arro3.compute¶

arro3.compute ¶

add ¶

add(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs + rhs, returning an error on overflow

add_wrapping ¶

add_wrapping(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs + rhs, wrapping on overflow for integer data types.

can_cast_types ¶

can_cast_types(
    from_type: ArrowSchemaExportable, to_type: ArrowSchemaExportable
) -> bool

Return true if a value of type from_type can be cast into a value of to_type.

Parameters:

from_type (ArrowSchemaExportable) –

Source type
to_type (ArrowSchemaExportable) –

Destination type

Returns:

bool –

True if can be casted.

cast ¶

cast(input: ArrayInput, to_type: ArrowSchemaExportable) -> Array

cast(
    input: ArrowStreamExportable, to_type: ArrowSchemaExportable
) -> ArrayReader

cast(
    input: ArrayInput | ArrowStreamExportable, to_type: ArrowSchemaExportable
) -> Array | ArrayReader

Cast input to the provided data type and return a new Array with type to_type, if possible.

If input is an Array, an Array will be returned. If input is a ChunkedArray or ArrayReader, an ArrayReader will be returned.

Parameters:

input (ArrayInput | ArrowStreamExportable) –

Input data to cast.
to_type (ArrowSchemaExportable) –

The target data type to cast to. You may pass in a Field here if you wish to include Arrow extension metadata on the output array.

Returns:

Array | ArrayReader –

The casted Arrow data.

date_part ¶

date_part(input: ArrowArrayExportable, part: DatePart | DatePartT) -> Array

date_part(
    input: ArrowStreamExportable, part: DatePart | DatePartT
) -> ArrayReader

date_part(
    input: ArrowArrayExportable | ArrowStreamExportable,
    part: DatePart | DatePartT,
) -> Array | ArrayReader

Given an array, return a new array with the extracted [DatePart] as signed 32-bit integer values.

Currently only supports temporal types

Date32/Date64
Time32/Time64
Timestamp
Interval
Duration

Returns an int32-typed array unless input was a dictionary type, in which case returns the dictionary but with this function applied onto its values.

If array passed in is not of the above listed types (or is a dictionary array where the values array isn't of the above listed types), then this function will return an error.

Parameters:

array –

Argument to compute function.

Returns:

Array | ArrayReader –

The extracted date part.

dictionary_encode ¶

dictionary_encode(array: ArrayInput) -> Array

dictionary_encode(array: ArrowStreamExportable) -> ArrayReader

dictionary_encode(
    array: ArrayInput | ArrowStreamExportable,
) -> Array | ArrayReader

Dictionary-encode array.

Return a dictionary-encoded version of the input array. This function does nothing if the input is already a dictionary array.

Note: for stream input, each output array will not necessarily have the same dictionary.

Parameters:

array (ArrayInput | ArrowStreamExportable) –

Argument to compute function.

Returns:

Array | ArrayReader –

The dictionary-encoded array.

div ¶

div(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs / rhs

filter ¶

filter(values: ArrayInput, predicate: ArrayInput) -> Array

filter(
    values: ArrowStreamExportable, predicate: ArrowStreamExportable
) -> ArrayReader

filter(
    values: ArrayInput | ArrowStreamExportable,
    predicate: ArrayInput | ArrowStreamExportable,
) -> Array | ArrayReader

Returns a filtered values array where the corresponding elements of predicate are true.

If input is an Array, an Array will be returned. If input is a ChunkedArray or ArrayReader, an ArrayReader will be returned.

is_not_null ¶

is_not_null(input: ArrayInput) -> Array

is_not_null(input: ArrowStreamExportable) -> ArrayReader

is_not_null(input: ArrayInput | ArrowStreamExportable) -> Array | ArrayReader

Returns a non-null boolean-typed array with whether each value of the array is not null.

If input is an Array, an Array will be returned. If input is a ChunkedArray or ArrayReader, an ArrayReader will be returned.

Parameters:

input (ArrayInput | ArrowStreamExportable) –

Input data

Returns:

Array | ArrayReader –

Output

is_null ¶

is_null(input: ArrayInput) -> Array

is_null(input: ArrowStreamExportable) -> ArrayReader

is_null(input: ArrayInput | ArrowStreamExportable) -> Array | ArrayReader

Returns a non-null boolean-typed array with whether each value of the array is null.

If input is an Array, an Array will be returned. If input is a ChunkedArray or ArrayReader, an ArrayReader will be returned.

Parameters:

input (ArrayInput | ArrowStreamExportable) –

Input data

Returns:

Array | ArrayReader –

Output

max ¶

max(input: ArrayInput | ArrowStreamExportable) -> Scalar

Returns the max of values in the array.

min ¶

min(input: ArrayInput | ArrowStreamExportable) -> Scalar

Returns the min of values in the array.

mul ¶

mul(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs * rhs, returning an error on overflow

mul_wrapping ¶

mul_wrapping(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs * rhs, wrapping on overflow for integer data types.

neg ¶

neg(array: ArrayInput) -> Array

Negates each element of array, returning an error on overflow

neg_wrapping ¶

neg_wrapping(array: ArrayInput) -> Array

Negates each element of array, wrapping on overflow for integer data types.

rem ¶

rem(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs % rhs

sub ¶

sub(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs - rhs, returning an error on overflow

sub_wrapping ¶

sub_wrapping(lhs: ArrayInput, rhs: ArrayInput) -> Array

Perform lhs - rhs, wrapping on overflow for integer data types.

sum ¶

sum(input: ArrayInput | ArrowStreamExportable) -> Scalar

Returns the sum of values in the array.

take ¶

take(values: ArrayInput, indices: ArrayInput) -> Array

Take elements by index from Array, creating a new Array from those indexes.

┌─────────────────┐      ┌─────────┐                              ┌─────────────────┐
│        A        │      │    0    │                              │        A        │
├─────────────────┤      ├─────────┤                              ├─────────────────┤
│        D        │      │    2    │                              │        B        │
├─────────────────┤      ├─────────┤   take(values, indices)      ├─────────────────┤
│        B        │      │    3    │ ─────────────────────────▶   │        C        │
├─────────────────┤      ├─────────┤                              ├─────────────────┤
│        C        │      │    1    │                              │        D        │
├─────────────────┤      └─────────┘                              └─────────────────┘
│        E        │
└─────────────────┘
values array             indices array                            result

Parameters:

values (ArrayInput) –

The input Arrow data to select from.
indices (ArrayInput) –

The indices within values to take. This must be a numeric array.

Returns:

Array –

The selected arrow data.