Michael Dawson-Haggerty

Hey, yeah I would also like to drop or reduce the scope of `trimesh.interfaces` to OpenSCAD/blender if at all possible. Until recently the state of boolean operation backends available has...

Awesome thanks for the PR and test! The logic seems pretty similar to [`trimesh.intersections.slice_faces_plane`](https://trimsh.org/trimesh.intersections.html#trimesh.intersections.slice_faces_plane), I was wondering if it would be possible to use the outputs of that to simplify...

Awesome yeah if you wouldn't mind factoring it use `slice_plane` (maybe pulling some of the internals into a simpler function that returns only numpy arrays or something?). From a maintenance...

Yeah this is true, currently the only way to get a subdivision to maintain nice properties is vanilla `subdivide` for every face (i.e. without passing a subset of faces). I...

Hey, it's probably a memory leak in one of the c-libraries (rtree or embreex maybe, debugging welcome!). If you don't have embree that will also be an enormous ~50x speedup...

If you run it with pyinstrument can you tell which line is bogging down?

This would be awesome for utilities interacting with complicated proto messages! For what it's worth I hacked this for a demo using `pydantic == 1.10.7` by patching a no-op encoder,...

Yeah just looking quickly there's a number of operations that look like they'd produce offsets, like [here](https://github.com/mikedh/trimesh/blob/master/trimesh/smoothing.py#L76-L81) where it's just multiplying by a scaling factor. I think that logic should...

Hey, is this still happening on the latest versions? This should be unit tested and I just quickly ran ``` e = trimesh.load('models/fuze.obj') e.export('~/Downloads/textest.glb') ``` Which loaded fine on https://gltf-viewer.donmccurdy.com/...

Oh interesting, thanks for the report! The second log makes me deeply suspicious of the GEOS/Shapely build on different architectures and I might punt on that one as "upstream." The...