A Rust implementation of the Delaunay triangulation algorithm presented by Paul Bourke.
Find the crate documentation on docs.rs, or here on Github.
This was developed as an exercise to get more used to Rust. As far as I know, it works, but it might not. Also, this is a O(n1.5) (approximatively) algorithm, it's not parallelized, and it doesn't use the GPU at all.
Add the rtriangulate dependency to Cargo.toml
:
[dependencies]
rtriangulate = "0.3"
And use the crate as such:
extern crate rtriangulate;
use rtriangulate::{TriangulationPoint, triangulate};
fn main() {
// A list of points (which has to be sorted on x).
// Note that you can use your own point type, just implement the rtriangulate::Point trait.
let points = [
TriangulationPoint::new(10.0, 50.0),
TriangulationPoint::new(25.0, 40.0),
TriangulationPoint::new(30.0, 40.0)
];
// In case you need to sort your points:
// points.sort_unstable_by(rtriangulate::sort_points);
// Do the triangulation.
let triangles = triangulate(&points).unwrap();
println!("{:?}", triangles); // [Triangle(1, 0, 2)]
}
MIT - See LICENSE
file.