Michael Kirk

I think you'll need to run cargo fmt, otherwise LGTM I'll leave it open for a couple days before merging in case anyone else wants to weigh in.

Darn it. I'm sorry for not realizing this earlier. This seems related to https://github.com/georust/geo/issues/1181

Assuming #1216 gets merged, I've got a different approach that should supersede this, generically implementing Densify for all line measures (Haversine, Geodesic, Euclidean, etc.) You can see a WIP here:...

My alternative to this was merged in #1231. With that you can call: ```rust use geo::{Geodesic, Densify, Euclidean}; line_string.densify::(); line_string.densify::(); ``` Let me know if that works for your purposes...

With #1298 (not yet released) this changed to: ```rust use geo::{Geodesic, Densify, Euclidean}; Geodesic.densify(&line_string); Euclidean.densify(&line_string); ```

This is exciting! I might not have time to review it this week, but I'll take a look as soon as I can.

> This impl doesn't use SweepPoint, Cross, or PointOrLine: we only accept Line segments for checking. If we adopt this, we would either have to do more work to accept...

We don't necessarily have to deprecate the old sweep algorithm at this point. I did that in my commit mostly just to identify where all it was being used while...

> Wondering whether it's sensible to document the presence of this module in the line_intersection module docs: sweep isn't very intuitive, and it's already slightly confusing to have line_intersection and...

I'm not sure if you're still actively pushing code, so tag me when you're ready for re-review @urschrei