all2md.search.chunking

Chunk generation utilities for document indexing.

class all2md.search.chunking.ChunkingContext

Bases: object

Describe the document targeted for chunk generation.

document_id: str

document_path: Path | None

metadata: Mapping[str, object]

__init__(document_id: str, document_path: Path | None, metadata: Mapping[str, object]) → None

all2md.search.chunking.chunk_document(doc: Document, *, context: ChunkingContext, chunk_size_tokens: int, chunk_overlap_tokens: int, min_chunk_tokens: int, include_preamble: bool, heading_merge: bool, max_heading_level: int | None = None, progress_callback: Callable[[ProgressEvent], None] | None = None) → list[Chunk]: Split an AST document into search-friendly text chunks.