Vendorize StaticArrays.FixedSizeArrays

[aaronang]

This patch allows the JuliaArrays/StaticArrays.jl repository to deprecate StaticArrays.FixedSizeArrays without affecting GeometricTypes.jl.

Resolves #132.

I am very new to Julia and I am not sure if this patch makes any sense. I am looking forward to receiving constructive feedback 🙂

[rdeits]

Thanks! This seems perfectly reasonable to me.

For clarity and proper attribution, could you please add a comment at the beginning of FixedSizeArrays.jl saying something like:

This code was copied from https://github.com/JuliaArrays/StaticArrays.jl which is written by Andy Ferris and released under the MIT "Expat" License: 

> Copyright (c) 2016: Andy Ferris.
>
> Permission is hereby granted, free of charge, to any person obtaining
> a copy of this software and associated documentation files (the
> "Software"), to deal in the Software without restriction, including
> without limitation the rights to use, copy, modify, merge, publish,
> distribute, sublicense, and/or sell copies of the Software, and to
> permit persons to whom the Software is furnished to do so, subject to
> the following conditions:
>
> The above copyright notice and this permission notice shall be
> included in all copies or substantial portions of the Software.
>
> THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
> EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
> MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
> IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY
> CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION OF CONTRACT,
> TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION WITH THE
> SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.

(which is just a copy of the license file from https://github.com/JuliaArrays/StaticArrays.jl/blob/master/LICENSE.md )

[aaronang]

Done. Is there by the way anything we can do about the following warnings?

WARNING: Method definition isnan(StaticArrays.StaticArray{S, T, N} where N where T where S<:Tuple) in module FixedSizeArrays at /home/travis/.julia/packages/StaticArrays/Ze5H3/src/FixedSizeArrays.jl:60 overwritten in module FixedSizeArrays at /home/travis/build/JuliaGeometry/GeometryTypes.jl/src/FixedSizeArrays.jl:83.
WARNING: Method definition extrema(AbstractArray{T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N), 1}) where {T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N)} in module FixedSizeArrays at /home/travis/.julia/packages/StaticArrays/Ze5H3/src/FixedSizeArrays.jl:77 overwritten in module FixedSizeArrays at /home/travis/build/JuliaGeometry/GeometryTypes.jl/src/FixedSizeArrays.jl:100.
WARNING: Method definition minimum(AbstractArray{T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N), 1}) where {T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N)} in module FixedSizeArrays at /home/travis/.julia/packages/StaticArrays/Ze5H3/src/FixedSizeArrays.jl:81 overwritten in module FixedSizeArrays at /home/travis/build/JuliaGeometry/GeometryTypes.jl/src/FixedSizeArrays.jl:104.
WARNING: Method definition maximum(AbstractArray{T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N), 1}) where {T<:(StaticArrays.StaticArray{Tuple{N}, T, 1} where T where N)} in module FixedSizeArrays at /home/travis/.julia/packages/StaticArrays/Ze5H3/src/FixedSizeArrays.jl:84 overwritten in module FixedSizeArrays at /home/travis/build/JuliaGeometry/GeometryTypes.jl/src/FixedSizeArrays.jl:107.

[rdeits]

We shouldn't include those definitions in GeometryTypes as it's technically type piracy. They have nothing to do with FixedSizeArrays, so they should probably just be defined in StaticArrays.jl or not defined at all.

@andyferris will the methods here: https://github.com/JuliaArrays/StaticArrays.jl/blob/86097997bd846f0b9f4331cd55ba8a233375173b/src/FixedSizeArrays.jl#L59-L85 be preserved when you deprecate and remove FixedSizeArrays?

[aaronang]

@rdeits I removed the definitions that from FixedSizeArrays. I assume we will have to wait for @andyferris's reply before can merge it?

[andyferris]

Thanks for taking this step.

[andyferris]

You have my permission to drop the license and include the code in GeometryTypes. I'm assuming @SimonDanisch will similarly agree.

[SimonDanisch]

Sure :)

[andyferris]

We shouldn't include those definitions in GeometryTypes as it's technically type piracy. They have nothing to do with FixedSizeArrays, so they should probably just be defined in StaticArrays.jl or not defined at all.

There were always a couple small definitions in FixedSizedArrays which went beyond the typical AbstractArray interface (one explicit design goal for StaticArrays was to follow the AbstractArray interface a strictly as feasibly possible so that they can be used interchangeably). Personally, I'd recommend following what's in Base and StaticArrays and for whatever is missing you either add here as a geometric concept, or make a PR to Base or StaticArrays for generic concepts. For example, I still feel slightly guilty about the length-2 "cross" product... :)

[aaronang]

I removed the attribution based on @andyferris's feedback. Please let me know if I did it incorrectly 🙂

[c42f]

@rdeits I had a look at the functions in https://github.com/JuliaArrays/StaticArrays.jl/blob/86097997bd846f0b9f4331cd55ba8a233375173b/src/FixedSizeArrays.jl#L59-L85. IMO most of these should not be preserved in StaticArrays:

• There is no canonical notion of ordering in a multidimensional space, so we should not provide maximum, minimum or extrema. extrema is really calculating the axis aligned bounding box, and should be renamed appropriately and implemented in GeometryTypes if needed.
• Base.isnan does not apply to arrays (an array can contain NaNs but cannot be a NaN), and we should follow the lead there. any(isnan, v) appears to generate efficient code for that.

We might think about retaining unit in StaticArrays proper, though I'd want to rename it to something less generic, perhaps unitvec.

[SimonDanisch]

Thank you, this seems to work fine :)