[SciPy-dev] Re: f2py and linalg (version 1)

Pearu Peterson peterson at math.utwente.nl
Thu Mar 7 13:52:42 EST 2002


On Thu, 7 Mar 2002, Travis Oliphant wrote:

>
> O.K.
>
> So, now I see that inspite of the error message.
>
> ev(a) works if a is Not Contiguous
>                    =====
> but
>
> ev(A) does not work if it is contiguous
>                          ====
>
> This seems to have something to do with
>
> intent(inout)
>
> I know, we aren't supposed to use this anymore.  So, is behavior broken if
> the variable is defined this way?
>
> -Travis
>
>
> Is this the same behavior on your system?

Yes, the error messages from old interface+new f2py are totally
misleading.

Pearu





More information about the SciPy-Dev mailing list