Skip to content
Open
Show file tree
Hide file tree
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
5 changes: 5 additions & 0 deletions rag/retriever/keyword_search.py
Original file line number Diff line number Diff line change
Expand Up @@ -21,6 +21,11 @@ def index(self, chunks: list[dict]) -> None:
chunks: List of chunk dicts with 'text' field
"""
self.chunks = chunks
if not chunks:
# BM25Okapi([]) divides by zero in IDF; keep an empty, searchable index.
self.bm25 = None
logger.info("keyword_index_built", chunk_count=0)
return
tokenized_corpus = [self._tokenize(chunk["text"]) for chunk in chunks]
self.bm25 = BM25Okapi(tokenized_corpus)
logger.info("keyword_index_built", chunk_count=len(chunks))
Expand Down
4 changes: 0 additions & 4 deletions tests/unit/test_keyword_search.py
Original file line number Diff line number Diff line change
Expand Up @@ -131,10 +131,6 @@ def test_results_preserve_chunk_fields(self, searcher):
assert result["source"] == "readme"
assert result["custom"] == "value"

@pytest.mark.xfail(
strict=True,
reason="issue #68 (manifest H-01): BM25 keyword search raises ZeroDivisionError on an empty index",
)
def test_empty_index(self, searcher):
"""Test searching on empty index."""
searcher.index([])
Expand Down
Loading