Andrew Port

Looks like we have a solution to this now thanks the new Document class pushed recently by @mxgrey ! I haven't done much testing, but I'm impressed -- it looks...

Hi @Esser50K, thanks for the suggestion and PR. If there's a significant speedup, yeah, that sounds great. I think there's pretty good test coverage for intersections, so we could even...

Do you have any examples to demonstrate the speedup @Esser50K ? Running `python -m unittest test.test_path.Test_intersect` shows no significant improvement. I'm tempted to drop 2.7 support just to have type...

Thanks to some great work by @tatarize , you can now run `python -m unittest test.test_path.Test_intersect.test_random_intersections` to test how fast intersections are found. By adjusting the `count` variable you can...

Regardless, thanks for your work on it. You are appreciated!

Hi @piscvau, this `tests_path_intersect_with_horizontal.py` file looks quite nice. Could you give me a simple example or two I can paste into my terminal to play around with?

I'm usually on an apple SoC so I know this isn't all non-x86 architectures. There's a known issue regarding the __hash__ method being OS (and maybe also architecture) dependent. I'm...

I wish (though keep reading). If I could go back in time, `svgpathtools` would use numpy much more ubiquitously and take advantage of vectorization in many more places. I just...

@tatarize thanks for bringing this up -- that was pretty sloppy of me. Switching `math.sqrt` to `numpy.sqrt` in path.py caused the problem (which was related to `numpy.sqrt` returning an `nan`...

Yeah, I'm surprised no one has commented on this before (as far as I recall). If you make a PR and include a unit test to check order is preserved,...