Use MongoDB Altas without required Documents (#1538)

* Working on mongodb

* Working on retriever tool

* Add vectorstore retriever

* Fix format

---------

Co-authored-by: Remco Goyvaerts <remco.goyvaerts@acagroup.be>

src/backend/langflow/components/retrievers/VectorStoreRetriever.py

@@ -0,0 +1,17 @@
+from langchain_core.vectorstores import VectorStoreRetriever
+
+from langflow import CustomComponent
+from langflow.field_typing import VectorStore
+
+
+class VectoStoreRetrieverComponent(CustomComponent):
+    display_name = "VectorStore Retriever"
+    description = "A vector store retriever"
+
+    def build_config(self):
+        return {
+            "vectorstore": {"display_name": "Vector Store", "type": VectorStore},
+        }
+
+    def build(self, vectorstore: VectorStore) -> VectorStoreRetriever:
+        return vectorstore.as_retriever()

src/backend/langflow/components/tools/RetrieverTool.py

@@ -0,0 +1,32 @@
+from langchain.tools.retriever import create_retriever_tool
+
+from langflow import CustomComponent
+from langflow.field_typing import BaseRetriever, Tool
+
+
+class RetrieverToolComponent(CustomComponent):
+    display_name = "RetrieverTool"
+    description = "Tool for interacting with retriever"
+
+    def build_config(self):
+        return {
+            "retriever": {
+                "display_name": "Retriever",
+                "info": "Retriever to interact with",
+                "type": BaseRetriever,
+            },
+            "name": {"display_name": "Name", "info": "Name of the tool"},
+            "description": {"display_name": "Description", "info": "Description of the tool"},
+        }
+
+    def build(
+        self,
+        retriever: BaseRetriever,
+        name: str,
+        description: str,
+    ) -> Tool:
+        return create_retriever_tool(
+            retriever=retriever,
+            name=name,
+            description=description,
+        )

src/backend/langflow/components/vectorstores/MongoDBAtlasVectorSearch.py

@@ -1,22 +1,18 @@
 from typing import List, Optional
 
-from langchain_community.vectorstores import MongoDBAtlasVectorSearch
+from langchain_community.vectorstores.mongodb_atlas import MongoDBAtlasVectorSearch
 
 from langflow import CustomComponent
-from langflow.field_typing import (
-    Document,
-    Embeddings,
-    NestedDict,
-)
+from langflow.field_typing import Document, Embeddings, NestedDict
 
 
 class MongoDBAtlasComponent(CustomComponent):
     display_name = "MongoDB Atlas"
-    description = "Construct a `MongoDB Atlas Vector Search` vector store from raw documents."
+    description = "a `MongoDB Atlas Vector Search` vector store from raw documents."
 
     def build_config(self):
         return {
-            "documents": {"display_name": "Documents"},
+            "documents": {"display_name": "Documents", "is_list": True},
             "embedding": {"display_name": "Embedding"},
             "collection_name": {"display_name": "Collection Name"},
             "db_name": {"display_name": "Database Name"},
@@ -27,8 +23,8 @@ class MongoDBAtlasComponent(CustomComponent):
 
     def build(
         self,
-        documents: List[Document],
         embedding: Embeddings,
+        documents: Optional[List[Document]] = None,
         collection_name: str = "",
         db_name: str = "",
         index_name: str = "",
@@ -36,12 +32,17 @@ class MongoDBAtlasComponent(CustomComponent):
         search_kwargs: Optional[NestedDict] = None,
     ) -> MongoDBAtlasVectorSearch:
         search_kwargs = search_kwargs or {}
-        return MongoDBAtlasVectorSearch(
-            documents=documents,
+        vector_store = MongoDBAtlasVectorSearch.from_connection_string(
+            connection_string=mongodb_atlas_cluster_uri,
+            namespace=f"{db_name}.{collection_name}",
             embedding=embedding,
-            collection_name=collection_name,
-            db_name=db_name,
             index_name=index_name,
-            mongodb_atlas_cluster_uri=mongodb_atlas_cluster_uri,
-            search_kwargs=search_kwargs,
         )
+
+        if documents is not None:
+            if len(documents) == 0:
+                raise ValueError("If documents are provided, there must be at least one document.")
+
+            vector_store.add_documents(documents)
+
+        return vector_store