nathadfield commented on code in PR #67285:
URL: https://github.com/apache/airflow/pull/67285#discussion_r3373690941
##########
airflow-core/docs/migrations-ref.rst:
##########
@@ -39,7 +39,10 @@ Here's the list of all the Database Migrations that are
executed via when you ru
+-------------------------+------------------+-------------------+--------------------------------------------------------------+
| Revision ID | Revises ID | Airflow Version | Description
|
+=========================+==================+===================+==============================================================+
-| ``9ff64e1c35d3`` (head) | ``dd5f3a8e2b91`` | ``3.3.0`` | Add indexes
on dag_run.created_dag_version_id and |
+| ``d2f4e1b3c5a7`` (head) | ``9ff64e1c35d3`` | ``3.3.0`` | Add
partition_date to asset_event and |
Review Comment:
`AssetEvent.partition_date` carries the *source-side* datetime so the
consumer can resolve its own `partition_date` without a second lookup. It's
consumed in `_compute_target_partition_date`: for `IdentityMapper` the
downstream date is the upstream date, read straight off the event. Keeping it
on the event (rather than re-deriving from
`event.source_dag_id`/`source_run_id`) means the value survives source-DagRun
pruning and avoids an extra query per APDR at scheduler-tick time. The
`StartOf*Mapper` family re-derives from the key instead, and rollup/other
mappers leave it `None`. Happy to drop the column and re-derive via the source
DagRun if you'd prefer less denormalization as it's a small change.
--
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]
