This issue is used to trigger TagBot; feel free to unsubscribe.

If you haven't already, you should update your TagBot.yml to include issue comment triggers.
Please see this post on Discourse for instructions and more details.

If you'd like for me to do this for you, comment TagBot fix on this issue.
I'll open a PR within a few hours, please be patient!

Currently some tests are marked as @test_broken, we should investigate which cases this occurs and fix the underlying issues. Nothing in our functionality is random and using any convex norm should not break any of the algorithms we currently implement.

Any chance this effort could be directed to reviving VoronoiDelaunay.jl? I think VoronoiDelaunay.jl had state-of-the-art performance across all languages, and it would be nice to get some people there maintaining it instead of starting a new package from scratch.

Also, as far as I remember, computing the Delaunay tessellation to extract the Voronoi diagrams is more efficient than trying to compute the Voronoi diagrams directly from the set of points.

Opening this issue to have a place to discuss what we need to to before initially putting out the code we already have. It's cool that our algorithms are already faster than what's out there so we should let others use it as soon as we can.

The minimum that needs to happen is:

• Decision on what the public API looks like and what functionality we want to offer,
• Documentation (with Documenter.jl? I'm not sure I've never generated documentation before),
• Reorganising the package into the stable branch and then our own experimental branches.

After this we can register the package.

In terms of functionality, what do we want to make the user do? Should we have out-of-place versions of our in-place jfa! and jdac! functions?

I really should have done this from the beginning as it makes everything so much simpler conceptually. We should make sure that they're not causing any allocations but I doubt that will be the case.