Adds periodic cleanup task for project documents

Introduces a background job that regularly prunes project documents from Redis,
improving resource management and preventing data accumulation.
Integrates the new cleanup routine into the existing garbage collector plugin.

Author: matusdrobuliak66

services/web/server/src/simcore_service_webserver/garbage_collector/_tasks_documents.py

@@ -0,0 +1,43 @@
+"""
+Scheduled tasks addressing users
+
+"""
+
+import logging
+from collections.abc import AsyncIterator
+from datetime import timedelta
+
+from aiohttp import web
+from servicelib.background_task_utils import exclusive_periodic
+from servicelib.logging_utils import log_context
+
+from ..projects import projects_documents_service
+from ..redis import get_redis_lock_manager_client_sdk
+from ._tasks_utils import CleanupContextFunc, periodic_task_lifespan
+
+_logger = logging.getLogger(__name__)
+
+
+def create_background_task_to_prune_documents(wait_s: float) -> CleanupContextFunc:
+
+    async def _cleanup_ctx_fun(app: web.Application) -> AsyncIterator[None]:
+        interval = timedelta(seconds=wait_s)
+
+        @exclusive_periodic(
+            # Function-exclusiveness is required to avoid multiple tasks like thisone running concurrently
+            get_redis_lock_manager_client_sdk(app),
+            task_interval=interval,
+            retry_after=min(timedelta(seconds=10), interval / 10),
+        )
+        async def _prune_documents_periodically() -> None:
+            with log_context(
+                _logger,
+                logging.INFO,
+                "Deleting project documents in Redis `documents` table started",
+            ):
+                await projects_documents_service.remove_project_documents_as_admin(app)
+
+        async for _ in periodic_task_lifespan(app, _prune_documents_periodically):
+            yield
+
+    return _cleanup_ctx_fun

services/web/server/src/simcore_service_webserver/garbage_collector/plugin.py

@@ -10,7 +10,7 @@
 from ..projects._projects_repository_legacy import setup_projects_db
 from ..redis import setup_redis
 from ..socketio.plugin import setup_socketio
-from . import _tasks_api_keys, _tasks_core, _tasks_trash, _tasks_users
+from . import _tasks_api_keys, _tasks_core, _tasks_documents, _tasks_trash, _tasks_users
 from .settings import get_plugin_settings
 
 _logger = logging.getLogger(__name__)
@@ -66,3 +66,8 @@ def setup_garbage_collector(app: web.Application) -> None:
     app.cleanup_ctx.append(
         _tasks_trash.create_background_task_to_prune_trash(wait_period_s)
     )
+
+    wait_period_s = 600
+    app.cleanup_ctx.append(
+        _tasks_documents.create_background_task_to_prune_documents(wait_period_s)
+    )

services/web/server/src/simcore_service_webserver/projects/_project_document_service.py

@@ -84,3 +84,6 @@ async def _create_project_document_and_increment_version() -> (
         return project_document, document_version
 
     return await _create_project_document_and_increment_version()
+
+
+async def remove_project_documents_as_admin(app: web.Application) -> None: ...

services/web/server/src/simcore_service_webserver/projects/projects_documents_service.py

@@ -0,0 +1,8 @@
+from ._project_document_service import (
+    remove_project_documents_as_admin,
+)
+
+__all__: tuple[str, ...] = ("remove_project_documents_as_admin",)
+
+
+# nopycln: file