Andrii Doroshenko

Not until it's actually finished, it's still in development state (in sandbox). And it's still has some bugs in C++ implementation. Triangulation unit might fail occasionally. I wonder whether Angus Johnson is still working on it, though.

If triangles don't overlap for some reason and you want to ensure that you can merge them all to get a single polygon, you could pass them to ClipperOffset with miter parameter, and then grow them by a tiny amount and merge afterwards with Clipper.

Make Clipper C++ 6.4.2 to be compiled with exception disabled

Simplify ClipperCore by using C++11 features

For integers this is already done. Clipping is based on integer coordinates anyway. As for changing default floating point type, Angus said that the performance difference is negligible on modern CPU's, but should be easy to adapt it anyway.

Looks neat, apart from CamelCase vs snake_case naming convention which is a good idea to keep for now and will have to be converted as final touch in order to be more like C++'s STL. I've noticed you've removed ClipperLibException from core. In fact I think it might be a good idea to define a macro so users can choose whether to have exceptions in code, the reason for that is that Clipper can be compiled for various platforms, and not all of them support exceptions (Android, iPhone etc) While compiling...

I'm sorry but it seems like you've uploaded the original version by mistake, I don't see any changes there, also make sure you downloaded ClipperCppR555.zip that I uploaded before which incorporates more changes to clipper_core. clipper.h/clipper.cpp are a bit messy because I've given up trying to fix the bugs, so see for yourself, if you manage to fix it I'll be happy to clean it up if needed...

Hehe yes initally I did exactly that, renaming Rect methods and making them work from inside struct, but there was RotateRect that couldn't be overloaded (?) Anyway that's good idea. Same for making path(s) related functions to become member functions, but some technicalities could arrive from that as templates work the same for all types, and I've noticed some custom logic between scaling methods, for instance. They could be moved inside struct with little modifications though. Also operator=()...