Michael Dawson-Haggerty

Hey, yeah agreed that they probably shouldn't be settable and just be strictly defined as "the cross product of the edges" to avoid confusion from functions that expect them to...

Hey, pyglet can't connect to a display to get an OpenGL context, which can happen for a lot of reasons. Can you open windows from that terminal (i.e. `glxgears`)?

Yeah, unfortunately `save_image` is pretty unreliable in general because of platform differences. I've found the most reliable thing for headless rendering in general is using LLVMpipe software rasterization which works...

Hey, not totally sure why it's not watertight... I was able to get it to be watertight with `triangle` (free for non-commercial use, `pip install triangle`) but not the permissibly...

Hey, I think what may be happening is `edge` doesn't exist in `face_adjacency_edges` because presumably it's on a boundary so it's giving a wrong index: ``` sliced.face_adjacency_edges_tree.query(edge) Out[21]: (6.708203932499369, 126)...

Hey, it's actually printing a `log.debug` message which is a little buried admittedly, if we changed it to a `log.warning` it would be a lot more obvious but spam users...

Glad it's helpful! I think you should just need to turn off the default merge-vertex-on-creation behavior as your scale is smaller than `trimesh.tol.merge` (or set that value to something smaller...

Hey, did you get a traceback (`python -X faulthandler script.py`)? Not to punt to upstream, but this is likely to be deep inside manifold3d and they would definitely appreciate isolated...

Hey @anandolee the CLA check is now passing thanks!

Awesome thanks for the PR! A few quick notes: - Can we change the parameter from `sidewall_angle: float` -> `taper: float | None`? - Angle should be in radians rather...