feat: Add data deletion endpoint

Add ability to delete data instance through endpoint

Feature #COG-455

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

@@ -52,6 +52,31 @@ def get_datasets_router():
 
         await delete_dataset(dataset)
 
+    @router.delete("/{dataset_id}/data/{data_id}", response_model=None, responses={404: {"model": ErrorResponseDTO}})
+    async def delete_data(dataset_id: str, data_id: str, user: User = Depends(get_authenticated_user)):
+        from cognee.modules.data.methods import get_data, delete_data
+        from cognee.modules.data.methods import get_dataset
+
+        # Check if user has permission to access dataset and data by trying to get the dataset
+        dataset = await get_dataset(user.id, dataset_id)
+
+        #TODO: Handle situation differently if user doesn't have permission to access data?
+        if dataset is None:
+            raise HTTPException(
+                status_code=404,
+                detail=f"Dataset ({dataset_id}) not found."
+            )
+
+        data = await get_data(user.id, data_id)
+
+        if data is None:
+            raise HTTPException(
+                status_code=404,
+                detail=f"Dataset ({data_id}) not found."
+            )
+
+        await delete_data(data)
+
     @router.get("/{dataset_id}/graph", response_model=str)
     async def get_dataset_graph(dataset_id: str, user: User = Depends(get_authenticated_user)):
         from cognee.shared.utils import render_graph

cognee/modules/data/methods/__init__.py

@@ -6,6 +6,8 @@ from .get_dataset import get_dataset
 from .get_datasets import get_datasets
 from .get_datasets_by_name import get_datasets_by_name
 from .get_dataset_data import get_dataset_data
+from .get_data import get_data
 
 # Delete
 from .delete_dataset import delete_dataset
+from .delete_data import delete_data

cognee/modules/data/methods/delete_data.py

@@ -0,0 +1,8 @@
+from cognee.modules.data.models import Data
+from cognee.infrastructure.databases.relational import get_relational_engine
+
+
+async def delete_data(data: Data):
+    db_engine = get_relational_engine()
+
+    return await db_engine.delete_data_by_id(data.__tablename__, data.id)

cognee/modules/data/methods/delete_dataset.py

@@ -4,4 +4,4 @@ from cognee.infrastructure.databases.relational import get_relational_engine
 async def delete_dataset(dataset: Dataset):
     db_engine = get_relational_engine()
 
-    return await db_engine.delete_table(dataset.id)
+    return await db_engine.delete_data_by_id(dataset.__tablename__, dataset.id)

cognee/modules/data/methods/get_data.py

@@ -0,0 +1,12 @@
+from uuid import UUID
+from typing import Optional
+from cognee.infrastructure.databases.relational import get_relational_engine
+from ..models import Data
+
+async def get_data(user_id: UUID, data_id: UUID) -> Optional[Data]:
+    db_engine = get_relational_engine()
+
+    async with db_engine.get_async_session() as session:
+        data = await session.get(Data, data_id)
+
+        return data