Trait Compressor 

pub trait Compressor {
    type ArrayVTable: VTable;
    type SchemeType: Scheme<StatsType = Self::StatsType> + ?Sized;
    type StatsType: CompressorStats<ArrayVTable = Self::ArrayVTable>;

    // Required methods
    fn schemes() -> &'static [&'static Self::SchemeType];
    fn default_scheme() -> &'static Self::SchemeType;
    fn dict_scheme_code() -> <Self::SchemeType as Scheme>::CodeType;

    // Provided methods
    fn compress(
        array: &<Self::ArrayVTable as VTable>::Array,
        is_sample: bool,
        allowed_cascading: usize,
        excludes: &[<Self::SchemeType as Scheme>::CodeType],
    ) -> VortexResult<ArrayRef>
       where Self::SchemeType: 'static { ... }
    fn choose_scheme(
        stats: &Self::StatsType,
        is_sample: bool,
        allowed_cascading: usize,
        excludes: &[<Self::SchemeType as Scheme>::CodeType],
    ) -> VortexResult<&'static Self::SchemeType> { ... }
}

A compressor for a particular input type.

This trait defines the interface for type-specific compressors that can adaptively choose and apply compression schemes based on data characteristics. Compressors analyze input arrays, select optimal compression schemes, and handle cascading compression with multiple encoding layers.

The compressor works by generating statistics on the input data, evaluating available compression schemes, and selecting the one with the best compression ratio.

Required Associated Types

type ArrayVTable: VTable

The VTable type for arrays this compressor operates on.

type SchemeType: Scheme<StatsType = Self::StatsType> + ?Sized

The compression scheme type used by this compressor.

type StatsType: CompressorStats<ArrayVTable = Self::ArrayVTable>

The statistics type used to analyze arrays for compression.

Required Methods

fn schemes() -> &'static [&'static Self::SchemeType]

Returns all available compression schemes for this compressor.

fn default_scheme() -> &'static Self::SchemeType

Returns the default fallback compression scheme.

fn dict_scheme_code() -> <Self::SchemeType as Scheme>::CodeType

Returns the scheme code for dictionary compression.

Provided Methods

fn compress( array: &<Self::ArrayVTable as VTable>::Array, is_sample: bool, allowed_cascading: usize, excludes: &[<Self::SchemeType as Scheme>::CodeType], ) -> VortexResult<ArrayRef>

where Self::SchemeType: 'static,

Compresses an array using the optimal compression scheme.

Generates statistics on the input array, selects the best compression scheme, and applies it. Returns the original array if compression would increase size.

fn choose_scheme( stats: &Self::StatsType, is_sample: bool, allowed_cascading: usize, excludes: &[<Self::SchemeType as Scheme>::CodeType], ) -> VortexResult<&'static Self::SchemeType>

Selects the best compression scheme based on expected compression ratios.

Evaluates all available schemes against the provided statistics and returns the one with the highest compression ratio. Falls back to the default scheme if no scheme provides compression benefits.

Dyn Compatibility

This trait is not dyn compatible.

In older versions of Rust, dyn compatibility was called "object safety", so this trait is not object safe.

Implementors

impl Compressor for FloatCompressor

type ArrayVTable = PrimitiveVTable

type SchemeType = dyn FloatScheme<StatsType = FloatStats, CodeType = FloatCode>

type StatsType = FloatStats

impl Compressor for IntCompressor

type ArrayVTable = PrimitiveVTable

type SchemeType = dyn IntegerScheme<StatsType = IntegerStats, CodeType = IntCode>

type StatsType = IntegerStats

impl Compressor for StringCompressor

type ArrayVTable = VarBinViewVTable

type SchemeType = dyn StringScheme<StatsType = StringStats, CodeType = StringCode>

type StatsType = StringStats