Apple M1 support

src/Arrow.jl

@@ -53,6 +53,13 @@ import Base: ==
 
 const DEBUG_LEVEL = Ref(0)
 
+const DEFAULT_BYTE_ALIGNMENT = let
+    require_16 = Base.BinaryPlatforms.Platform("aarch64", "macos";
+                                               libgfortran_version = "5.0.0",
+                                               cxxstring_abi = "cxx11")
+    Base.BinaryPlatforms.HostPlatform() == require_16 ? 16 : 8
+end
+
 function setdebug!(level::Int)
     DEBUG_LEVEL[] = level
     return

src/append.jl

@@ -43,7 +43,7 @@ record batches simultaneously (e.g. if julia is started with `julia -t 8`
 or the `JULIA_NUM_THREADS` environment variable is set).
 
 Supported keyword arguments to `Arrow.append` include:
-  * `alignment::Int=8`: specify the number of bytes to align buffers to when written in messages; strongly recommended to only use alignment values of 8 or 64 for modern memory cache line optimization
+  * `alignment::Int=$DEFAULT_BYTE_ALIGNMENT`: specify the number of bytes to align buffers to when written in messages; strongly recommended to only use alignment values of 8 or 64 for modern memory cache line optimization or 16 on Apple silicon.
   * `colmetadata=nothing`: the metadata that should be written as the table's columns' `custom_metadata` fields; must either be `nothing` or an `AbstractDict` of `column_name::Symbol => column_metadata` where `column_metadata` is an iterable of `<:AbstractString` pairs.
   * `dictencode::Bool=false`: whether all columns should use dictionary encoding when being written; to dict encode specific columns, wrap the column/array in `Arrow.DictEncode(col)`
   * `dictencodenested::Bool=false`: whether nested data type columns should also dict encode nested arrays/buffers; other language implementations [may not support this](https://arrow.apache.org/docs/status.html)
@@ -74,7 +74,7 @@ function append(io::IO, tbl;
         denseunions::Bool=true,
         dictencode::Bool=false,
         dictencodenested::Bool=false,
-        alignment::Int=8,
+        alignment::Int=DEFAULT_BYTE_ALIGNMENT,
         maxdepth::Int=DEFAULT_MAX_DEPTH,
         ntasks=Inf,
         convert::Bool=true,

src/write.jl

@@ -39,7 +39,7 @@ record batches simultaneously (e.g. if julia is started with `julia -t 8` or the
 Supported keyword arguments to `Arrow.write` include:
   * `colmetadata=nothing`: the metadata that should be written as the table's columns' `custom_metadata` fields; must either be `nothing` or an `AbstractDict` of `column_name::Symbol => column_metadata` where `column_metadata` is an iterable of `<:AbstractString` pairs.
   * `compress`: possible values include `:lz4`, `:zstd`, or your own initialized `LZ4FrameCompressor` or `ZstdCompressor` objects; will cause all buffers in each record batch to use the respective compression encoding
-  * `alignment::Int=8`: specify the number of bytes to align buffers to when written in messages; strongly recommended to only use alignment values of 8 or 64 for modern memory cache line optimization
+  * `alignment::Int=$DEFAULT_BYTE_ALIGNMENT`: specify the number of bytes to align buffers to when written in messages; strongly recommended to only use alignment values of 8 or 64 for modern memory cache line optimization
   * `dictencode::Bool=false`: whether all columns should use dictionary encoding when being written; to dict encode specific columns, wrap the column/array in `Arrow.DictEncode(col)`
   * `dictencodenested::Bool=false`: whether nested data type columns should also dict encode nested arrays/buffers; other language implementations [may not support this](https://arrow.apache.org/docs/status.html)
   * `denseunions::Bool=true`: whether Julia `Vector{<:Union}` arrays should be written using the dense union layout; passing `false` will result in the sparse union layout
@@ -167,7 +167,7 @@ function Base.open(::Type{Writer}, io::IO, compress::Symbol, args...)
     open(Writer, io, compressor, args...)
 end
 
-function Base.open(::Type{Writer}, io::IO; compress::Union{Nothing,Symbol,LZ4FrameCompressor,<:AbstractVector{LZ4FrameCompressor},ZstdCompressor,<:AbstractVector{ZstdCompressor}}=nothing, file::Bool=true, largelists::Bool=false, denseunions::Bool=true, dictencode::Bool=false, dictencodenested::Bool=false, alignment::Integer=8, maxdepth::Integer=DEFAULT_MAX_DEPTH, ntasks::Integer=typemax(Int32), metadata::Union{Nothing,Any}=nothing, colmetadata::Union{Nothing,Any}=nothing, closeio::Bool=false)
+function Base.open(::Type{Writer}, io::IO; compress::Union{Nothing,Symbol,LZ4FrameCompressor,<:AbstractVector{LZ4FrameCompressor},ZstdCompressor,<:AbstractVector{ZstdCompressor}}=nothing, file::Bool=true, largelists::Bool=false, denseunions::Bool=true, dictencode::Bool=false, dictencodenested::Bool=false, alignment::Integer=DEFAULT_BYTE_ALIGNMENT, maxdepth::Integer=DEFAULT_MAX_DEPTH, ntasks::Integer=typemax(Int32), metadata::Union{Nothing,Any}=nothing, colmetadata::Union{Nothing,Any}=nothing, closeio::Bool=false)
     open(Writer, io, compress, file, largelists, denseunions, dictencode, dictencodenested, alignment, maxdepth, ntasks, metadata, colmetadata, closeio)
 end