Coverage for /pythoncovmergedfiles/medio/medio/usr/local/lib/python3.8/site-packages/airflow/utils/dag_cycle_tester.py: 28%

1# Licensed to the Apache Software Foundation (ASF) under one 

2# or more contributor license agreements. See the NOTICE file 

3# distributed with this work for additional information 

4# regarding copyright ownership. The ASF licenses this file 

5# to you under the Apache License, Version 2.0 (the 

6# "License"); you may not use this file except in compliance 

7# with the License. You may obtain a copy of the License at 

8# 

9# http://www.apache.org/licenses/LICENSE-2.0 

10# 

11# Unless required by applicable law or agreed to in writing, 

12# software distributed under the License is distributed on an 

13# "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY 

14# KIND, either express or implied. See the License for the 

15# specific language governing permissions and limitations 

16# under the License. 

17"""DAG Cycle tester.""" 

18from __future__ import annotations 

19 

20from collections import defaultdict, deque 

21from typing import TYPE_CHECKING, Deque 

22 

23from airflow.exceptions import AirflowDagCycleException, RemovedInAirflow3Warning 

24 

25if TYPE_CHECKING: 

26 from airflow.models.dag import DAG 

27 

28CYCLE_NEW = 0 

29CYCLE_IN_PROGRESS = 1 

30CYCLE_DONE = 2 

31 

32 

33def test_cycle(dag: DAG) -> None: 

34 """ 

35 A wrapper function of `check_cycle` for backward compatibility purpose. 

36 New code should use `check_cycle` instead since this function name `test_cycle` starts with 'test_' and 

37 will be considered as a unit test by pytest, resulting in failure. 

38 """ 

39 from warnings import warn 

40 

41 warn( 

42 "Deprecated, please use `check_cycle` at the same module instead.", 

43 RemovedInAirflow3Warning, 

44 stacklevel=2, 

45 ) 

46 return check_cycle(dag) 

47 

48 

49def check_cycle(dag: DAG) -> None: 

50 """Check to see if there are any cycles in the DAG. 

51 

52 :raises AirflowDagCycleException: If cycle is found in the DAG. 

53 """ 

54 # default of int is 0 which corresponds to CYCLE_NEW 

55 visited: dict[str, int] = defaultdict(int) 

56 path_stack: Deque[str] = deque() 

57 task_dict = dag.task_dict 

58 

59 def _check_adjacent_tasks(task_id, current_task): 

60 """Returns first untraversed child task, else None if all tasks traversed.""" 

61 for adjacent_task in current_task.get_direct_relative_ids(): 

62 if visited[adjacent_task] == CYCLE_IN_PROGRESS: 

63 msg = f"Cycle detected in DAG: {dag.dag_id}. Faulty task: {task_id}" 

64 raise AirflowDagCycleException(msg) 

65 elif visited[adjacent_task] == CYCLE_NEW: 

66 return adjacent_task 

67 return None 

68 

69 for dag_task_id in dag.task_dict.keys(): 

70 if visited[dag_task_id] == CYCLE_DONE: 

71 continue 

72 path_stack.append(dag_task_id) 

73 while path_stack: 

74 current_task_id = path_stack[-1] 

75 if visited[current_task_id] == CYCLE_NEW: 

76 visited[current_task_id] = CYCLE_IN_PROGRESS 

77 task = task_dict[current_task_id] 

78 child_to_check = _check_adjacent_tasks(current_task_id, task) 

79 if not child_to_check: 

80 visited[current_task_id] = CYCLE_DONE 

81 path_stack.pop() 

82 else: 

83 path_stack.append(child_to_check)