JuliaArrays / JuliaArrays/StaticArrays.jl

`reinterpret` failure due to missing `axes` specialization

Open
#1,269 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

Dominant language
Julia
Stars
844
Forks
159
Avg merge
3d 21h
Merged PRs (30d)
3

Description

julia> reinterpret(Float64, SA[im,1.0im])
Error showing value of type Base.ReinterpretArray{Float64, 1, ComplexF64, SVector{2, ComplexF64}, false}:
ERROR: DimensionMismatch("1:4 is inconsistent with SOneTo{2}")
Stacktrace:
  [1] (::StaticArrays.var"#errmsg#1"{2})(r::UnitRange{Int64})
    @ StaticArrays ~/.julia/packages/StaticArrays/80e5O/src/SOneTo.jl:14
  [2] SOneTo
    @ ~/.julia/packages/StaticArrays/80e5O/src/SOneTo.jl:15 [inlined]
  [3] convert
    @ ./range.jl:148 [inlined]
  [4] oftype
    @ ./essentials.jl:375 [inlined]
  [5] axes
    @ ./reinterpretarray.jl:302 [inlined]
  [6] summary
    @ ./show.jl:2664 [inlined]
  [7] show(io::IOContext{Base.TTY}, #unused#::MIME{Symbol("text/plain")}, X::Base.ReinterpretArray{Float64, 1, ComplexF64, SVector{2, ComplexF64}, false})
    @ Base ./arrayshow.jl:337
...

As @MasonProtter points out in slack, this is the offending method in Base

function axes(a::NonReshapedReinterpretArray{T,N,S} where {N}) where {T,S}
    paxs = axes(a.parent)
    f, l = first(paxs[1]), length(paxs[1])
    size1 = issingletontype(T) ? l : div(l*sizeof(S), sizeof(T))
    tuple(oftype(paxs[1], f:f+size1-1), tail(paxs)...)
end

The error stems from oftype not receiving enough type-domain information to yield the correct SOneTo axis object. Mason also suggests the following kind of fix in StaticArrays.jl

function Base.axes(::Base.NonReshapedReinterpretArray{T, N, S, SA}) where {T, N, S, SA <: StaticArray{Tup}} where {Tup}
    paxs = axes(SA)
    old_sizes = size(SA)
    l = old_sizes[1]
    size1 = Base.issingletontype(T) ? l : div(l*sizeof(S), sizeof(T))
    (SOneTo(size1), SOneTo.(Base.tail(old_sizes))...)
end

and an analogous method for Base.axes(::ReshapedReinterpretArray{T,N,S} where {N}) where {T,S}

Contributor guide

No contributing guide indexed for this repository

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start with the Julia reproducer and the Base axes methods shown in the issue, then inspect StaticArrays' SOneTo implementation and existing axes specializations. The analogous methods for NonReshapedReinterpretArray and ReshapedReinterpretArray should make reinterpret(Float64, SA[im, 1.0im]) display without a DimensionMismatch and report compatible axes.

Written by the indexing model from the issue text.

Assessment

Tech stack
julia
Domain
data
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
45/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.