3b

Thanks for the patch, didn't realize anyone was still using this :) Might take me a while to look at it, but will try to get it merged eventually.

If we try to fix this in cl-glut, need to be careful about the case where the user doesn't actually want the window shown to start with (assuming that is...

added gl:+timeout-ignored+ and a few similar constants for those (leaving it in %gl:enum for now since that is autogenerated though)

skipping some of these for now, since they might conflict with the `glarray` buffer management stuff in cl-opengl, and that needs more thought: `glNamedBufferData`, `glBindVertexBuffer`, `glVertexArrayVertexBuffer` (and plural variants) filed...

Not so much "deliberate" as "not finished". For example final wrappers might want to parse or wrap results in some way, in which case there would be a "nicer" function...

added an option to mask traps at coarser granularity without disabling it completely, so for example you can wrap your main loop or render function with `%GL:WITH-FLOAT-TRAPS-MASKED` and get most...

yeah, probably would be reasonable to use `ecase` instead for those functions dispatching on `length` like that.

I think the reason I added `with-pushed-matrix*` instead of adding `unwind-protect` to `with-pushed-matrix` was due to the possibility of the current matrix changing before the non-local exit, in which case...

Right, push without pop is bad, but we can't (efficiently) be sure we are popping the right matrix stack, which is why `with-pushed-matrix*` requires explicitly specifying which matrix stack to...

Actually looks like it uses a special instead of a closer now, so it can work with multiple contexts with changes to user code. Probably needs better documentation though, and...