[NumPy-Tickets] [NumPy] #1863: Buggy results when operating on array copied with astype()

[NumPy Trac]

#1863: Buggy results when operating on array copied with astype()
-------------------------+--------------------------------------------------
  Reporter:  embray      |       Owner:  somebody
      Type:  defect      |      Status:  closed  
  Priority:  normal      |   Milestone:  1.6.1   
 Component:  numpy.core  |     Version:  1.6.0   
Resolution:  fixed       |    Keywords:          
-------------------------+--------------------------------------------------
Changes (by mwiebe):

  * status:  new => closed
  * version:  devel => 1.6.0
  * resolution:  => fixed
  * milestone:  Unscheduled => 1.6.1


Comment:

 This bug turns out to be much more longstanding than is implied by it
 showing up in astype. It's a bit trickier, but possible to expose it in
 1.6 as well:

 {{{
 >>> np.__version__
 '1.6.1.dev-878ab94'
 >>> a = np.arange(24).reshape(2,3,4).swapaxes(0,1)
 >>> a.strides
 (16, 48, 4)
 >>> b = np.zeros_like(a)
 >>> b.strides
 (16, 48, 4)
 >>> a.flags.c_contiguous
 False
 >>> b.flags.c_contiguous
 True
 }}}

 The C_CONTIGUOUS flag should not be set in this case, and that's what was
 happening in the bug report example too, just with a simpler array and
 different function call.

 Fixed in commit:a1e7be3 (master) and commit:5765325 (1.6.1).

-- 
Ticket URL: <http://projects.scipy.org/numpy/ticket/1863#comment:1>
NumPy <http://projects.scipy.org/numpy>
My example project