Skip to content

Conversation

@bkamins
Copy link
Member

@bkamins bkamins commented Apr 8, 2024

Fixes #354.

@bkamins bkamins requested review from nalimilan and quinnj April 8, 2024 19:29
@bkamins
Copy link
Member Author

bkamins commented Apr 8, 2024

Current behavior under Julia nightly:

julia> Tables.columntable(Vector{Any}[])
(ref = MemoryRef{Any}[], size = Tuple{Int64}[])

julia> T = typeof(view(Vector{Any}[], :))
SubArray{Vector{Any}, 1, Vector{Vector{Any}}, Tuple{Base.Slice{Base.OneTo{Int64}}}, true}

julia> Tables.columntable(T[])
(parent = Vector{Vector{Any}}[], indices = Tuple{Base.Slice{Base.OneTo{Int64}}}[], offset1 = Int64[], stride1 = Int64[])

@nalimilan
Copy link
Member

Maybe we should apply this to any AbstractVector? There's a small risk of breaking something, but it would sound more consistent.

@bkamins
Copy link
Member Author

bkamins commented Apr 10, 2024

We patched this in DataFrames.jl in JuliaData/DataFrames.jl#3435, so making adjustments here is not critical for DataFrames.jl.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

materializer behavior differs from its docs

3 participants