[htgb]

Do you have an image of the reversing portion? Any examples of pitfalls?

On the face of it, it seems like you'd define paths and sweep profiles for the material to remove. Is the difficulty in defining the path of the reversing portion, where it's not a helix?

[contingencies]

Threads have a range of properties, all of which must be supported, then it's generally the capacity to mirror a thread along the same face (exactly), and at either end add a transition zone that reverses, preferably with configurable characteristics (straight portion, rounding, etc.). Many approaches fail because the reversing transition zone is not readily aligned with the existing thread except through hacks or because reversing is incompatible with the programmatic approach used. Once that's all done, there are additional complexities for a captured element to ride in it which typically has a geometric relationship with the thread pitch, crossovers and transition zone. It's one of those "harder than it looks" problems, even for machinists. That's why, if the codebase can achieve it, I'd give it respect.

PS. I also use a lot of OpenSCAD. Liked your comment. I started in POVRay so it made a lot of sense after managing non-FOSS CAD people for years.