Types

Array layout classes

In Python, we make a distinction between high-level ak.Array (for data analysts) and low-level Content memory layouts (for downstream developers). In Julia, it's more advantageous to expose the concrete type details to all users, particularly for defining functions with multiple dispatch. Thus, there is no ak.Array equivalent.

The layout classes (subclasses of AwkwardArray.Content) are:

Julia class	corresponding Python	corresponding Arrow	description
`PrimitiveArray`	NumpyArray	primitive	one-dimensional array of booleans, numbers, date-times, or time-differences
`EmptyArray`	EmptyArray	(none)	length-zero array with unknown type (usually derived from untyped sources)
`ListOffsetArray`	ListOffsetArray	list	variable-length lists defined by an index of `offsets`
`ListArray`	ListArray	(none)	variable-length lists defined by more general `starts` and `stops` indexes
`RegularArray`	RegularArray	fixed-size	lists of uniform `size`
`RecordArray`	RecordArray with `fields`	struct	struct-like records with named fields of different types
`TupleArray`	RecordArray with `fields=None`	(none)	tuples of unnamed fields of different types
`IndexedArray`	IndexedArray	dictionary	data that are lazily filtered, duplicated, and/or rearranged by an integer `index`
`IndexedOptionArray`	IndexedOptionArray	(none)	same but negative values in the `index` correspond to `Missing` values
`ByteMaskedArray`	ByteMaskedArray	(none)	possibly-missing data, defined by a byte `mask`
`BitMaskedArray` (only `lsb_order = true`)	BitMaskedArray	bitmaps	same, defined by a `BitVector`
`UnmaskedArray`	UnmaskedArray	same	in-principle missing data, but none are actually missing so no mask
`UnionArray`	UnionArray	dense union	data of different types in the same array

Any node in the data-type tree can carry Dict{String,Any} metadata as parameters, as well as a behavior::Symbol that can be used to define specialized behaviors. For instance, arrays of strings (constructed with StringOffsetArray, StringArray, or StringRegularArray) are defined by behavior = :string (instead of behavior = :default).

Types specification

AwkwardArray.BitMaskedArray — Type

BitMaskedArray{CONTENT<:Content,BEHAVIOR} <: OptionType{BEHAVIOR}

Specialized array type designed to handle masked arrays, where certain elements can be marked as valid or invalid using a BitVector.

Inherits from OptionType.

struct BitMaskedArray{CONTENT<:Content,BEHAVIOR} <: OptionType{BEHAVIOR}
    mask::BitVector
    content::CONTENT
    valid_when::Bool
    parameters::Parameters

    BitMaskedArray(
        mask::BitVector,
        content::CONTENT;
        valid_when::Bool = false,
        parameters::Parameters = Parameters(),
        behavior::Symbol = :default,
    ) where {CONTENT<:Content} =
        new{CONTENT,behavior}(mask, content, valid_when, parameters)
end

Type Parameters:

CONTENT<:Content: The CONTENT type parameter is constrained to be a subtype of Content.
BEHAVIOR: A type parameter that can represent different behaviors associated with the array.

Inheritance:

<: OptionType{BEHAVIOR}: Indicates that BitMaskedArray is a subtype of OptionType parameterized by BEHAVIOR.

Fields:

mask::BitVector: A BitVector indicating which elements are valid or invalid.
content::CONTENT: The actual data content, constrained to be a subtype of Content.
valid_when::Bool: A flag indicating when the mask is valid (by default false).

Note

NumPy MaskedArray's convention; note that Arrow's is true.

parameters::Parameters: Additional parameters associated with the array, defined elsewhere.

Constructor:

BitMaskedArray(
    mask::BitVector, 
    content::CONTENT; 
    valid_when::Bool = false, 
    parameters::Parameters = Parameters(), 
    behavior::Symbol = :default
) where {CONTENT<:Content}

This is the outer constructor for the BitMaskedArray struct. It initializes a new instance of BitMaskedArray with the given mask, content, and optional valid_when, parameters, and behavior. The where {CONTENT<:Content} clause ensures that CONTENT satisfies the specified constraint.

new{CONTENT,behavior}(mask, content, valid_when, parameters)

creates a new instance of BitMaskedArray with the specified type parameters and field values.

Note

All Python BitMaskedArrays must be converted to lsb_order = true.

Types

Array layout classes

Types specification

Examples

Index