enable semantic cache

backend/app/api/admin_routes/semantic_cache.py

@@ -1,46 +1,57 @@
-from typing import Optional, List
+from typing import Optional, List, Dict
 
 from fastapi import APIRouter
 from app.models import Document
 from app.api.deps import SessionDep, CurrentSuperuserDep
-from app.rag.chat_config import ChatEngineConfig, get_default_embedding_model
+from app.rag.chat_config import ChatEngineConfig
 from app.rag.semantic_cache import SemanticCacheManager, SemanticItem
 
 router = APIRouter()
 
+
 @router.post("/admin/semantic_cache")
-async def retrieve_documents(
+async def add_semantic_cache(
     session: SessionDep,
     user: CurrentSuperuserDep,
     question: str,
     answer: str,
     namespace: str = "default",
     metadata: Optional[dict] = None,
     chat_engine: str = "default",
-) -> List[Document]:
+) -> Dict:
     chat_engine_config = ChatEngineConfig.load_from_db(session, chat_engine)
     _dspy_lm = chat_engine_config.get_dspy_lm(session)
 
     scm = SemanticCacheManager(
         dspy_llm=_dspy_lm,
     )
 
-    return scm.add_cache(
-        session,
-        item=SemanticItem(question=question, answer=answer),
-        namespace=namespace,
-        metadata=metadata,
-    )
+    try:
+        scm.add_cache(
+            session,
+            item=SemanticItem(question=question, answer=answer),
+            namespace=namespace,
+            metadata=metadata,
+        )
+    except Exception as e:
+        return {
+            "status": "failed",
+            "message": str(e),
+        }
+
+    return {
+        "status": "success",
+    }
 
 
 @router.get("/admin/semantic_cache")
-async def retrieve_documents(
+async def search_semantic_cache(
     session: SessionDep,
     user: CurrentSuperuserDep,
     query: str,
     namespace: str = "default",
     chat_engine: str = "default",
-) -> List[Document]:
+) -> Dict:
     chat_engine_config = ChatEngineConfig.load_from_db(session, chat_engine)
     _dspy_lm = chat_engine_config.get_dspy_lm(session)