Exact quadratic prediction by the Sibson interpolation method

[stla]

Hello,

The doc claims:

However I tried with a general quadratic form a + bx*x + by*y + cxx*x^2 + cyy*y^2 + cxy*x*y and I reproduced the exact values by interpolation. Are you sure of this restriction?

[stla]

Still checked and that's not true. It was a particular case.

[stla]

But seems ok when cxy=0.

[lrineau]

I doubt there are many CGAL developers familiar with interpolation methods. Can you please post your C++ code?

[stla]

My code is the same as the one in this example. But maybe I still was in a particular case.