cognee 0.2.1.dev6__py3-none-any.whl → 0.2.2.dev0__py3-none-any.whl

cognee/api/.env.example

@@ -0,0 +1,5 @@
+# Comma-separated list of allowed origins for CORS (leave empty to block all cross-origin requests)
+# Example:
+# CORS_ALLOWED_ORIGINS="https://yourdomain.com,https://another.com"
+# For local development, you might use:
+# CORS_ALLOWED_ORIGINS="http://localhost:3000"

cognee/api/client.py

@@ -67,13 +67,26 @@ async def lifespan(app: FastAPI):
 app = FastAPI(debug=app_environment != "prod", lifespan=lifespan)
 
 
+# Read allowed origins from environment variable (comma-separated)
+CORS_ALLOWED_ORIGINS = os.getenv("CORS_ALLOWED_ORIGINS")
+if CORS_ALLOWED_ORIGINS:
+    allowed_origins = [
+        origin.strip() for origin in CORS_ALLOWED_ORIGINS.split(",") if origin.strip()
+    ]
+else:
+    allowed_origins = [
+        "http://localhost:3000",
+    ]  # Block all except explicitly set origins
+
 app.add_middleware(
     CORSMiddleware,
-    allow_origins=["http://localhost:3000", "http://127.0.0.1:3000"],
+    allow_origins=allowed_origins,  # Now controlled by env var
     allow_credentials=True,
     allow_methods=["OPTIONS", "GET", "POST", "DELETE"],
     allow_headers=["*"],
 )
+# To allow origins, set CORS_ALLOWED_ORIGINS env variable to a comma-separated list, e.g.:
+# CORS_ALLOWED_ORIGINS="https://yourdomain.com,https://another.com"
 
 
 def custom_openapi():

cognee/api/v1/add/add.py

@@ -14,7 +14,7 @@ async def add(
     node_set: Optional[List[str]] = None,
     vector_db_config: dict = None,
     graph_db_config: dict = None,
-    dataset_id: UUID = None,
+    dataset_id: Optional[UUID] = None,
 ):
     """
     Add data to Cognee for knowledge graph processing.

cognee/api/v1/add/routers/get_add_router.py

@@ -1,16 +1,17 @@
 import os
+import requests
+import subprocess
 from uuid import UUID
 
-from fastapi import Form, UploadFile, Depends
-from fastapi.responses import JSONResponse
 from fastapi import APIRouter
-from typing import List, Optional
-import subprocess
-from cognee.shared.logging_utils import get_logger
-import requests
+from fastapi.responses import JSONResponse
+from fastapi import Form, File, UploadFile, Depends
+from typing import List, Optional, Union, Literal
 
 from cognee.modules.users.models import User
 from cognee.modules.users.methods import get_authenticated_user
+from cognee.shared.utils import send_telemetry
+from cognee.shared.logging_utils import get_logger
 
 logger = get_logger()
 
@@ -20,9 +21,9 @@ def get_add_router() -> APIRouter:
 
     @router.post("", response_model=dict)
     async def add(
-        data: List[UploadFile],
+        data: List[UploadFile] = File(default=None),
         datasetName: Optional[str] = Form(default=None),
-        datasetId: Optional[UUID] = Form(default=None),
+        datasetId: Union[UUID, Literal[""], None] = Form(default=None, examples=[""]),
         user: User = Depends(get_authenticated_user),
     ):
         """
@@ -38,7 +39,7 @@ def get_add_router() -> APIRouter:
           - GitHub repository URLs (will be cloned and processed)
           - Regular file uploads
         - **datasetName** (Optional[str]): Name of the dataset to add data to
-        - **datasetId** (Optional[UUID]): UUID of the dataset to add data to
+        - **datasetId** (Optional[UUID]): UUID of an already existing dataset
 
         Either datasetName or datasetId must be provided.
 
@@ -58,7 +59,16 @@ def get_add_router() -> APIRouter:
         - GitHub repositories are cloned and all files are processed
         - HTTP URLs are fetched and their content is processed
         - The ALLOW_HTTP_REQUESTS environment variable controls URL processing
+        - datasetId value can only be the UUID of an already existing dataset
         """
+        send_telemetry(
+            "Add API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/add",
+            },
+        )
+
         from cognee.api.v1.add import add as cognee_add
 
         if not datasetId and not datasetName:

cognee/api/v1/cognify/routers/get_cognify_router.py

@@ -1,7 +1,7 @@
 import os
 import asyncio
 from uuid import UUID
-from pydantic import BaseModel
+from pydantic import Field
 from typing import List, Optional
 from fastapi.responses import JSONResponse
 from fastapi import APIRouter, WebSocket, Depends, WebSocketDisconnect
@@ -10,7 +10,6 @@ from starlette.status import WS_1000_NORMAL_CLOSURE, WS_1008_POLICY_VIOLATION
 from cognee.api.DTO import InDTO
 from cognee.modules.pipelines.methods import get_pipeline_run
 from cognee.modules.users.models import User
-from cognee.shared.data_models import KnowledgeGraph
 from cognee.modules.users.methods import get_authenticated_user
 from cognee.modules.users.get_user_db import get_user_db_context
 from cognee.modules.graph.methods import get_formatted_graph_data
@@ -24,15 +23,16 @@ from cognee.modules.pipelines.queues.pipeline_run_info_queues import (
     remove_queue,
 )
 from cognee.shared.logging_utils import get_logger
+from cognee.shared.utils import send_telemetry
 
 
 logger = get_logger("api.cognify")
 
 
 class CognifyPayloadDTO(InDTO):
-    datasets: Optional[List[str]] = None
-    dataset_ids: Optional[List[UUID]] = None
-    run_in_background: Optional[bool] = False
+    datasets: Optional[List[str]] = Field(default=None)
+    dataset_ids: Optional[List[UUID]] = Field(default=None, examples=[[]])
+    run_in_background: Optional[bool] = Field(default=False)
 
 
 def get_cognify_router() -> APIRouter:
@@ -57,8 +57,7 @@ def get_cognify_router() -> APIRouter:
 
         ## Request Parameters
         - **datasets** (Optional[List[str]]): List of dataset names to process. Dataset names are resolved to datasets owned by the authenticated user.
-        - **dataset_ids** (Optional[List[UUID]]): List of dataset UUIDs to process. UUIDs allow processing of datasets not owned by the user (if permitted).
-        - **graph_model** (Optional[BaseModel]): Custom Pydantic model defining the knowledge graph schema. Defaults to KnowledgeGraph for general-purpose processing.
+        - **dataset_ids** (Optional[List[UUID]]): List of existing dataset UUIDs to process. UUIDs allow processing of datasets not owned by the user (if permitted).
         - **run_in_background** (Optional[bool]): Whether to execute processing asynchronously. Defaults to False (blocking).
 
         ## Response
@@ -84,6 +83,14 @@ def get_cognify_router() -> APIRouter:
         ## Next Steps
         After successful processing, use the search endpoints to query the generated knowledge graph for insights, relationships, and semantic search.
         """
+        send_telemetry(
+            "Cognify API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/cognify",
+            },
+        )
+
         if not payload.datasets and not payload.dataset_ids:
             return JSONResponse(
                 status_code=400, content={"error": "No datasets or dataset_ids provided"}

cognee/api/v1/config/config.py

@@ -182,15 +182,3 @@ class config:
     def set_vector_db_url(db_url: str):
         vector_db_config = get_vectordb_config()
         vector_db_config.vector_db_url = db_url
-
-    @staticmethod
-    def set_graphistry_config(graphistry_config: dict[str, str]):
-        base_config = get_base_config()
-
-        if "username" not in graphistry_config or "password" not in graphistry_config:
-            raise InvalidValueError(
-                message="graphistry_config dictionary must contain 'username' and 'password' keys."
-            )
-
-        base_config.graphistry_username = graphistry_config.get("username")
-        base_config.graphistry_password = graphistry_config.get("password")

cognee/api/v1/datasets/routers/get_datasets_router.py

@@ -22,6 +22,7 @@ from cognee.modules.users.permissions.methods import (
 )
 from cognee.modules.graph.methods import get_formatted_graph_data
 from cognee.modules.pipelines.models import PipelineRunStatus
+from cognee.shared.utils import send_telemetry
 
 logger = get_logger()
 
@@ -92,6 +93,14 @@ def get_datasets_router() -> APIRouter:
         ## Error Codes
         - **418 I'm a teapot**: Error retrieving datasets
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "GET /v1/datasets",
+            },
+        )
+
         try:
             datasets = await get_all_user_permission_datasets(user, "read")
 
@@ -130,6 +139,14 @@ def get_datasets_router() -> APIRouter:
         ## Error Codes
         - **418 I'm a teapot**: Error creating dataset
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/datasets",
+            },
+        )
+
         try:
             datasets = await get_datasets_by_name([dataset_data.name], user.id)
 
@@ -175,6 +192,15 @@ def get_datasets_router() -> APIRouter:
         - **404 Not Found**: Dataset doesn't exist or user doesn't have access
         - **500 Internal Server Error**: Error during deletion
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"DELETE /v1/datasets/{str(dataset_id)}",
+                "dataset_id": str(dataset_id),
+            },
+        )
+
         from cognee.modules.data.methods import get_dataset, delete_dataset
 
         dataset = await get_dataset(user.id, dataset_id)
@@ -210,6 +236,16 @@ def get_datasets_router() -> APIRouter:
         - **404 Not Found**: Dataset or data item doesn't exist, or user doesn't have access
         - **500 Internal Server Error**: Error during deletion
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"DELETE /v1/datasets/{str(dataset_id)}/data/{str(data_id)}",
+                "dataset_id": str(dataset_id),
+                "data_id": str(data_id),
+            },
+        )
+
         from cognee.modules.data.methods import get_data, delete_data
         from cognee.modules.data.methods import get_dataset
 
@@ -254,7 +290,7 @@ def get_datasets_router() -> APIRouter:
         if dataset is None:
             raise DatasetNotFoundError(message=f"Dataset ({str(dataset_id)}) not found.")
 
-        graph_data = await get_formatted_graph_data(dataset)
+        graph_data = await get_formatted_graph_data(dataset.id, user.id)
 
         return graph_data
 
@@ -288,6 +324,15 @@ def get_datasets_router() -> APIRouter:
         - **404 Not Found**: Dataset doesn't exist or user doesn't have access
         - **500 Internal Server Error**: Error retrieving data
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"GET /v1/datasets/{str(dataset_id)}/data",
+                "dataset_id": str(dataset_id),
+            },
+        )
+
         from cognee.modules.data.methods import get_dataset_data, get_dataset
 
         # Verify user has permission to read dataset
@@ -308,7 +353,7 @@ def get_datasets_router() -> APIRouter:
 
     @router.get("/status", response_model=dict[str, PipelineRunStatus])
     async def get_dataset_status(
-        datasets: Annotated[List[UUID], Query(alias="dataset")] = None,
+        datasets: Annotated[List[UUID], Query(alias="dataset")] = [],
         user: User = Depends(get_authenticated_user),
     ):
         """
@@ -331,6 +376,15 @@ def get_datasets_router() -> APIRouter:
         ## Error Codes
         - **500 Internal Server Error**: Error retrieving status information
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "GET /v1/datasets/status",
+                "datasets": [str(dataset_id) for dataset_id in datasets],
+            },
+        )
+
         from cognee.api.v1.datasets.datasets import datasets as cognee_datasets
 
         try:
@@ -367,6 +421,16 @@ def get_datasets_router() -> APIRouter:
         - **404 Not Found**: Dataset or data item doesn't exist, or user doesn't have access
         - **500 Internal Server Error**: Error accessing the raw data file
         """
+        send_telemetry(
+            "Datasets API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"GET /v1/datasets/{str(dataset_id)}/data/{str(data_id)}/raw",
+                "dataset_id": str(dataset_id),
+                "data_id": str(data_id),
+            },
+        )
+
         from cognee.modules.data.methods import get_data
         from cognee.modules.data.methods import get_dataset_data
 

cognee/api/v1/delete/routers/get_delete_router.py

@@ -5,6 +5,7 @@ from uuid import UUID
 from cognee.shared.logging_utils import get_logger
 from cognee.modules.users.models import User
 from cognee.modules.users.methods import get_authenticated_user
+from cognee.shared.utils import send_telemetry
 
 logger = get_logger()
 
@@ -31,6 +32,16 @@ def get_delete_router() -> APIRouter:
             JSON response indicating success or failure
 
         """
+        send_telemetry(
+            "Delete API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "DELETE /v1/delete",
+                "dataset_id": str(dataset_id),
+                "data_id": str(data_id),
+            },
+        )
+
         from cognee.api.v1.delete import delete as cognee_delete
 
         try:

cognee/api/v1/permissions/routers/get_permissions_router.py

@@ -6,12 +6,13 @@ from fastapi.responses import JSONResponse
 
 from cognee.modules.users.models import User
 from cognee.modules.users.methods import get_authenticated_user
+from cognee.shared.utils import send_telemetry
 
 
 def get_permissions_router() -> APIRouter:
     permissions_router = APIRouter()
 
-    @permissions_router.post("/datasets/{principal_id}/")
+    @permissions_router.post("/datasets/{principal_id}")
     async def give_datasets_permission_to_principal(
         permission_name: str,
         dataset_ids: List[UUID],
@@ -40,6 +41,16 @@ def get_permissions_router() -> APIRouter:
         - **403 Forbidden**: User doesn't have permission to grant access
         - **500 Internal Server Error**: Error granting permission
         """
+        send_telemetry(
+            "Permissions API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"POST /v1/permissions/datasets/{str(principal_id)}",
+                "dataset_ids": str(dataset_ids),
+                "principal_id": str(principal_id),
+            },
+        )
+
         from cognee.modules.users.permissions.methods import authorized_give_permission_on_datasets
 
         await authorized_give_permission_on_datasets(
@@ -72,6 +83,15 @@ def get_permissions_router() -> APIRouter:
         - **400 Bad Request**: Invalid role name or role already exists
         - **500 Internal Server Error**: Error creating the role
         """
+        send_telemetry(
+            "Permissions API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/permissions/roles",
+                "role_name": role_name,
+            },
+        )
+
         from cognee.modules.users.roles.methods import create_role as create_role_method
 
         await create_role_method(role_name=role_name, owner_id=user.id)
@@ -104,6 +124,16 @@ def get_permissions_router() -> APIRouter:
         - **404 Not Found**: User or role doesn't exist
         - **500 Internal Server Error**: Error adding user to role
         """
+        send_telemetry(
+            "Permissions API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"POST /v1/permissions/users/{str(user_id)}/roles",
+                "user_id": str(user_id),
+                "role_id": str(role_id),
+            },
+        )
+
         from cognee.modules.users.roles.methods import add_user_to_role as add_user_to_role_method
 
         await add_user_to_role_method(user_id=user_id, role_id=role_id, owner_id=user.id)
@@ -136,6 +166,16 @@ def get_permissions_router() -> APIRouter:
         - **404 Not Found**: User or tenant doesn't exist
         - **500 Internal Server Error**: Error adding user to tenant
         """
+        send_telemetry(
+            "Permissions API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": f"POST /v1/permissions/users/{str(user_id)}/tenants",
+                "user_id": str(user_id),
+                "tenant_id": str(tenant_id),
+            },
+        )
+
         from cognee.modules.users.tenants.methods import add_user_to_tenant
 
         await add_user_to_tenant(user_id=user_id, tenant_id=tenant_id, owner_id=user.id)
@@ -161,6 +201,15 @@ def get_permissions_router() -> APIRouter:
         - **400 Bad Request**: Invalid tenant name or tenant already exists
         - **500 Internal Server Error**: Error creating the tenant
         """
+        send_telemetry(
+            "Permissions API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/permissions/tenants",
+                "tenant_name": tenant_name,
+            },
+        )
+
         from cognee.modules.users.tenants.methods import create_tenant as create_tenant_method
 
         await create_tenant_method(tenant_name=tenant_name, user_id=user.id)

cognee/api/v1/search/routers/get_search_router.py

@@ -1,6 +1,7 @@
 from uuid import UUID
 from typing import Optional
 from datetime import datetime
+from pydantic import Field
 from fastapi import Depends, APIRouter
 from fastapi.responses import JSONResponse
 from cognee.modules.search.types import SearchType
@@ -9,16 +10,17 @@ from cognee.modules.users.exceptions.exceptions import PermissionDeniedError
 from cognee.modules.users.models import User
 from cognee.modules.search.operations import get_history
 from cognee.modules.users.methods import get_authenticated_user
+from cognee.shared.utils import send_telemetry
 
 
 # Note: Datasets sent by name will only map to datasets owned by the request sender
 #       To search for datasets not owned by the request sender dataset UUID is needed
 class SearchPayloadDTO(InDTO):
-    search_type: SearchType
-    datasets: Optional[list[str]] = None
-    dataset_ids: Optional[list[UUID]] = None
-    query: str
-    top_k: Optional[int] = 10
+    search_type: SearchType = Field(default=SearchType.GRAPH_COMPLETION)
+    datasets: Optional[list[str]] = Field(default=None)
+    dataset_ids: Optional[list[UUID]] = Field(default=None, examples=[[]])
+    query: str = Field(default="What is in the document?")
+    top_k: Optional[int] = Field(default=10)
 
 
 def get_search_router() -> APIRouter:
@@ -48,6 +50,14 @@ def get_search_router() -> APIRouter:
         ## Error Codes
         - **500 Internal Server Error**: Error retrieving search history
         """
+        send_telemetry(
+            "Search API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "GET /v1/search",
+            },
+        )
+
         try:
             history = await get_history(user.id, limit=0)
 
@@ -83,6 +93,19 @@ def get_search_router() -> APIRouter:
         - To search datasets not owned by the request sender, dataset UUID is needed
         - If permission is denied, returns empty list instead of error
         """
+        send_telemetry(
+            "Search API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "POST /v1/search",
+                "search_type": str(payload.search_type),
+                "datasets": payload.datasets,
+                "dataset_ids": [str(dataset_id) for dataset_id in payload.dataset_ids or []],
+                "query": payload.query,
+                "top_k": payload.top_k,
+            },
+        )
+
         from cognee.api.v1.search import search as cognee_search
 
         try:

cognee/api/v1/users/routers/get_visualize_router.py

@@ -7,6 +7,7 @@ from cognee.modules.data.methods import get_authorized_existing_datasets
 from cognee.modules.users.models import User
 
 from cognee.context_global_variables import set_database_global_context_variables
+from cognee.shared.utils import send_telemetry
 
 logger = get_logger()
 
@@ -39,6 +40,15 @@ def get_visualize_router() -> APIRouter:
         - User must have read permissions on the dataset
         - Visualization is interactive and allows graph exploration
         """
+        send_telemetry(
+            "Visualize API Endpoint Invoked",
+            user.id,
+            additional_properties={
+                "endpoint": "GET /v1/visualize",
+                "dataset_id": str(dataset_id),
+            },
+        )
+
         from cognee.api.v1.visualize import visualize_graph
 
         try:

cognee/base_config.py

@@ -10,8 +10,6 @@ class BaseConfig(BaseSettings):
     data_root_directory: str = get_absolute_path(".data_storage")
     system_root_directory: str = get_absolute_path(".cognee_system")
     monitoring_tool: object = Observer.LANGFUSE
-    graphistry_username: Optional[str] = os.getenv("GRAPHISTRY_USERNAME")
-    graphistry_password: Optional[str] = os.getenv("GRAPHISTRY_PASSWORD")
     langfuse_public_key: Optional[str] = os.getenv("LANGFUSE_PUBLIC_KEY")
     langfuse_secret_key: Optional[str] = os.getenv("LANGFUSE_SECRET_KEY")
     langfuse_host: Optional[str] = os.getenv("LANGFUSE_HOST")

cognee/infrastructure/databases/graph/kuzu/adapter.py

@@ -72,11 +72,36 @@ class KuzuAdapter(GraphDBInterface):
 
                 run_sync(file_storage.ensure_directory_exists())
 
-                self.db = Database(
-                    self.db_path,
-                    buffer_pool_size=256 * 1024 * 1024,  # 256MB buffer pool
-                    max_db_size=1024 * 1024 * 1024,
-                )
+                try:
+                    self.db = Database(
+                        self.db_path,
+                        buffer_pool_size=2048 * 1024 * 1024,  # 2048MB buffer pool
+                        max_db_size=4096 * 1024 * 1024,
+                    )
+                except RuntimeError:
+                    from .kuzu_migrate import read_kuzu_storage_version
+                    import kuzu
+
+                    kuzu_db_version = read_kuzu_storage_version(self.db_path)
+                    if (
+                        kuzu_db_version == "0.9.0" or kuzu_db_version == "0.8.2"
+                    ) and kuzu_db_version != kuzu.__version__:
+                        # Try to migrate kuzu database to latest version
+                        from .kuzu_migrate import kuzu_migration
+
+                        kuzu_migration(
+                            new_db=self.db_path + "_new",
+                            old_db=self.db_path,
+                            new_version=kuzu.__version__,
+                            old_version=kuzu_db_version,
+                            overwrite=True,
+                        )
+
+                    self.db = Database(
+                        self.db_path,
+                        buffer_pool_size=2048 * 1024 * 1024,  # 2048MB buffer pool
+                        max_db_size=4096 * 1024 * 1024,
+                    )
 
             self.db.init_database()
             self.connection = Connection(self.db)
@@ -1438,11 +1463,8 @@ class KuzuAdapter(GraphDBInterface):
         It raises exceptions for failures occurring during deletion processes.
         """
         try:
-            # Use DETACH DELETE to remove both nodes and their relationships in one operation
-            await self.query("MATCH (n:Node) DETACH DELETE n")
-            logger.info("Cleared all data from graph while preserving structure")
-
             if self.connection:
+                self.connection.close()
                 self.connection = None
             if self.db:
                 self.db.close()