casassg commented on a change in pull request #8962:
URL: https://github.com/apache/airflow/pull/8962#discussion_r437767677
##########
File path: airflow/operators/python.py
##########
@@ -145,6 +149,131 @@ def execute_callable(self):
return self.python_callable(*self.op_args, **self.op_kwargs)
+class _PythonFunctionalOperator(BaseOperator):
+ """
+ Wraps a Python callable and captures args/kwargs when called for execution.
+
+ :param python_callable: A reference to an object that is callable
+ :type python_callable: python callable
+ :param op_kwargs: a dictionary of keyword arguments that will get unpacked
+ in your function
+ :type op_kwargs: dict (templated)
+ :param op_args: a list of positional arguments that will get unpacked when
+ calling your callable
+ :type op_args: list (templated)
+ :param multiple_outputs: if set, function return value will be
+ unrolled to multiple XCom values. Dict will unroll to xcom values with
keys as keys.
+ Defaults to False.
+ :type multiple_outputs: bool
+ """
+
+ template_fields = ('op_args', 'op_kwargs')
+ ui_color = '#ffefeb'
+
+ # since we won't mutate the arguments, we should just do the shallow copy
+ # there are some cases we can't deepcopy the objects(e.g protobuf).
+ shallow_copy_attrs = ('python_callable',)
+
+ @apply_defaults
+ def __init__(
+ self,
+ python_callable: Callable,
+ op_args: Tuple[Any],
+ op_kwargs: Dict[str, Any],
+ multiple_outputs: bool = False,
+ **kwargs
+ ) -> None:
+ kwargs['task_id'] = self._get_unique_task_id(kwargs['task_id'],
kwargs.get('dag', None))
+ super().__init__(**kwargs)
+ self.python_callable = python_callable
+
+ # Check that arguments can be binded
+ signature(python_callable).bind(*op_args, **op_kwargs)
+ self.multiple_outputs = multiple_outputs
+ self.op_args = op_args
+ self.op_kwargs = op_kwargs
+
+ @staticmethod
+ def _get_unique_task_id(task_id: str, dag: Optional[DAG]) -> str:
+ dag = dag or DagContext.get_current_dag()
+ if not dag or task_id not in dag.task_ids:
+ return task_id
+ core = re.split(r'__\d+$', task_id)[0]
+ suffixes = sorted(
+ [int(re.split(r'^.+__', task_id)[1])
+ for task_id in dag.task_ids
+ if re.match(rf'^{core}__\d+$', task_id)]
+ )
+ if not suffixes:
+ return f'{core}__1'
+ return f'{core}__{suffixes[-1] + 1}'
+
+ @staticmethod
+ def validate_python_callable(python_callable):
+ """Validate that python callable can be wrapped by operator.
+ Raises exception if invalid.
+
+ :param python_callable: Python object to be validated
+ :raises: TypeError, AirflowException
+ """
+ if not callable(python_callable):
+ raise TypeError('`python_callable` param must be callable')
+ if 'self' in signature(python_callable).parameters.keys():
+ raise AirflowException('@task does not support methods')
Review comment:
It will fail as `@classmethod` does not return a callable it seems
(surprising to me as well, but it fails on checking if python_callable is a
callable).
##########
File path: airflow/operators/python.py
##########
@@ -145,6 +149,131 @@ def execute_callable(self):
return self.python_callable(*self.op_args, **self.op_kwargs)
+class _PythonFunctionalOperator(BaseOperator):
+ """
+ Wraps a Python callable and captures args/kwargs when called for execution.
+
+ :param python_callable: A reference to an object that is callable
+ :type python_callable: python callable
+ :param op_kwargs: a dictionary of keyword arguments that will get unpacked
+ in your function
+ :type op_kwargs: dict (templated)
+ :param op_args: a list of positional arguments that will get unpacked when
+ calling your callable
+ :type op_args: list (templated)
+ :param multiple_outputs: if set, function return value will be
+ unrolled to multiple XCom values. Dict will unroll to xcom values with
keys as keys.
+ Defaults to False.
+ :type multiple_outputs: bool
+ """
+
+ template_fields = ('op_args', 'op_kwargs')
+ ui_color = '#ffefeb'
+
+ # since we won't mutate the arguments, we should just do the shallow copy
+ # there are some cases we can't deepcopy the objects(e.g protobuf).
+ shallow_copy_attrs = ('python_callable',)
+
+ @apply_defaults
+ def __init__(
+ self,
+ python_callable: Callable,
+ op_args: Tuple[Any],
+ op_kwargs: Dict[str, Any],
+ multiple_outputs: bool = False,
+ **kwargs
+ ) -> None:
+ kwargs['task_id'] = self._get_unique_task_id(kwargs['task_id'],
kwargs.get('dag', None))
+ super().__init__(**kwargs)
Review comment:
Nvm, seems this proposed change fails due to task_id being checked if
repeated on ` super().__init__(**kwargs)`
Resolving.
----------------------------------------------------------------
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]
