Nataneljpwd commented on code in PR #62114:
URL: https://github.com/apache/airflow/pull/62114#discussion_r2879622614
##########
airflow-core/src/airflow/jobs/scheduler_job_runner.py:
##########
@@ -3028,44 +3021,41 @@ def _update_asset_orphanage(self, session: Session =
NEW_SESSION) -> None:
)
== 0
).label("orphaned")
- asset_reference_query = session.execute(
- select(orphaned, AssetModel)
+ asset_reference_query = (
+ select(AssetModel)
.outerjoin(DagScheduleAssetReference)
.outerjoin(TaskOutletAssetReference)
.outerjoin(TaskInletAssetReference)
.group_by(AssetModel.id)
- .order_by(orphaned)
)
- asset_orphanation: dict[bool, Collection[AssetModel]] = {
- orphaned: [asset for _, asset in group]
- for orphaned, group in itertools.groupby(asset_reference_query,
key=operator.itemgetter(0))
- }
- self._orphan_unreferenced_assets(asset_orphanation.get(True, ()),
session=session)
- self._activate_referenced_assets(asset_orphanation.get(False, ()),
session=session)
+
+ orphan_query = asset_reference_query.having(orphaned).cte()
+ activate_query = asset_reference_query.having(~orphaned).cte()
+
+ self._orphan_unreferenced_assets(orphan_query, session=session)
+ self._activate_referenced_assets(activate_query, session=session)
@staticmethod
- def _orphan_unreferenced_assets(assets: Collection[AssetModel], *,
session: Session) -> None:
- if assets:
- session.execute(
- delete(AssetActive).where(
- tuple_(AssetActive.name, AssetActive.uri).in_((a.name,
a.uri) for a in assets)
+ def _orphan_unreferenced_assets(assets_query: CTE, *, session: Session) ->
None:
Review Comment:
Simplicity is important, yet I do not see how it is making it less simple by
passing the cte, adding another method increases the mental load, and either
way the method would be called in the same place, yet now, instead of just
seeing the query I have to go to another function
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]
