INNER CODE UNIT · Python

elements

Laurent00TT/PharosRAG · src/chunker/__init__.py:12

    elements = from_mineru(content_list_json, layout_json)     # parse-stage adapter
    result = Chunker().chunk(elements, doc_id="d1", doc_type="academic_paper", lang="en")
    # -> embed [c.text for c in result.chunks]; persist result.sections
    big = Chunker().assemble_big(hit_chunk, result, elements)  # query-time small-to-big
"""
from .core import RESTRICTED_ACL, Chunker
from .meta import extract_doc_meta
from .retrieve import assemble_big
from .types import BigBlock, Chunk, ChunkResult, Element, Section

__all__ = ["Chunker", "assemble_big", "extract_doc_meta", "RESTRICTED_ACL",
           "Element", "Chunk", "Section", "ChunkResult", "BigBlock"]

View source record →