[f2py] column major versus row major storage question...

[pearu at cens.ioc.ee]

On Tue, 8 Feb 2005, Louis Wicker wrote:

> All:
> 
> this is probably another one of my silly questions.....
> 
> If you create an array in Python, such as
> 
> sC = array([[1,2,3],[4,5,6]])
> 
> and print it out it shows
> 
> [[1 2 3]
>   [4 5 6]]
> 
> now, suppose I try and convert that array to a column major storage via
> 
> sF = fortran.as_column_major_storage(sC)
> 
> and if the data transposition is done, then should not the print out of  
> sF be
> 
> [[1 3 5]
>   [2 4 6]]   ???
> 
> its not...

And should not be. The argument of as_column_major_storage and its return
value are equivalent when accessing from Pyhton side. Just when argument
is C contiguous then the return value will be Fortran contiguous. So, the
following behaviour is correct:

In [4]: a=array([[1,2,3],[4,5,6]])

In [5]: a 
Out[5]: 
array([[1, 2, 3],
       [4, 5, 6]])

In [6]: b=as_column_major_storage(a)

In [7]: b
Out[7]: 
array([[1, 2, 3],
       [4, 5, 6]])

In [8]: transpose(a)
Out[8]: 
array([[1, 4],
       [2, 5],
       [3, 6]])

In [9]: transpose(b)
Out[9]: 
array([[1, 4],
       [2, 5],
       [3, 6]])

In [11]: a.iscontiguous()
Out[11]: 1

In [12]: b.iscontiguous()
Out[12]: 0

In [13]: transpose(b).iscontiguous()
Out[13]: 1

Pearu