[NumPy-Tickets] [NumPy] #2047: f2py needs to be updated to use non-deprecated macros/functions

[NumPy Trac]

#2047: f2py needs to be updated to use non-deprecated macros/functions
------------------------+---------------------------------------------------
 Reporter:  charris     |       Owner:  pearu         
     Type:  task        |      Status:  needs_decision
 Priority:  normal      |   Milestone:  1.7.0         
Component:  numpy.f2py  |     Version:  devel         
 Keywords:              |  
------------------------+---------------------------------------------------

Comment(by charris):

 Hi Pearu,

 Sorry I missed your first updates here. I think starting with master is a
 good place, I've already merged all the non-f2py stuff. As to the larger
 problem of swapping, perhaps we need some support from Numpy, which is why
 I signalled Mark when I put up the branch. The problem is that the ndarray
 internals are eventually going to be hidden, so we need some sort of
 function to do what you are doing, if that is really how it needs to be
 done, and that function needs to be designed.

 3) would be preferable, but I don't think you can pull it off.
 2) how would this break code? Do coders reference swap_array directly?
 1) I'm thinking this is the way we might have to go. Why can't this be in
 1.7?

 Swapping smart pointer contents is pretty standard in C++, even in the
 stdlib. Call with one pointer, initialize your new structure, swap
 pointers, exit, and the destructor cleans up the passed structure. I don't
 see why Numpy shouldn't have something with similar functionality, but as
 you say, it needs discussion and design.

 Perhaps the best thing at this point would be to post on the list,
 describe what you do and why you want to do it, and ask for input. I'd
 really like Mark to comment, but I'm sure other folks would have something
 to say.

-- 
Ticket URL: <http://projects.scipy.org/numpy/ticket/2047#comment:8>
NumPy <http://projects.scipy.org/numpy>
My example project