Hypergraph is more powerful. Transform unstructured text into structured knowledge with LLMs. Graphs, hypergraphs, and spatio-temporal extractions — with one command.
fix: forward source/tag scope through Cog_RAG.chat
Frankie-Xu4ff75e63f0b8
diff --git a/hyperextract/methods/rag/cog_rag.py b/hyperextract/methods/rag/cog_rag.pyindex e174f9c..da9a12c 100644--- a/hyperextract/methods/rag/cog_rag.py+++ b/hyperextract/methods/rag/cog_rag.py@@ -429,9 +429,23 @@ class Cog_RAG:return {"themes": themes, "entities": entities}- def chat(self, query: str, top_k_themes: int = 3, top_k_entities: int = 3):+ def chat(+ self,+ query: str,+ top_k_themes: int = 3,+ top_k_entities: int = 3,+ *,+ source_ids: list[str] | None = None,+ tags: list[str] | None = None,+ ):"""Generate an answer using context from both layers."""- results = self.search(query, top_k_themes, top_k_entities)+ results = self.search(+ query,+ top_k_themes,+ top_k_entities,+ source_ids=source_ids,+ tags=tags,+ )context_parts = []if results["themes"]:diff --git a/tests/methods/test_search_scope_forwarding.py b/tests/methods/test_search_scope_forwarding.pyindex c482eb9..3c627a4 100644--- a/tests/methods/test_search_scope_forwarding.py+++ b/tests/methods/test_search_scope_forwarding.py@@ -79,6 +79,23 @@ class TestCogRagScope:assert calls["edges"] == {"source_ids": ["s1"], "tags": ["t1"]}assert calls["nodes"] == {"source_ids": ["s1"], "tags": ["t1"]}+ def test_chat_forwards_scope_to_search(self):+ cog = Cog_RAG.__new__(Cog_RAG)+ seen = {}++ def _search(query, top_k_themes=3, top_k_entities=3, **kwargs):+ seen["kwargs"] = kwargs+ return {"themes": [], "entities": []}++ cog.search = _search+ from langchain_core.runnables import RunnableLambda++ cog.llm = RunnableLambda(lambda _: "ok")++ cog.chat("q", source_ids=["s1"], tags=["t1"])++ assert seen["kwargs"] == {"source_ids": ["s1"], "tags": ["t1"]}+def test_cog_rag_uses_hypergraph_layers(self):"""ThemeSchema participates in the theme layer's hyperedge schema."""assert issubclass(ThemeSchema, object)