XD-DENG commented on a change in pull request #12574:
URL: https://github.com/apache/airflow/pull/12574#discussion_r528974178
##########
File path: airflow/sensors/external_task_sensor.py
##########
@@ -149,38 +148,22 @@ def poke(self, context, session=None):
dttm = context['execution_date']
dttm_filter = dttm if isinstance(dttm, list) else [dttm]
Review comment:
The use case I can think of to justify this setting-up/design is that a
user is checking & depending a series of external DAG/Task. For example, I
want to depend on DAG A, and what I want to ensure is DAG A has succeeded in
the past a few consecutive days (or maybe more complex case, not consecutive).
Using Multiple Sensors may multiply the resource usage and may make the DAG
authoring more complex in this scenario?
From another perspective, if someone prefers to have multiple sensors each
checking single exec date, the current design can meet his/her requirement as
well.
So personally this doesn't bother me (but as you mentioned, not giving
information about how many TI failed or failed for which dates is not nice
enough)
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
[email protected]
