Hey everyone,
First off, a huge thank you to cavalier_contours project.
Our team does a lot of high-precision layout work in AutoCAD. For the longest time, we struggled with the native Region API , which were proving to be a major performance bottleneck. As an alternative, we tried the well-known Clipper2 library. While it was incredibly fast, the floating-point precision wasn't sufficient for our needs, leading to unacceptable deviations in our final layouts.
Then we discovered cavalier_contours, and it hit the sweet spot for us — it's both remarkably fast and highly accurate. It solved our core problem beautifully.
To integrate it into our C#/.NET toolchain, we've created a C# wrapper for the library, which we've open-sourced here:
https://github.com/FsDiG/CavalierContoursSharp
So far, everything is working great on our end. We wanted to share this with the community in case anyone else finds themselves in a similar situation. We're happy to open this up for discussion and welcome any collaboration or feedback.
Thanks again for the fantastic library!

Hey everyone,
First off, a huge thank you to cavalier_contours project.
Our team does a lot of high-precision layout work in AutoCAD. For the longest time, we struggled with the native
RegionAPI , which were proving to be a major performance bottleneck. As an alternative, we tried the well-known Clipper2 library. While it was incredibly fast, the floating-point precision wasn't sufficient for our needs, leading to unacceptable deviations in our final layouts.Then we discovered
cavalier_contours, and it hit the sweet spot for us — it's both remarkably fast and highly accurate. It solved our core problem beautifully.To integrate it into our C#/.NET toolchain, we've created a C# wrapper for the library, which we've open-sourced here:
https://github.com/FsDiG/CavalierContoursSharp
So far, everything is working great on our end. We wanted to share this with the community in case anyone else finds themselves in a similar situation. We're happy to open this up for discussion and welcome any collaboration or feedback.
Thanks again for the fantastic library!