Struct SemanticChunkConfig 

pub struct SemanticChunkConfig {
    pub target_size: usize,
    pub min_size: usize,
    pub max_size: usize,
    pub similarity_threshold: f32,
    pub overlap: usize,
    pub fallback_to_sentences: bool,
}

Configuration for semantic chunking.

Fields

target_size: usize

Target chunk size in characters (soft limit)

min_size: usize

Minimum chunk size in characters (hard limit)

max_size: usize

Maximum chunk size in characters (hard limit)

similarity_threshold: f32

Similarity threshold for chunk boundaries (0.0-1.0) Lower = more chunks, Higher = fewer chunks

overlap: usize

Overlap between chunks in characters

fallback_to_sentences: bool

Use sentence boundaries as fallback when similarity is ambiguous

Implementations

impl SemanticChunkConfig

pub fn long_document() -> SemanticChunkConfig

Create config optimized for long documents.

pub fn coreference() -> SemanticChunkConfig

Create config for coreference resolution (smaller chunks, higher similarity).

Trait Implementations

impl Clone for SemanticChunkConfig

fn clone(&self) -> SemanticChunkConfig

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for SemanticChunkConfig

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

Formats the value using the given formatter.

impl Default for SemanticChunkConfig

fn default() -> SemanticChunkConfig

Returns the “default value” for a type.