Coverage for /pythoncovmergedfiles/medio/medio/src/airflow/airflow/policies.py: 42%

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 

18from __future__ import annotations 

19 

20from typing import TYPE_CHECKING 

21 

22import pluggy 

23 

24local_settings_hookspec = pluggy.HookspecMarker("airflow.policy") 

25hookimpl = pluggy.HookimplMarker("airflow.policy") 

26 

27__all__: list[str] = ["hookimpl"] 

28 

29if TYPE_CHECKING: 

30 from airflow.models.baseoperator import BaseOperator 

31 from airflow.models.dag import DAG 

32 from airflow.models.taskinstance import TaskInstance 

33 

34 

35@local_settings_hookspec 

36def task_policy(task: BaseOperator) -> None: 

37 """ 

38 This policy setting allows altering tasks after they are loaded in the DagBag. 

39 

40 It allows administrator to rewire some task's parameters. Alternatively you can raise 

41 ``AirflowClusterPolicyViolation`` exception to stop DAG from being executed. 

42 

43 Here are a few examples of how this can be useful: 

44 

45 * You could enforce a specific queue (say the ``spark`` queue) for tasks using the ``SparkOperator`` to 

46 make sure that these tasks get wired to the right workers 

47 * You could enforce a task timeout policy, making sure that no tasks run for more than 48 hours 

48 

49 :param task: task to be mutated 

50 """ 

51 

52 

53@local_settings_hookspec 

54def dag_policy(dag: DAG) -> None: 

55 """ 

56 This policy setting allows altering DAGs after they are loaded in the DagBag. 

57 

58 It allows administrator to rewire some DAG's parameters. 

59 Alternatively you can raise ``AirflowClusterPolicyViolation`` exception 

60 to stop DAG from being executed. 

61 

62 Here are a few examples of how this can be useful: 

63 

64 * You could enforce default user for DAGs 

65 * Check if every DAG has configured tags 

66 

67 :param dag: dag to be mutated 

68 """ 

69 

70 

71@local_settings_hookspec 

72def task_instance_mutation_hook(task_instance: TaskInstance) -> None: 

73 """ 

74 This setting allows altering task instances before being queued by the Airflow scheduler. 

75 

76 This could be used, for instance, to modify the task instance during retries. 

77 

78 :param task_instance: task instance to be mutated 

79 """ 

80 

81 

82@local_settings_hookspec 

83def pod_mutation_hook(pod) -> None: 

84 """ 

85 Mutate pod before scheduling. 

86 

87 This setting allows altering ``kubernetes.client.models.V1Pod`` object before they are passed to the 

88 Kubernetes client for scheduling. 

89 

90 This could be used, for instance, to add sidecar or init containers to every worker pod launched by 

91 KubernetesExecutor or KubernetesPodOperator. 

92 """ 

93 

94 

95@local_settings_hookspec(firstresult=True) 

96def get_airflow_context_vars(context) -> dict[str, str]: # type: ignore[empty-body] 

97 """ 

98 This setting allows getting the airflow context vars, which are key value pairs. They are then injected 

99 to default airflow context vars, which in the end are available as environment variables when running 

100 tasks dag_id, task_id, execution_date, dag_run_id, try_number are reserved keys. 

101 

102 :param context: The context for the task_instance of interest. 

103 """ 

104 

105 

106@local_settings_hookspec(firstresult=True) 

107def get_dagbag_import_timeout(dag_file_path: str) -> int | float: # type: ignore[empty-body] 

108 """ 

109 This setting allows for dynamic control of the DAG file parsing timeout based on the DAG file path. 

110 

111 It is useful when there are a few DAG files requiring longer parsing times, while others do not. 

112 You can control them separately instead of having one value for all DAG files. 

113 

114 If the return value is less than or equal to 0, it means no timeout during the DAG parsing. 

115 """ 

116 

117 

118class DefaultPolicy: 

119 """Default implementations of the policy functions. 

120 

121 :meta private: 

122 """ 

123 

124 # Default implementations of the policy functions 

125 

126 @staticmethod 

127 @hookimpl 

128 def get_dagbag_import_timeout(dag_file_path: str): 

129 from airflow.configuration import conf 

130 

131 return conf.getfloat("core", "DAGBAG_IMPORT_TIMEOUT") 

132 

133 @staticmethod 

134 @hookimpl 

135 def get_airflow_context_vars(context): 

136 return {} 

137 

138 

139def make_plugin_from_local_settings(pm: pluggy.PluginManager, module, names: list[str]): 

140 """ 

141 Turn the functions from airflow_local_settings module into a custom/local plugin, so that 

142 plugin-registered functions can co-operate with pluggy/setuptool entrypoint plugins of the same methods. 

143 

144 Airflow local settings will be "win" (i.e. they have the final say) as they are the last plugin 

145 registered. 

146 

147 :meta private: 

148 """ 

149 import inspect 

150 import textwrap 

151 

152 import attr 

153 

154 hook_methods = set() 

155 

156 def _make_shim_fn(name, desired_sig, target): 

157 # Functions defined in airflow_local_settings are called by positional parameters, so the names don't 

158 # have to match what we define in the "template" policy. 

159 # 

160 # However Pluggy validates the names match (and will raise an error if they don't!) 

161 # 

162 # To maintain compat, if we detect the names don't match, we will wrap it with a dynamically created 

163 # shim function that looks somewhat like this: 

164 # 

165 # def dag_policy_name_mismatch_shim(dag): 

166 # airflow_local_settings.dag_policy(dag) 

167 # 

168 codestr = textwrap.dedent( 

169 f""" 

170 def {name}_name_mismatch_shim{str(desired_sig)}: 

171 return __target({' ,'.join(desired_sig.parameters)}) 

172 """ 

173 ) 

174 code = compile(codestr, "<policy-shim>", "single") 

175 scope = {"__target": target} 

176 exec(code, scope, scope) 

177 return scope[f"{name}_name_mismatch_shim"] 

178 

179 @attr.define(frozen=True) 

180 class AirflowLocalSettingsPolicy: 

181 hook_methods: tuple[str, ...] 

182 

183 __name__ = "AirflowLocalSettingsPolicy" 

184 

185 def __dir__(self): 

186 return self.hook_methods 

187 

188 for name in names: 

189 if not hasattr(pm.hook, name): 

190 continue 

191 

192 policy = getattr(module, name) 

193 

194 if not policy: 

195 continue 

196 

197 local_sig = inspect.signature(policy) 

198 policy_sig = inspect.signature(globals()[name]) 

199 # We only care if names/order/number of parameters match, not type hints 

200 if local_sig.parameters.keys() != policy_sig.parameters.keys(): 

201 policy = _make_shim_fn(name, policy_sig, target=policy) 

202 

203 setattr(AirflowLocalSettingsPolicy, name, staticmethod(hookimpl(policy, specname=name))) 

204 hook_methods.add(name) 

205 

206 if hook_methods: 

207 pm.register(AirflowLocalSettingsPolicy(hook_methods=tuple(hook_methods))) 

208 

209 return hook_methods