Circonflexe

`AngleAxis` modified. Also, while I was looking in there, I was a bit surprised to see the details of this struct: wouldn't it have been a bit more natural to...

Exactitude of coefficients is lost for `AngleAxis`, since the matrix is obtained through quaternions; this involves both angle-halving and L^2 normalization. In the general case, since the axis itself is...

> I think we are not using quaternion here, just [Rodrigues' rotation formula](https://mathworld.wolfram.com/RodriguesRotationFormula.html), ([code in this repo](https://github.com/JuliaGeometry/Rotations.jl/blob/v1.0.4/src/angleaxis_types.jl#L52)). But just three lines above your quote: ```julia @inline Base.getindex(aa::AngleAxis, i::Int) = UnitQuaternion(aa)[i]...

> > i.e. `AngleAxis(60°, [1,1,1]//1)` could return a rational matrix > > Understood, but I think it will be hard for type stability. For example, `AngleAxis(59°, [1,1,1]//1)` has same argument...

> > I didn't know that the Unitful works exactly. > > In this case `Unitful` is not doing anything very fancy, it's just causing `sin(30°)` to dispatch to `sind(30)`...

So, after almost 6 years this still has not been solved (and yes, I just did write yet another `.ply` file reader). A way to ease the dispatch would be...

[Here is the PR](https://github.com/JuliaGeometry/Rotations.jl/pull/197).

In addition to this: may I suggest that the doc include a couple of paragraphs about “how to define your own tree type (indexed vs. struct-like)”? I know that there...

This is exactly what I suspected: [`RegistryCI.jl` has a hard-coded exception for Yggdrasil](https://github.com/JuliaRegistries/RegistryCI.jl/issues/437#issuecomment-950232221). Therefore, `BinaryBuilder` should also behave differently on this aspect when it is run outside of Yggdrasil (such...

I could not reproduce this — moreover, in the latest version, there is no `stl()` function. (It has been replaced by `save()`). Are you sure you have an up-to-date version?