Merge branch 'master' into feature/po-center-ii

Author: odeimaiz

services/storage/src/simcore_service_storage/modules/db/file_meta_data.py

@@ -2,15 +2,15 @@
 import datetime
 from collections.abc import AsyncGenerator
 from pathlib import Path
-from typing import TypeAlias
+from typing import Annotated, TypeAlias
 
 import sqlalchemy as sa
 from models_library.basic_types import SHA256Str
 from models_library.projects import ProjectID
 from models_library.projects_nodes_io import NodeID, SimcoreS3FileID
 from models_library.users import UserID
 from models_library.utils.fastapi_encoders import jsonable_encoder
-from pydantic import BaseModel
+from pydantic import BaseModel, Field, validate_call
 from simcore_postgres_database.storage_models import file_meta_data
 from simcore_postgres_database.utils_repos import (
     pass_or_acquire_connection,
@@ -35,16 +35,15 @@
 
 
 class _PathsCursorParameters(BaseModel):
+    # NOTE: this is a cursor do not put things that can grow unbounded as this goes then through REST APIs or such
     offset: int
     file_prefix: Path | None
-    project_ids: list[ProjectID] | None
     partial: bool
 
 
 def _init_pagination(
     cursor: GenericCursor | None,
     *,
-    filter_by_project_ids: list[ProjectID] | None,
     filter_by_file_prefix: Path | None,
     is_partial_prefix: bool,
 ) -> _PathsCursorParameters:
@@ -53,7 +52,6 @@ def _init_pagination(
     return _PathsCursorParameters(
         offset=0,
         file_prefix=filter_by_file_prefix,
-        project_ids=filter_by_project_ids,
         partial=is_partial_prefix,
     )
 
@@ -229,23 +227,28 @@ async def try_get_directory(
                         return None
         return None
 
+    @validate_call(config={"arbitrary_types_allowed": True})
     async def list_child_paths(
         self,
         *,
         connection: AsyncConnection | None = None,
-        filter_by_project_ids: list[ProjectID] | None,
+        filter_by_project_ids: Annotated[
+            list[ProjectID] | None, Field(max_length=10000)
+        ],
         filter_by_file_prefix: Path | None,
         cursor: GenericCursor | None,
         limit: int,
         is_partial_prefix: bool,
     ) -> tuple[list[PathMetaData], GenericCursor | None, TotalChildren]:
         """returns a list of FileMetaDataAtDB that are one level deep.
         e.g. when no filter is used, these are top level objects
+
+        NOTE: if filter_by_project_ids is huge, this will raise ValidationError and someone needs to fix it!
+        Maybe using a DB join
         """
 
         cursor_params = _init_pagination(
             cursor,
-            filter_by_project_ids=filter_by_project_ids,
             filter_by_file_prefix=filter_by_file_prefix,
             is_partial_prefix=is_partial_prefix,
         )
@@ -278,9 +281,9 @@ async def list_child_paths(
                         file_meta_data.c.file_id.like(search_prefix),
                         (
                             file_meta_data.c.project_id.in_(
-                                [f"{_}" for _ in cursor_params.project_ids]
+                                [f"{_}" for _ in filter_by_project_ids]
                             )
-                            if cursor_params.project_ids
+                            if filter_by_project_ids
                             else True
                         ),
                     )
@@ -303,9 +306,9 @@ async def list_child_paths(
                 )
                 .where(
                     file_meta_data.c.project_id.in_(
-                        [f"{_}" for _ in cursor_params.project_ids]
+                        [f"{_}" for _ in filter_by_project_ids]
                     )
-                    if cursor_params.project_ids
+                    if filter_by_project_ids
                     else True
                 )
                 .cte("ranked_files")

services/storage/src/simcore_service_storage/utils/simcore_s3_dsm_utils.py

@@ -1,3 +1,4 @@
+import logging
 from contextlib import suppress
 from pathlib import Path
 from typing import TypeAlias
@@ -31,6 +32,8 @@
 from ..modules.db.projects import ProjectRepository
 from .utils import convert_db_to_model
 
+_logger = logging.getLogger(__name__)
+
 
 async def _list_all_files_in_folder(
     *,
@@ -250,6 +253,11 @@ async def list_child_paths_from_s3(
     """
     objects_cursor = None
     if cursor is not None:
+        _logger.debug(
+            "Using cursor for listing child paths in S3 for filter '%s': %s",
+            file_filter,
+            cursor,
+        )
         cursor_params = json_loads(cursor)
         assert cursor_params["file_filter"] == f"{file_filter}"  # nosec
         objects_cursor = cursor_params["objects_next_cursor"]
@@ -277,6 +285,11 @@ async def list_child_paths_from_s3(
     ]
     next_cursor = None
     if objects_next_cursor:
+        _logger.debug(
+            "Next cursor for listing child paths in S3 for filter '%s': %s",
+            file_filter,
+            objects_next_cursor,
+        )
         next_cursor = json_dumps(
             {
                 "file_filter": f"{file_filter}",