Comparison of negative zero float

[Zubnix]

Currently the equals operation of MatN compares the raw bits of the float values wrapped by the MatN objects. However in case of -0.0f and 0.0f these are considered not equal, while in fact they should be (from a practical point of view).

Would it be beneficial to do a temp. negative to positive zero conversion while comparing MatN objects?

The reason I'm asking is because I implemented a Matrices.inverse(Mat4 mat) method on my fork, and my unit test fails because of the above description. I can fix it using this method it but I'd rather ask first.

[jroyalty]

I'll take a look. Did you implement that method in PR #9?

[Zubnix]

I adapted it from here:
http://www.sanfoundry.com/java-program-find-inverse-matrix/

Sinds is a very commonly known math algorithm and the code doesn't seem to be licensed in any way, I though it'd be ok.