Re: [Bug-gsl] Evaluation of B-Splines

[Rhys Ulerich]

Hi Iryna,

> I'm not really shure if it's a bug, but I cann't find other reason for the
> following error.
>
> ...
> //construct five knots {0, 1, 2, 3, 4}
> gsl_bspline_knots_uniform(0, 4, w);
>
> The values of the first and the last two knots don't get
> evaluated properly. Does anybody know the reasons for such a behavior?

As we've been discussing in the thread
http://lists.gnu.org/archive/html/help-gsl/2012-01/msg00009.html, the
behavior is correct and comes from the distinction between "knots" and
"breakpoints".

Let's keep the discussion in that original thread.

Thanks,
Rhys