Struct RecursiveCharSplitter 

pub struct RecursiveCharSplitter { /* private fields */ }

Splits text by character count, falling back through a list of separators until each chunk is at most chunk_size characters.

Adjacent chunks share a chunk_overlap window so context isn’t lost at boundaries. Mirrors LangChain’s RecursiveCharacterTextSplitter in behaviour but operates on Rust &str and counts characters, not tokens.

Implementations

impl RecursiveCharSplitter

pub fn new() -> Self

Construct with default settings.

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

Set the target maximum chunk size (in chars).

pub fn with_overlap(self, n: usize) -> Self

Set the overlap window between adjacent chunks.

pub fn with_separators<I, S>(self, seps: I) -> Self

where I: IntoIterator<Item = S>, S: Into<String>,

Override the separator list (tried in order, coarsest first).

Trait Implementations

impl Default for RecursiveCharSplitter

fn default() -> Self

Returns the “default value” for a type.

impl TextSplitter for RecursiveCharSplitter

fn split(&self, doc: &Document) -> Vec<Document>

Split doc into chunks.

fn split_all(&self, docs: &[Document]) -> Vec<Document>

Convenience: split many documents and concatenate the chunks.