Tinko Bartels

I have made the following changes to the PR: - The template-based error bound calculation has been replaced by a constexpr based implementation. This improves compile-time, template instantiation depth and...

Thanks for all the comments, I already addressed a couple of them in comments and some with a commit. I think the main question, though, is the matter of strategies...

@awulkiew thank you for your elaborate comment. I'll address the easy questions first: - Is `uniform_convex_polygon_sampler` cartesian-only or CS-agnostic? I think `uniform_convex_polygon_sampler` in itself is CS-agnostic, however, it requires a...

I am sorry for the long delay. I will resume work on this PR and I will attempt to implement the policy solution in the near future and commit it...

Hello @awulkiew , I have reconsidered your solutions 2.1 und 2.2 and I genereally lean towards solution 2.1. I have tried to consider the sensible design space for uniform random...

I have committed and pushed changes that move non-specific code into policies. Please consider this commit WIP. I opted to push this early, so that it can discussed whether this...

@vissarion Thanks for the test results with long double and multi-precision. I tested it with double before mentioning it in the PR and saw the line in the bottom right...

Hi, I cannot tell you why you get the inconsistency without further analysis (I suspect numerical issues, more on that further below), but I have a strong suspicion why you...

I am not a maintainer, so I don't know if union_ deletes collinear vertices and if there is a way to stop it. But point 7 is not strictly collinear...

Thanks. Unfortunately, my knowledge of the implementation of union_ is insufficient to tell how to force it to keep collinear vertices. I hope somebody else can help you with that....