molcay commented on PR #37087:
URL: https://github.com/apache/airflow/pull/37087#issuecomment-1916616603
Hi @uranusjr,
I think the use case is a bit different from the datasets. Let's have an
following DAG as an example;
```python3
import random
from airflow import models
from airflow.operators.empty import EmptyOperator
from airflow.operators.python import BranchPythonOperator
from airflow.operators.trigger_dagrun import TriggerDagRunOperator
import pendulum
def check_for_file():
i = random.randint(1, 8)
if i % 2 == 1:
return "do_nothing"
else:
return "trigger_dag_again"
with models.DAG(dag_id="Trigger_dag_test",
start_date=pendulum.datetime(2021, 1, 1, tz="UTC"),
schedule_interval="*/3 * * * *",
catchup=False,
tags=['TriggerDagRunOperator']) as dag:
first_task = EmptyOperator(
task_id="first_task"
)
checkforfile = BranchPythonOperator(task_id='check_for_file',
python_callable=check_for_file)
trigger_dag_again = TriggerDagRunOperator(
task_id="trigger_dag_again",
trigger_dag_id="Trigger_dag_test",
wait_for_completion=False
)
do_nothing = EmptyOperator(
task_id="do_nothing"
)
final = EmptyOperator(
task_id="final"
)
first_task >> checkforfile >> [trigger_dag_again, do_nothing] >> final
```
For this DAG, every time the DAG run in scheduled fashion and select the
trigger_dag_again path, than we see `manual__<date>` as `run_id`, to
distinguish this programmatically triggered DAG runs (using
`TriggerDagRunOperator`) we need to introduce a new run type.
--
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]
