unstructured
82714cad - rfctr(chunking): extract BasePreChunker (#2294)

Commit
2 years ago
rfctr(chunking): extract BasePreChunker (#2294) The `_split_elements_by_title_and_table()` function fulfills the pre-chunker role for `chunk_by_title()`, but most of its operation is not strategy-specific and can be reused by other chunking strategies. Extract `BasePreChunker` and use it as the base class for `_ByTitlePreChunker` which now only needs to provide the boundary predicates specific to that strategy.
Author
Parents
Loading