Re: [Bug-gsl] interpolation function

[Michael Ng]

Thanks Brian. I have found a solution to my problem. Basically, when two 
lines intersect (LineA is a flat line and LineB is a slanted line) at a 
given y-coordinate, instead of getting the x-coordinate from 
interpolation in LineA (my old problematic approach) I simply get the 
x-coordinate from interpolation in LineB (new approach). Hence, in this 
case, the initial problem is solved. In fact, it is not possible to get 
a specific x-coordinate from a flat line, since all x-coordinates has 
the same y-coordinate!

Thanks for your help.
  

Regards,
Michael Ng



Brian Gough wrote:
> At Fri, 27 Mar 2009 00:25:28 +0000,
> Michael Ng wrote:
>   
> > Dear gurus,
> >     I am using the following function to get interpolation points in a 
> > 3D surface. I am doing it for the sake of finding the intersection line 
> > between two surfaces:
> >
> >     gsl_spline_eval(spline, x, acc);
> >
> >     Unfortunately, the abave line failed to work when the data 'x' is a 
> > constant, since it is from a flat surface. I understand that x has to be 
> > in ascending order, but I hope to find a quick fix when x is a constant...
> >     
>
> As you say x must be increasing, so the only way is to remove any
> equal points before doing the interpolation.
>
>