RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
fix(metadata): preserve empty AND results across filter conditions (#15386)
## Summary - Fix `meta_filter()` AND logic so an empty result from an early condition is not overwritten when a later condition matches. - Add regression tests for empty-first AND, successful AND intersection, and OR behavior after an empty first condition. Fixes incorrect `/retrieval` metadata filtering when multiple AND conditions are used and the first condition matches no documents. Closes #15360 ## Test plan - [x] `pytest test/unit_test/common/test_metadata_filter_operators.py -v` (19/19 passed)
M
monsterDavid committed
53bb2bd9e8ad52acf04833136dac6f648174e35d
Parent: 2d229dd
Committed by GitHub <noreply@github.com>
on 5/29/2026, 11:33:26 AM