cdt

Fast, robust constrained Delaunay triangulation

1 unstable release

 0.1.0 Jun 25, 2021

#2022 in Algorithms

Used in 4 crates (2 directly)

MIT/Apache

105KB
1.5K SLoC

`cdt`

`cdt` is a library for calculating Delaunay and constrained Delaunay triangulations.

It is optimized for correctness and speed, using exact predicates to perform point-in-circle and orientation tests.

`lib.rs`:

`cdt` is a library for calculating Delaunay and constrained Delaunay triangulations.

It is optimized for correctness and speed, using exact predicates to perform point-in-circle and orientation tests.

Examples

Delaunay triangulation

This triangulates a set of four points in a square

``````let pts = vec![(0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0)];
let triangles = cdt::triangulate_points(&pts).unwrap();
assert!(triangles.len() == 2);
for t in triangles {
println!("{:?} {:?} {:?}", pts[t.0], pts[t.1], pts[t.2])
}
``````

Constrained Delaunay triangulation

This triangulates an inner and outer square

``````let pts = vec![(0.0, 0.0), (1.0, 0.0), (1.0, 1.0), (0.0, 1.0),
(0.2, 0.2), (0.8, 0.2), (0.8, 0.8), (0.2, 0.8)];
let triangles = cdt::triangulate_contours(&pts,
&[vec![0, 1, 2, 3, 0], vec![4, 5, 6, 7, 4]])
.unwrap();
for t in triangles {
println!("{:?} {:?} {:?}", pts[t.0], pts[t.1], pts[t.2])
}
``````

Crate features

By default, the library uses `u32` indexes for internal data structures, to improve performance. If you are planning to triangulate more than 500M points in a single pass, you should enable the `long-indexes` feature.

~1.1–1.6MB
~38K SLoC