Skip to content

improved support for struct columns with missing values #498

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
baumgold wants to merge 1 commit into
base: main
Choose a base branch
from
Open

improved support for struct columns with missing values #498

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 1 addition & 1 deletion Project.toml
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -17,7 +17,7 @@
name = "Arrow"
uuid = "69666777-d1a9-59fb-9406-91d4454c9d45"
authors = ["quinnj <quinn.jacobd@gmail.com>"]
version = "2.7.1"
version = "2.7.2"

[deps]
ArrowTypes = "31f734f8-188a-4ce0-8406-c8a06bd891cd"
Expand Down
14 changes: 9 additions & 5 deletions src/arraytypes/struct.jl
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -77,16 +77,19 @@ struct ToStruct{T,i,A} <: AbstractVector{T}
data::A # eltype is NamedTuple or some struct
end

ToStruct(x::A, j::Integer) where {A} =
ToStruct{fieldtype(Base.nonmissingtype(eltype(A)), j),j,A}(x)
function ToStruct(x::A, j::Integer, hasmissing::Bool=false) where {A}
AT = fieldtype(eltype(A), j)
Copy link
Member

@ericphanson ericphanson Feb 8, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

do we still want Base.nonmissingtype(eltype(A)) inside the field type?

Copy link
Member Author

@baumgold baumgold Feb 9, 2024

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I don't think so. My change makes it so missing types are properly supported.

T = hasmissing ? Union{Missing,AT} : AT
ToStruct{T,j,A}(x)
end

Base.IndexStyle(::Type{<:ToStruct}) = Base.IndexLinear()
Base.size(x::ToStruct) = (length(x.data),)

Base.@propagate_inbounds function Base.getindex(A::ToStruct{T,j}, i::Integer) where {T,j}
@boundscheck checkbounds(A, i)
@inbounds x = A.data[i]
return x === missing ? ArrowTypes.default(T) : getfield(x, j)
ismissing(x) ? missing : getfield(x, j)
end

arrowvector(::StructKind, x::Struct, i, nl, fi, de, ded, meta; kw...) = x
Expand All @@ -102,9 +105,10 @@ function arrowvector(::StructKind, x, i, nl, fi, de, ded, meta; kw...)
len = length(x)
validity = ValidityBitmap(x)
T = Base.nonmissingtype(eltype(x))
hasmissing = Missing <: eltype(x)
data = Tuple(
arrowvector(ToStruct(x, j), i, nl + 1, j, de, ded, nothing; kw...) for
j = 1:fieldcount(T)
arrowvector(ToStruct(x, j, hasmissing), i, nl + 1, j, de, ded, nothing; kw...)
for j = 1:fieldcount(T)
)
NT = namedtupletype(T, data)
return Struct{withmissing(eltype(x), NT),typeof(data),fieldnames(NT)}(
Expand Down