I hope that one of the developers would be able to share information on how current implementation of spline components works under the hood.
More specifically what kind of class of curves is used - is it one of the Bézier, Cubic Hermite or maybe Catmull-Rom?
What is the mathematical relationship between tangents of control points - how they affect curve? Answer to this can be obvious from class of curve but maybe tangents are derived from some other parameters, just to make them easier to manipulate visually.
I could solve my problem by implementing my own splines, but there are so much functionality behind them already that it would be like re-inventing the bicycle.
The reason why I’m asking is that I need to develop (or find) an algorithm to manipulate looped spline with length constraint.
To illustrate this:
When control points at the bottom change their position, total length of the spline becomes longer or shorter, which is undesirable.
The purpose of the algorithm is to re-calculate positions and tangents of control points to conform to length constraint.
The idea is to have a several classes of control points:
- Static, which are locked into specific place with fixed tangents
- Static Position Scalable Tangents, locked into position but tangents can be scaled
- Restricted Position Scalable Tangent, this control points are allowed to move only in specific axis and their tangents can be scaled.
Right now I have an idea how to solve this iteratively but if I knew the math behind the current implementation of splines I could do it analytically.

