Expand description
contains a wide range of compute operations (e.g.
arithmetics
, aggregate
,
filter
, comparison
, and sort
)
This module’s general design is
that each operator has two interfaces, a statically-typed version and a dynamically-typed
version.
The statically-typed version expects concrete arrays (such as PrimitiveArray
);
the dynamically-typed version expects &dyn Array
and errors if the the type is not
supported.
Some dynamically-typed operators have an auxiliary function, can_*
, that returns
true if the operator can be applied to the particular DataType
.
Modules
compute_aggregate
Contains different aggregation functions
compute_arithmetics
Defines basic arithmetic kernels for PrimitiveArray
s.
Defines kernels suitable to perform operations to primitive arrays.
Defines generics suitable to perform operations to PrimitiveArray
in-place.
compute_boolean_kleene
Boolean operators of Kleene logic.
compute_cast
Defines different casting operators such as cast
or primitive_to_binary
.
compute_comparison
Contains comparison operators
compute_concatenate
Contains the concatenate kernel
compute_hash
Contains the hash
and typed (e.g. hash_primitive
) operators.
compute_if_then_else
Contains the operator if_then_else
.
compute_length
Defines kernel for length of composite arrays
compute_like
Contains “like” operators such as like_utf8
and like_utf8_scalar
.
compute_merge_sort
Functions to perform merge-sorts.
compute_partition
Defines partition kernel for crate::array::Array
compute_regex_match
Contains regex matching operators regex_match
and regex_match_scalar
.
compute_substring
Defines kernel to extract a substring of a [Large]StringArray
compute_temporal
Defines temporal kernels for time and date related functions.
compute_window
Defines windowing functions, like shift
ing