This is the mail archive of the
gsl-discuss@sourceware.cygnus.com
mailing list for the GSL project.
Re: Vector and matrix views (and bug report)
Ok, I have checked in a fix for that. At some point we'll need to
write tests for the new matrix/vector functions. I have been adding
quite a few untested functions in there too.
Fabrice Rossi writes:
> Brian Gough wrote:
> >
> > Thanks. I will fix the bug and add those functions.
>
> By the way, there is a bug in my code. I've used base->data rather than
> base->block to allow a view to be extracted from a vector that has been itself
> extracted from a block or a matrix. The problem is that I've not taken into
> account the stride in the original vector. For gsl_vector_view_from_vector it
> simply means that the function shall start with
>
> stride *= base->stride;
>
> For gsl_matrix_view_from_vector, it is not possible to use such a feature,
> because there is no support in the matrix structure for a stride on the column
> index, so I guess base->stride should be tested to refuse the view if it's
> different from 1.
>
> Sorry for the error.
>
> Fabrice Rossi