Module compute 

Compute kernels for NdArrowArray, with NumPy and xarray semantics.

Axes are addressed by dimension name. Broadcasting aligns axes by name. Missing values are Arrow nulls: kernels skip nulls like the NumPy nan* functions.

Attribute rules follow xarray defaults:

• layout kernels, cast, where_, fill_null and clip keep the attributes;
• arithmetic, comparisons, math functions and reductions drop them;
• decode_cf and encode_cf move the CF encoding keys.

Re-exports

pub use aggregate::CoarsenBoundary;

pub use aggregate::coarsen;

pub use aggregate::dot;

pub use aggregate::weighted_mean;

pub use aggregate::weighted_sum;

pub use cf::CfEncoding;

pub use cf::decode_cf;

pub use cf::encode_cf;

pub use column::column_binary;

pub use column::column_cast;

pub use column::column_decode_cf;

pub use column::column_unary;

pub use column::row_reduce;

pub use column::row_reduce_axes;

pub use groupby::groupby_reduce;

pub use histogram::bincount;

pub use histogram::histogram;

pub use interp::InterpMethod;

pub use interp::interp;

pub use interp::interp_with;

pub use partial::PartialReduction;

pub use partial::merge_all;

pub use resample::Frequency;

pub use resample::resample;

pub use search::SearchSide;

pub use search::argsort;

pub use search::argwhere;

pub use search::digitize;

pub use search::isin;

pub use search::searchsorted;

pub use search::sort;

pub use search::unique;

pub use time::CfTimeUnits;

pub use time::TimePeriod;

pub use time::decode_cf_time;

pub use time::encode_cf_time;

pub use time::floor_time;

pub use time::time_component;

pub use window::RollingOp;

pub use window::rolling;

pub use elementwise::*;

pub use reduce::*;

pub use shape::*;

Modules

aggregate

Block and weighted aggregation: coarsen, weighted_sum, weighted_mean, dot.

cf

CF conventions: mask fill values and unpack scale_factor / add_offset.

column

Kernels on a whole NdColumn: one kernel call for all rows.

elementwise

Elementwise kernels with name-based broadcasting and NumPy type promotion.

groupby

Group-by reductions along one dimension, like xarray.DataArray.groupby.

histogram

Counting kernels: histogram and bincount.

interp

Linear interpolation along one dimension.

partial

Reduction states that merge exactly across chunks.

reduce

Reductions along named axes, cumulative operations and differences.

resample

Resampling along a time dimension, like xarray.DataArray.resample.

search

Sorting and searching: searchsorted, sort, argsort, unique.

shape

Layout kernels: broadcast, transpose, squeeze, expand, select, take, concat.

time

CF time coordinates and time labels for resampling.

window

Moving windows along one dimension, like xarray.DataArray.rolling.

Enums

DatePart

Valid parts to extract from date/time/timestamp arrays.