Struct LateChunkConfig 

pub struct LateChunkConfig {
    pub chunk_size: usize,
    pub chunk_overlap: usize,
}

Configuration for late chunking.

Fields

chunk_size: usize

Chunk size in bytes (chunks are byte windows over the document).

chunk_overlap: usize

Overlap between consecutive chunks in bytes.

Implementations

impl LateChunkConfig

pub fn new() -> Self

Creates a config with defaults.

pub fn with_chunk_size(self, chunk_size: usize) -> Self

Sets the chunk size in bytes.

pub fn with_chunk_overlap(self, chunk_overlap: usize) -> Self

Sets the chunk overlap in bytes.

pub fn validate(&self) -> Result<(), EmbeddingError>

Validates the config: chunk_overlap < chunk_size (otherwise chunks would not advance).

pub fn chunk_ranges(&self, text: &str) -> Vec<(usize, usize)>

Computes the chunk byte ranges [start, end) over text.

Ranges are monotonically increasing, non-empty, cover the whole document, and — 0.22.0 C6 fix — are always snapped to UTF-8 char boundaries (the window advances in bytes for stable sizing, then edges snap inward; slicing text[start..end] can no longer panic on CJK / emoji text whose character straddles a window edge).

Trait Implementations

impl Clone for LateChunkConfig

fn clone(&self) -> LateChunkConfig

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for LateChunkConfig

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for LateChunkConfig

fn default() -> Self

Returns the “default value” for a type.