Lcbx

just found a bug using this. some arguments expect pointers but can be given non ffi pointers ; however since signatures expects a pointer it calls `ffi.addressof` on the argument...

not the best fix but it works

found a way to check on type rather than str(type) should be robust and work with dynamic since the C_POINTER type we check is determined in pyRay init

The implementation change is around 15 loc, only in [pyray/__init__.py](https://github.com/electronstudio/raylib-python-cffi/pull/53/files#diff-377dcfbce9c77e1cfa24dd4f9889403c63211acae5d7cec83fa13913248ee229) other files are changes to examples

oh yeah and it's pretty much finished afaik, tested it on a few examples (your pyrayTest, bunnyMark, boxCollisions and animation)

> For me the not-so-ideal work around is to export all variables that may have documentation, then use `_validate_property` to change the flags and other pieces of the properties. Interesting...

Hello @Daylily-Zeleen, for thread-safety, you might want to call the functions ```sync()``` and ```end_sync()``` of the PhysicsServer classes an example of usage is in the function ```Main::iteration()``` in file ```main/main.cpp```...

For the first error I think it should be ```get_node()```, the compiler says it can't find the template type (called 'T' in godot source code) The second is because c++...

closing, the get_node not having the type argument is a bit too complex to put in the transpiler, and the 'new' keyword would have to be handled by hand in...