Jul 11, 2011
  1. @rgommers

    REL: set version to 1.6.1rc3

    rgommers committed
  2. @charris

    BUG: iter: writeonly operands weren't always being buffered correctly…

    Mark Wiebe committed with charris
    … (Ticket #1895/1896)
    
    This interacted with the NPY_ITER_GROWINNER flag which would incorrectly
    grow the length because it thought nothing was being buffered.
Commits on Jul 1, 2011
  1. @rgommers

    REL: set version to 1.6.1rc2.

    rgommers committed
  2. @rgommers

    BUG: deal with shared lib extension in a single place. Fix ctypes.loa…

    rgommers committed
    …d_library.
    
    This is related to PEP 3149, tags in shared library extensions.  Only applies
    to Linux (for now).  See also #1749.
  3. @rgommers @charris
Commits on Jun 29, 2011
  1. @mwiebe

    BUG: nditer: Error in detecting double reduction loop (Fixes #1885)

    mwiebe committed
    This bug was in the double reduction loop optimization used during
    buffering, and was triggered by the bug report example because it
    reduced everything to a single value.
Commits on Jun 22, 2011
  1. @pearu @rgommers

    BUG[f2py]: fix --include_paths bug. Deprecated --include_paths in fav…

    pearu committed with rgommers
    …or of --include-paths. Updated docs.
  2. @mwiebe @rgommers

    BUG: core: Fix corner case of object array assignment (#1870)

    mwiebe committed with rgommers
    This general aspect of object arrays is fraught with errors, for
    example, eval(repr(x)) produces a 2 instead of 1 dimensional array
    when given the x in the first example of the bug. This fix at least
    makes the result requested in the bug report work.
  3. @charris @rgommers

    ENH: Add the polynomial module to the documentation.

    charris committed with rgommers
    Also:
    1. Note that the polynomial package is preferred over poly1d.
    
    2. Remove deprecation of mapparms in __init__.py as it interferes with the
       documention of the method of the same name. This is probably safe as it
       is unlikely to be used yet.
    
    3. Make some improvements to the documentation in polytemplate.
  4. @rgommers
Commits on Jun 13, 2011
  1. @rgommers

    REL: tag 1.6.1rc1

    rgommers committed
Commits on Jun 9, 2011
  1. BUG: core: PyArray_NewFromDescr needs to update flags when strides !=…

    Mark Wiebe committed
    … NULL (fixes #1863
    
    This appears to have been a longstanding bug, but has come to the
    surface because PyArray_NewLikeArray uses the function with a non-NULL
    strides parameter and is used more frequently then other such uses.
Commits on Jun 2, 2011
  1. @rgommers

    ENH: distutils: do the correct thing when swig_opts is a string in se…

    rgommers committed
    …tup.py
    
    It should be a list, but until numpy 1.5.1 a string was accepted. This broke in
    1.6.0, this commit unbreaks things.
    
    Closes #1851. (backport of bf1e6f3)
Commits on May 30, 2011
  1. @pearu @rgommers

    BUG: fix f2py size variadic macro for Visual C++ 2008 compiler. Also …

    pearu committed with rgommers
    …be verbose on unspecified use modules.
    
    (backport of 026bc92)
Commits on May 21, 2011
  1. @mwiebe
  2. @mwiebe

    BUG: Iterator reduction buffering bug when the inner loop is bigger t…

    mwiebe committed
    …han the buffer size (ticket #1837)
    
    The symptoms of this bug were showing up only for a greater number of operands
    because einsum runs specialized loops for fewer operands and dimensions.
Commits on May 18, 2011
  1. @mwiebe

    BUG: PyArray_FillWithZero didn't work in the general case

    mwiebe committed
    The call to the stransfer function had incorrect arguments. This
    didn't show up because all the tests of einsum, the only current
    user of the function, produced contiguous arrays which are special
    cased.
  2. @mwiebe
  3. @mwiebe

    TST: Reduced test case for ticket #1834

    mwiebe committed
    The ticket's bug is visible in einsum, but is caused by an error in how
    the reduction double loop interacts with buffering in the nditer. This
    test case directly tests the nditer triggering the bad code.
Commits on May 14, 2011
  1. @rgommers
  2. @rgommers

    REL: release Numpy 1.6.0

    rgommers committed
Commits on May 7, 2011
  1. @rgommers

    REL: set version to 1.6.0rc3.

    rgommers committed
  2. @rgommers

    use np.atleast_Nd() to boost dimensions to ndmin

    Derek Homeier committed with rgommers
  3. @rgommers

    changed ndmin option in loadtxt to return shape (1, X.size) for singl…

    Derek Homeier committed with rgommers
    …e-row inputs
  4. @pearu @rgommers

    BUG: Fix the order of declaring variables in f2py generated code. The…

    pearu committed with rgommers
    … bug was noticable with ifort but not with gfortran.
  5. @pearu @rgommers
  6. @pearu @rgommers

    BUG: Fix two argument size support for Fortran module routines. Rever…

    pearu committed with rgommers
    …ted size-to-shape mapping patch and implemented two argument size function in C.
  7. @pearu @rgommers
Commits on May 3, 2011
  1. @rgommers

    REL: set version to 1.6.0rc2.

    rgommers committed
Commits on May 2, 2011
  1. @mdboom @mwiebe
  2. @mdboom @mwiebe

    BUG: Fix a bug where memory was being read after being freed.

    mdboom committed with mwiebe
    On my system (Python 2.7, RHEL5), the call to func.__name__
    returns a temporary Python string object with only a single
    reference.  Dereferencing it before copying out its contents
    results in reading freed memory.
  3. @mdboom @mwiebe
  4. @mdboom @mwiebe
  5. @mdboom @mwiebe
  6. @mwiebe

    BUG: Fix 1.6rc1 crash in result_type on 32-bit Windows MKL builds

    mwiebe committed
    Thanks to Christoph Gohlke for tracking down the error. It appears to
    stem from the C preprocessor making the use of an undefined symbol
    become arbitrary behavior rather than a compile error.
Something went wrong with that request. Please try again.