在TriggerDagRunOp中提供上下文

Question

我有一个由另一个dag触发的dag。我已经通过DagRunOrder().payload字典向这个dag传递了一些配置变量，就像official example那样。

现在在这个dag中，我有另一个TriggerDagRunOperator来启动第二个dag，并希望通过这些相同的配置变量。

我成功地访问了PythonOperator中的负载变量，如下所示：

def run_this_func(ds, **kwargs):
    print("Remotely received value of {} for message and {} for day".format(
        kwargs["dag_run"].conf["message"], kwargs["dag_run"].conf["day"])
    )

run_this = PythonOperator(
    task_id='run_this',
    provide_context=True,
    python_callable=run_this_func,
    dag=dag
)

但同样的模式在TriggerDagRunOperator中不起作用：

def trigger(context, dag_run_obj, **kwargs):
    dag_run_obj.payload = {
        "message": kwargs["dag_run"].conf["message"],
        "day": kwargs["dag_run"].conf["day"]
    }
    return dag_run_obj

trigger_step = TriggerDagRunOperator(
    task_id="trigger_modelling",
    trigger_dag_id="Dummy_Modelling",
    provide_context=True,
    python_callable=trigger,
    dag=dag
)

它会生成有关使用provide_context的警告：

INFO - Subtask: /usr/local/lib/python2.7/dist-packages/airflow/models.py:1927: PendingDeprecationWarning: Invalid arguments were passed to TriggerDagRunOperator. Support for passing such arguments will be dropped in Airflow 2.0. Invalid arguments were:
INFO - Subtask: *args: ()
INFO - Subtask: **kwargs: {'provide_context': True}
INFO - Subtask:   category=PendingDeprecationWarning

这个错误表明我没有通过conf：

INFO - Subtask: Traceback (most recent call last):
INFO - Subtask:   File "/usr/local/lib/python2.7/dist-packages/airflow/models.py", line 1374, in run
INFO - Subtask:     result = task_copy.execute(context=context)
INFO - Subtask:   File "/usr/local/lib/python2.7/dist-packages/airflow/operators/dagrun_operator.py", line 64, in execute
INFO - Subtask:     dro = self.python_callable(context, dro)
INFO - Subtask:   File "/home/user/airflow/dags/dummy_responses.py", line 28, in trigger
INFO - Subtask:     "message": kwargs["dag_run"].conf["message"],
INFO - Subtask: KeyError: 'dag_run'

---

我尝试过的第二种模式也没有成功，它使用了params参数，如下所示：

def trigger(context, dag_run_obj):
    dag_run_obj.payload = {
        "message": context['params']['message'],
        "day": context['params']['day']
    }
    return dag_run_obj

trigger_step = TriggerDagRunOperator(
    task_id="trigger_modelling",
    trigger_dag_id="Dummy_Modelling",
    python_callable=trigger,
    params={
        "message": "{{ dag_run.conf['message'] }}",
        "day": "{{ dag_run.conf['day'] }}"
    },
    dag=dag
)

此模式不产生错误，而是将参数作为字符串传递给下一个dag，即它不计算表达式。

---

如何访问第二个dag的TriggerDagRunOperator中的配置变量？