Ambiguity error between CategoricalArrays and SentinelArrays

[JackDunnNZ]

(Apologies that this isn't strictly related to DataFrames, but I'm not sure where the fix for this needs to be)

The following code works for me when using a single thread:

using CSV, DataFrames, CategoricalArrays
data = CSV.read("kc_house_data.csv", DataFrame)
select(data, :waterfront => categorical)

But if I run with multiple threads, it uses SentinelArrays when reading in the CSV and I get the following method ambiguity error between CategoricalArrays and SentinelArrays during the transform:

ERROR: MethodError: copyto!(::CategoricalVector{Int64, UInt32, Int64, CategoricalValue{Int64, UInt32}, Union{}}, ::SentinelArrays.ChainedVector{Int64, Vector{Int64}}) is ambiguous. Candidates:
  copyto!(dest::Union{SubArray{var"#s5", N, var"#s4", I, L} where {var"#s5", var"#s4"<:(CategoricalArray{T, var"#s3", R, V, C, U} where {var"#s3", V, C, U}), I, L}, CategoricalArray{T, N, R, V, C, U} where {V, C, U}} where {T, N, R<:Integer}, src::AbstractArray) in CategoricalArrays at /Users/jack/.julia/packages/CategoricalArrays/rDwMt/src/array.jl:615
  copyto!(dest::AbstractVector{T} where T, src::SentinelArrays.ChainedVector) in SentinelArrays at /Users/jack/.julia/packages/SentinelArrays/VbnBp/src/chainedvector.jl:335
Possible fix, define
  copyto!(::Union{SubArray{T, 1, var"#s4", I, L} where {T, T1, R<:Integer, var"#s4"<:(CategoricalArray{T1, var"#s3", R, V, C, U} where {var"#s3", V, C, U}), I, L}, CategoricalVector{T, R, V, C, U} where {C, U, T, R<:Integer, V}}, ::SentinelArrays.ChainedVector)
Stacktrace:
  [1] _convert(::Type{CategoricalVector{Int64, UInt32, V, C, U} where {V, C, U}}, A::SentinelArrays.ChainedVector{Int64, Vector{Int64}}; levels::Nothing)
    @ CategoricalArrays ~/.julia/packages/CategoricalArrays/rDwMt/src/array.jl:343
  [2] (CategoricalVector{Int64, UInt32, V, C, U} where {V, C, U})(A::SentinelArrays.ChainedVector{Int64, Vector{Int64}}; levels::Nothing, ordered::Bool)
    @ CategoricalArrays ~/.julia/packages/CategoricalArrays/rDwMt/src/array.jl:253
  [3] #categorical#72
    @ ~/.julia/packages/CategoricalArrays/rDwMt/src/array.jl:927 [inlined]
  [4] categorical(A::SentinelArrays.ChainedVector{Int64, Vector{Int64}})
    @ CategoricalArrays ~/.julia/packages/CategoricalArrays/rDwMt/src/array.jl:926
  [5] _transformation_helper(df::DataFrame, col_idx::Int64, ::Base.RefValue{Any})
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:368
  [6] select_transform!(::Base.RefValue{Any}, df::DataFrame, newdf::DataFrame, transformed_cols::Set{Symbol}, copycols::Bool, allow_resizing_newdf::Base.RefValue{Bool})
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:546
  [7] _manipulate(df::DataFrame, normalized_cs::Vector{Any}, copycols::Bool, keeprows::Bool)
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:1383
  [8] manipulate(df::DataFrame, cs::Any; copycols::Bool, keeprows::Bool, renamecols::Bool)
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:1311
  [9] select(df::DataFrame, args::Any; copycols::Bool, renamecols::Bool)
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:940
 [10] select(df::DataFrame, args::Any)
    @ DataFrames ~/.julia/packages/DataFrames/vuMM8/src/abstractdataframe/selection.jl:940
 [11] top-level scope
    @ REPL[3]:1

If helpful, the data is available here: kc_house_data.csv

All packages are up to date (CSV 0.9.3, DataFrames 1.2.2, CategoricalArrays 0.10.0, Sentinel Arrays 1.3.7)

[bkamins]

Thank you for an excellent issue report. This is a known problem. It is tracked here JuliaData/CategoricalArrays.jl#361 (and it is good that you reported it so that we will get back to fixing it)

[JackDunnNZ]

Thank you @bkamins, glad to hear you are all on top of it!