search

kvark / mint

Math Interoperability Types
MIT License
253 stars 20 forks source link

Generic basis for Euler angles and left-handed quaternions #8

Closed kvark closed 7 years ago

kvark commented 7 years ago

Fixes #7 Fixes #9

r? @Ralith

kvark commented 7 years ago

also r? @nox @nical if interested for a bit

nox commented 7 years ago

Why can't the marker be a private field?

nox commented 7 years ago

I agree with @Ralith that units look out of scope in that crate.

kvark commented 7 years ago

@nox Thanks for taking a look!

Why can't the marker be a private field?

I originally thought that hiding markers is a bad idea because there will be a need for new() constructors, and methods/functions are out of scope for the library. But then, I think it's fine, given that we can do Type::from([...]). The markers are hidden now.

I agree with @Ralith that units look out of scope in that crate.

I don't think that's what @Ralith says. I read their words as - providing Euler angles is unfortunate, but if that's what's needed, then the way this PR choose to do it is right.

It's not a discussion/PR about units.

nox commented 7 years ago

methods/functions are out of scope for the library.

Then IMO units are out of scope too.

It's not a discussion/PR about units.

This discussion needs to happen nonetheless, given you add units in it.

nical commented 7 years ago

This discussion needs to happen nonetheless, given you add units in it.

I think that the PhantomData is about the left/right-handedness of the quaternions, and not units.

nox commented 7 years ago

Oh! Makes sense. Disregard me.

kvark commented 7 years ago

Thanks everyone!