refactor: Add better handling of configuration for dataset to database handler

.env.template

@@ -93,6 +93,8 @@ DB_NAME=cognee_db
 
 # Default (local file-based)
 GRAPH_DATABASE_PROVIDER="kuzu"
+# Handler for multi-user access control mode, it handles how should the mapping/creation of separate DBs be handled per Cognee dataset
+GRAPH_DATASET_DATABASE_HANDLER="kuzu"
 
 # -- To switch to Remote Kuzu uncomment and fill these: -------------------------------------------------------------
 #GRAPH_DATABASE_PROVIDER="kuzu"
@@ -117,6 +119,8 @@ VECTOR_DB_PROVIDER="lancedb"
 # Not needed if a cloud vector database is not used
 VECTOR_DB_URL=
 VECTOR_DB_KEY=
+# Handler for multi-user access control mode, it handles how should the mapping/creation of separate DBs be handled per Cognee dataset
+VECTOR_DATASET_DATABASE_HANDLER="lancedb"
 
 ################################################################################
 # 🧩 Ontology resolver settings

cognee/context_global_variables.py

@@ -4,8 +4,8 @@ from typing import Union
 from uuid import UUID
 
 from cognee.base_config import get_base_config
-from cognee.infrastructure.databases.vector.config import get_vectordb_context_config
-from cognee.infrastructure.databases.graph.config import get_graph_context_config
+from cognee.infrastructure.databases.vector.config import get_vectordb_config
+from cognee.infrastructure.databases.graph.config import get_graph_config
 from cognee.infrastructure.databases.utils import get_or_create_dataset_database
 from cognee.infrastructure.files.storage.config import file_storage_config
 from cognee.modules.users.methods import get_user
@@ -16,23 +16,59 @@ vector_db_config = ContextVar("vector_db_config", default=None)
 graph_db_config = ContextVar("graph_db_config", default=None)
 session_user = ContextVar("session_user", default=None)
 
-VECTOR_DBS_WITH_MULTI_USER_SUPPORT = ["lancedb", "falkor"]
-GRAPH_DBS_WITH_MULTI_USER_SUPPORT = ["kuzu", "falkor", "neo4j"]
-
 
 async def set_session_user_context_variable(user):
     session_user.set(user)
 
 
 def multi_user_support_possible():
-    graph_db_config = get_graph_context_config()
-    vector_db_config = get_vectordb_context_config()
-    # TODO: Make sure dataset database handler and provider match, remove multi_user support check, add error if no dataset database handler exists for provider
-    return (
-        graph_db_config["graph_database_provider"] in GRAPH_DBS_WITH_MULTI_USER_SUPPORT
-        and vector_db_config["vector_db_provider"] in VECTOR_DBS_WITH_MULTI_USER_SUPPORT
+    graph_db_config = get_graph_config()
+    vector_db_config = get_vectordb_config()
+
+    graph_handler = graph_db_config.graph_dataset_database_handler
+    vector_handler = vector_db_config.vector_dataset_database_handler
+    from cognee.infrastructure.databases.dataset_database_handler import (
+        supported_dataset_database_handlers,
     )
 
+    if graph_handler not in supported_dataset_database_handlers:
+        raise EnvironmentError(
+            "Unsupported graph dataset to database handler configured. Cannot add support for multi-user access control mode. Please use a supported graph dataset to database handler or set the environment variables ENABLE_BACKEND_ACCESS_CONTROL to false to switch off multi-user access control mode.\n"
+            f"Selected graph dataset to database handler: {graph_handler}\n"
+            f"Supported dataset to database handlers: {list(supported_dataset_database_handlers.keys())}\n"
+        )
+
+    if vector_handler not in supported_dataset_database_handlers:
+        raise EnvironmentError(
+            "Unsupported vector dataset to database handler configured. Cannot add support for multi-user access control mode. Please use a supported vector dataset to database handler or set the environment variables ENABLE_BACKEND_ACCESS_CONTROL to false to switch off multi-user access control mode.\n"
+            f"Selected vector dataset to database handler: {vector_handler}\n"
+            f"Supported dataset to database handlers: {list(supported_dataset_database_handlers.keys())}\n"
+        )
+
+    if (
+        supported_dataset_database_handlers[graph_handler]["handler_provider"]
+        != graph_db_config.graph_database_provider
+    ):
+        raise EnvironmentError(
+            "The selected graph dataset to database handler does not work with the configured graph database provider. Cannot add support for multi-user access control mode. Please use a supported graph dataset to database handler or set the environment variables ENABLE_BACKEND_ACCESS_CONTROL to false to switch off multi-user access control mode.\n"
+            f"Selected graph database provider: {graph_db_config.graph_database_provider}\n"
+            f"Selected graph dataset to database handler: {graph_handler}\n"
+            f"Supported dataset to database handlers: {list(supported_dataset_database_handlers.keys())}\n"
+        )
+
+    if (
+        supported_dataset_database_handlers[vector_handler]["handler_provider"]
+        != vector_db_config.vector_db_provider
+    ):
+        raise EnvironmentError(
+            "The selected vector dataset to database handler does not work with the configured vector database provider. Cannot add support for multi-user access control mode. Please use a supported vector dataset to database handler or set the environment variables ENABLE_BACKEND_ACCESS_CONTROL to false to switch off multi-user access control mode.\n"
+            f"Selected vector database provider: {vector_db_config.vector_db_provider}\n"
+            f"Selected vector dataset to database handler: {vector_handler}\n"
+            f"Supported dataset to database handlers: {list(supported_dataset_database_handlers.keys())}\n"
+        )
+
+    return True
+
 
 def backend_access_control_enabled():
     backend_access_control = os.environ.get("ENABLE_BACKEND_ACCESS_CONTROL", None)
@@ -42,12 +78,7 @@ def backend_access_control_enabled():
         return multi_user_support_possible()
     elif backend_access_control.lower() == "true":
         # If enabled, ensure that the current graph and vector DBs can support it
-        multi_user_support = multi_user_support_possible()
-        if not multi_user_support:
-            raise EnvironmentError(
-                "ENABLE_BACKEND_ACCESS_CONTROL is set to true but the current graph and/or vector databases do not support multi-user access control. Please use supported databases or disable backend access control."
-            )
-        return True
+        return multi_user_support_possible()
     return False
 
 

cognee/infrastructure/databases/dataset_database_handler/supported_dataset_database_handlers.py

@@ -9,7 +9,10 @@ from cognee.infrastructure.databases.graph.kuzu.KuzuDatasetDatabaseHandler impor
 )
 
 supported_dataset_database_handlers = {
-    "neo4j_aura": Neo4jAuraDatasetDatabaseHandler,
-    "lancedb": LanceDBDatasetDatabaseHandler,
-    "kuzu": KuzuDatasetDatabaseHandler,
+    "neo4j_aura": {
+        "handler_instance": Neo4jAuraDatasetDatabaseHandler,
+        "handler_provider": "neo4j",
+    },
+    "lancedb": {"handler_instance": LanceDBDatasetDatabaseHandler, "handler_provider": "lancedb"},
+    "kuzu": {"handler_instance": KuzuDatasetDatabaseHandler, "handler_provider": "kuzu"},
 }

cognee/infrastructure/databases/dataset_database_handler/use_dataset_database_handler.py

@@ -1,5 +1,10 @@
 from .supported_dataset_database_handlers import supported_dataset_database_handlers
 
 
-def use_dataset_database_handler(dataset_database_handler_name, dataset_database_handler):
-    supported_dataset_database_handlers[dataset_database_handler_name] = dataset_database_handler
+def use_dataset_database_handler(
+    dataset_database_handler_name, dataset_database_handler, dataset_database_provider
+):
+    supported_dataset_database_handlers[dataset_database_handler_name] = {
+        "handler_instance": dataset_database_handler,
+        "handler_provider": dataset_database_provider,
+    }

cognee/infrastructure/databases/utils/get_or_create_dataset_database.py

@@ -1,13 +1,9 @@
-import os
-import asyncio
-import requests
 from uuid import UUID
 from typing import Union, Optional
 
 from sqlalchemy import select
 from sqlalchemy.exc import IntegrityError
 
-from cognee.base_config import get_base_config
 from cognee.modules.data.methods import create_dataset
 from cognee.infrastructure.databases.relational import get_relational_engine
 from cognee.infrastructure.databases.vector import get_vectordb_config
@@ -25,7 +21,7 @@ async def _get_vector_db_info(dataset_id: UUID, user: User) -> dict:
     )
 
     handler = supported_dataset_database_handlers[vector_config.vector_dataset_database_handler]
-    return await handler.create_dataset(dataset_id, user)
+    return await handler["handler_instance"].create_dataset(dataset_id, user)
 
 
 async def _get_graph_db_info(dataset_id: UUID, user: User) -> dict:
@@ -36,7 +32,7 @@ async def _get_graph_db_info(dataset_id: UUID, user: User) -> dict:
     )
 
     handler = supported_dataset_database_handlers[graph_config.graph_dataset_database_handler]
-    return await handler.create_dataset(dataset_id, user)
+    return await handler["handler_instance"].create_dataset(dataset_id, user)
 
 
 async def _existing_dataset_database(

cognee/shared/logging_utils.py

@@ -534,6 +534,10 @@ def setup_logging(log_level=None, name=None):
     # Get a configured logger and log system information
     logger = structlog.get_logger(name if name else __name__)
 
+    logger.warning(
+        "From version 0.5.0 onwards, Cognee will run with multi-user access control mode set to on by default. Data isolation between different users and datasets will be enforced and data created before multi-user access control mode was turned on won't be accessible by default. To disable multi-user access control mode and regain access to old data set the environment variable ENABLE_BACKEND_ACCESS_CONTROL to false before starting Cognee. For more information, please refer to the Cognee documentation."
+    )
+
     if logs_dir is not None:
         logger.info(f"Log file created at: {log_file_path}", log_file=log_file_path)