Merge pull request #1784 from fractal-analytics-platform/1783-make-da…

…taset-history-validation-accept-workflowtasks-that-ran-with-legacy-tasks

Fix v2.5.0 bug in DatasetRead response validation

CHANGELOG.md

@@ -1,15 +1,18 @@
 **Note**: Numbers like (\#1234) point to closed Pull Requests on the fractal-server repository.
 
 
-# 2.5.1 (Unreleased)
+# 2.5.1
 
 * API:
+    * Make `WorkflowTaskDumpV2` attributes `task_id` and `task` optional (\#1784).
     * Add validation for user-provided strings that execute commands with subprocess or remote-shell (\#1767).
 * Runner and task collection:
     * Validate commands before running them via `subprocess` or `fabric` (\#1767).
 
 # 2.5.0
 
+> WARNING: This release has a minor API bug when displaying a V2 dataset with a history that contains legacy tasks. It's recommended to update to 2.5.1.
+
 This release removes support for including V1 tasks in V2 workflows. This comes
 with changes to the database (data and metadata), to the API, and to the V2
 runner.

fractal_server/app/schemas/v2/dumps.py

@@ -39,14 +39,23 @@ class TaskDumpV2(BaseModel):
 
 
 class WorkflowTaskDumpV2(BaseModel):
+    """
+    Before v2.5.0, WorkflowTaskV2 could have `task_id=task=None` and
+    non-`None` `task_legacy_id` and `task_legacy`. Since these objects
+    may still exist in the database after version updates, we are setting
+    `task_id` and `task` to `Optional` to avoid response-validation errors
+    for the endpoints that GET datasets.
+    Ref issue #1783.
+    """
+
     id: int
     workflow_id: int
     order: Optional[int]
 
     input_filters: Filters
 
-    task_id: int
-    task: TaskDumpV2
+    task_id: Optional[int]
+    task: Optional[TaskDumpV2]
 
 
 class WorkflowDumpV2(BaseModel, extra=Extra.forbid):

tests/v2/03_api/test_unit_issue_1783.py

@@ -0,0 +1,55 @@
+from fractal_server.app.schemas.v2.dataset import _DatasetHistoryItemV2
+from fractal_server.app.schemas.v2.workflowtask import WorkflowTaskStatusTypeV2
+
+
+def test_issue_1783():
+    """
+    Given a dataset that was processed with legacy tasks from within V2
+    workflows, verify that the responds of its GET endpoint is valid.
+
+    The differences
+    """
+
+    wftask1 = dict(
+        id=1,
+        workflow_id=1,
+        order=0,
+        input_filters=dict(attributes=dict(), types=dict()),
+        task_id=1,
+        task=dict(
+            id=1,
+            name="name",
+            type="parallel",
+            source="source",
+            input_types={},
+            output_types={},
+        ),
+    )
+    history_item_1 = dict(
+        workflowtask=wftask1,
+        status=WorkflowTaskStatusTypeV2.FAILED,
+        parallelization=dict(),
+    )
+    _DatasetHistoryItemV2(**history_item_1)
+
+    wftask2 = dict(
+        id=1,
+        workflow_id=1,
+        order=0,
+        input_filters=dict(attributes=dict(), types=dict()),
+        task_legacy_id=1,
+        task_legacy=dict(
+            id=1,
+            input_type="image",
+            output_type="zarr",
+            command="echo",
+            name="name",
+            source="source",
+        ),
+    )
+    history_item_2 = dict(
+        workflowtask=wftask2,
+        status=WorkflowTaskStatusTypeV2.FAILED,
+        parallelization=dict(),
+    )
+    _DatasetHistoryItemV2(**history_item_2)