Re: [Swig-user] Returning an array to python numpy gives segmentation fault

[David F. <dav...@gm...>]

On Mon, 30 Jul 2012 00:04:34 +0200, Oliver Willekens <oli...@el...> wrote:
> 2012/7/29 David Froger <dav...@gm...>
> 
> > Hi Oliver,
> >
> > > the numpy.i interface provides us with most of the items I need for a
> > > project, however, I have a case whereby I build an array in C++ and need
> > to
> > > pass it to numpy. The size of the array is unknown at the beginning, so I
> > > cannot use the numpy.i typedefs for ARGOUT_ARRAY1 (as these require you
> > to
> > > supply the dimension of the array to be returned).
> >
> > Could argoutview arrays be what you need?
> >
> > http://docs.scipy.org/doc/numpy/reference/swig.interface-file.html#argoutview-arrays
> >
> 
> I don't think so, unless I'm missing the point of argoutview. From the
> example at http://www.scipy.org/Cookbook/SWIG_NumPy_examples , it looks
> like here too the size of the array to be returned is known at compile
> time, which is not the case.

Actually,  is this example,  the  typemap is not applied on  my_ones (which take
the size of the array as argument),  but on my_set_ones (which does not take the
size of the array as argument), and then my_set_ones is rename to set_ones...

But another question:  Do you want the C++ array to be copied in the numpy array
(safe,  ARGOUT_ARRAY1 + a helper function  should words),  or the numpy array to
point on the  C++ array (hard to  make it safe,  only required is  array is very
big and can not be duplicated, ARGOUTVIEW_ARRAY1 should works) ?