Skip to content

RAGFlow is an agent-native database: a Retrieval-Augmented Generation (RAG) engine that grew into context infrastructure.

fix(test): give the rag_agent dialog stub the retrieval settings it now reads (#19081)

MarMar Labsbf5cf26df6af
diff --git a/test/unit_test/api/db/services/test_dialog_service_rag_agent_image_routing.py b/test/unit_test/api/db/services/test_dialog_service_rag_agent_image_routing.py
index 8805a59292..527ad29cda 100644
--- a/test/unit_test/api/db/services/test_dialog_service_rag_agent_image_routing.py
+++ b/test/unit_test/api/db/services/test_dialog_service_rag_agent_image_routing.py
@@ -103,6 +103,14 @@ def _dialog(llm_id="qwen3-vl-plus@Tongyi-Qianwen"):
llm_setting={"temperature": 0.1},
prompt_config={},
meta_data_filter=None,
+ # rag_agent forwards the assistant's retrieval settings to RAGTools, so a
+ # dialog stub without them raises AttributeError before reaching anything
+ # these tests assert on. Values are the Dialog model's own defaults.
+ similarity_threshold=0.2,
+ vector_similarity_weight=0.3,
+ top_n=6,
+ rerank_candidates_count=64,
+ top_k=1024,
)