refactoring for collection_name input

utils/retriever.py

@@ -209,7 +209,7 @@ def get_context(
                 search_kwargs["filter"] = filter_obj
         
         # Perform initial retrieval
-        retrieved_docs = vectorstore.search(query, top_k, **search_kwargs)
+        retrieved_docs = vectorstore.search(query, collection_name,  top_k,  **search_kwargs)
         
         logging.info(f"Retrieved {len(retrieved_docs)} documents for query: {query[:50]}...")
         

utils/vectorstore_interface.py

@@ -55,7 +55,7 @@ class HuggingFaceSpacesVectorStore(VectorStoreInterface):
 class QdrantVectorStore(VectorStoreInterface):
     """Vector store implementation for direct Qdrant connection."""
     
-    def __init__(self, url: str, collection_name: str, api_key: Optional[str] = None):
+    def __init__(self, url: str, api_key: Optional[str] = None):
         from qdrant_client import QdrantClient
         from sentence_transformers import SentenceTransformer
 
@@ -101,7 +101,7 @@ class QdrantVectorStore(VectorStoreInterface):
             
         return self._embedding_model
         
-    def search(self, query: str, top_k: int, **kwargs) -> List[Dict[str, Any]]:
+    def search(self, query: str, collection_name:str, top_k: int, **kwargs) -> List[Dict[str, Any]]:
         """Search using direct Qdrant connection."""
         try:
             # Get embedding model
@@ -118,7 +118,7 @@ class QdrantVectorStore(VectorStoreInterface):
             # Perform vector search
             logging.info(f"Searching Qdrant collection '{self.collection_name}' for top {top_k} results")
             search_result = self.client.search(
-                collection_name=self.collection_name,
+                collection_name=collection_name,
                 query_vector=query_embedding,
                 query_filter=filter_obj,  # Add filter support
                 limit=top_k,
@@ -162,10 +162,10 @@ def create_vectorstore(config: Any) -> VectorStoreInterface:
     
     elif vectorstore_type.lower() == "qdrant":
         url = config.get("vectorstore", "URL")  # Use the full URL
-        collection_name = config.get("vectorstore", "COLLECTION_NAME")
+        #collection_name = config.get("vectorstore", "COLLECTION_NAME")
         api_key = auth_config["api_key"]
         # Remove port parameter since it's included in the URL
-        return QdrantVectorStore(url, collection_name, api_key)
+        return QdrantVectorStore(url, api_key)
     
     else:
         raise ValueError(f"Unsupported vector store type: {vectorstore_type}")