Awesome

Clipper2-java

A Java port of Clipper2.

Usage

Overview

The interface of Clipper2-java is identical to the original C# version.

The Clipper class provides static methods for clipping, path-offsetting, minkowski-sums and path simplification. For more complex clipping operations (e.g. when clipping open paths or when outputs are expected to include polygons nested within holes of others), use the Clipper64 or ClipperD classes directly.

Maven

Clipper2-java is available as Maven/Gradle artifact via Jitpack.

Example

Paths64 subj = new Paths64();
Paths64 clip = new Paths64();
subj.add(Clipper.MakePath(new int[] { 100, 50, 10, 79, 65, 2, 65, 98, 10, 21 }));
clip.add(Clipper.MakePath(new int[] { 98, 63, 4, 68, 77, 8, 52, 100, 19, 12 }));
Paths64 solution = Clipper.Union(subj, clip, FillRule.NonZero);
solution.get(0).forEach(p -> System.out.println(p.toString()));

Port Info

• tangiblesoftwaresolutions' C# to Java Converter did the heavy lifting (but then a lot of manual work was required).
• Wrapper objects are used to replicate C# ref (pass-by-reference) behaviour. This isn't very Java-esque but avoids an unmanageable refactoring effort.
• Code passes all tests: polygon, line and polytree.
• Uses lower-case (x, y) for point coordinates.
• Private local variables have been renamed to their camelCase variant but public methods (i.e. those of Clipper.class) retain their C# PascalCase names (for now...).
• Benchmarks can be run by including jmh:benchmark to the chosen maven goal.
• scanlineList from ClipperBase uses Java TreeSet (variable renamed to scanlineSet).

Benchmark

lightbringer's Java port of Clipper1 is benchmarked against this project in the benchmarks. Clipper2-java is faster, which becomes more pronounced input size grows.

Benchmark               (edgeCount)  Mode  Cnt  Score   Error  Units
Clipper1.Intersection         1000  avgt    2  0.209           s/op
Clipper1.Intersection         2000  avgt    2  1.123           s/op
Clipper1.Intersection         4000  avgt    2  9.691           s/op
Clipper2.Intersection         1000  avgt    2  0.130           s/op
Clipper2.Intersection         2000  avgt    2  0.852           s/op
Clipper2.Intersection         4000  avgt    2  3.465           s/op