def test_run_cli(self, mock_popen, mock_temp_dir):
mock_subprocess = MockSubProcess()
mock_popen.return_value = mock_subprocess
mock_temp_dir.return_value = "test_run_cli"
with mock.patch.dict('os.environ', {
'AIRFLOW_CTX_DAG_ID': 'test_dag_id',
'AIRFLOW_CTX_TASK_ID': 'test_task_id',
'AIRFLOW_CTX_EXECUTION_DATE': '2015-01-01T00:00:00+00:00',
'AIRFLOW_CTX_DAG_RUN_ID': '55',
'AIRFLOW_CTX_DAG_OWNER': 'airflow',
'AIRFLOW_CTX_DAG_EMAIL': '*****@*****.**',
}):
hook = MockHiveCliHook()
hook.run_cli("SHOW DATABASES")
hive_cmd = ['beeline', '-u', '"jdbc:hive2://localhost:10000/default"', '-hiveconf',
'airflow.ctx.dag_id=test_dag_id', '-hiveconf', 'airflow.ctx.task_id=test_task_id',
'-hiveconf', 'airflow.ctx.execution_date=2015-01-01T00:00:00+00:00', '-hiveconf',
'airflow.ctx.dag_run_id=55', '-hiveconf', 'airflow.ctx.dag_owner=airflow',
'-hiveconf', '[email protected]', '-hiveconf',
'mapreduce.job.queuename=airflow', '-hiveconf', 'mapred.job.queue.name=airflow',
'-hiveconf', 'tez.queue.name=airflow', '-f',
'/tmp/airflow_hiveop_test_run_cli/tmptest_run_cli']
mock_popen.assert_called_with(
hive_cmd,
stdout=mock_subprocess.PIPE,
stderr=mock_subprocess.STDOUT,
cwd="/tmp/airflow_hiveop_test_run_cli",
close_fds=True
)
def test_run_cli_with_hive_conf(self, mock_popen):
hql = (
"set key;\n"
"set airflow.ctx.dag_id;\nset airflow.ctx.dag_run_id;\n"
"set airflow.ctx.task_id;\nset airflow.ctx.execution_date;\n"
)
dag_id_ctx_var_name = AIRFLOW_VAR_NAME_FORMAT_MAPPING['AIRFLOW_CONTEXT_DAG_ID']['env_var_format']
task_id_ctx_var_name = AIRFLOW_VAR_NAME_FORMAT_MAPPING['AIRFLOW_CONTEXT_TASK_ID']['env_var_format']
execution_date_ctx_var_name = AIRFLOW_VAR_NAME_FORMAT_MAPPING['AIRFLOW_CONTEXT_EXECUTION_DATE'][
'env_var_format'
]
dag_run_id_ctx_var_name = AIRFLOW_VAR_NAME_FORMAT_MAPPING['AIRFLOW_CONTEXT_DAG_RUN_ID'][
'env_var_format'
]
mock_output = [
'Connecting to jdbc:hive2://localhost:10000/default',
'log4j:WARN No appenders could be found for logger (org.apache.hive.jdbc.Utils).',
'log4j:WARN Please initialize the log4j system properly.',
'log4j:WARN See http://logging.apache.org/log4j/1.2/faq.html#noconfig for more info.',
'Connected to: Apache Hive (version 1.2.1.2.3.2.0-2950)',
'Driver: Hive JDBC (version 1.2.1.spark2)',
'Transaction isolation: TRANSACTION_REPEATABLE_READ',
'0: jdbc:hive2://localhost:10000/default> USE default;',
'No rows affected (0.37 seconds)',
'0: jdbc:hive2://localhost:10000/default> set key;',
'+------------+--+',
'| set |',
'+------------+--+',
'| key=value |',
'+------------+--+',
'1 row selected (0.133 seconds)',
'0: jdbc:hive2://localhost:10000/default> set airflow.ctx.dag_id;',
'+---------------------------------+--+',
'| set |',
'+---------------------------------+--+',
'| airflow.ctx.dag_id=test_dag_id |',
'+---------------------------------+--+',
'1 row selected (0.008 seconds)',
'0: jdbc:hive2://localhost:10000/default> set airflow.ctx.dag_run_id;',
'+-----------------------------------------+--+',
'| set |',
'+-----------------------------------------+--+',
'| airflow.ctx.dag_run_id=test_dag_run_id |',
'+-----------------------------------------+--+',
'1 row selected (0.007 seconds)',
'0: jdbc:hive2://localhost:10000/default> set airflow.ctx.task_id;',
'+-----------------------------------+--+',
'| set |',
'+-----------------------------------+--+',
'| airflow.ctx.task_id=test_task_id |',
'+-----------------------------------+--+',
'1 row selected (0.009 seconds)',
'0: jdbc:hive2://localhost:10000/default> set airflow.ctx.execution_date;',
'+-------------------------------------------------+--+',
'| set |',
'+-------------------------------------------------+--+',
'| airflow.ctx.execution_date=test_execution_date |',
'+-------------------------------------------------+--+',
'1 row selected (0.006 seconds)',
'0: jdbc:hive2://localhost:10000/default> ',
'0: jdbc:hive2://localhost:10000/default> ',
'Closing: 0: jdbc:hive2://localhost:10000/default',
'',
]
with mock.patch.dict(
'os.environ',
{
dag_id_ctx_var_name: 'test_dag_id',
task_id_ctx_var_name: 'test_task_id',
execution_date_ctx_var_name: 'test_execution_date',
dag_run_id_ctx_var_name: 'test_dag_run_id',
},
):
hook = MockHiveCliHook()
mock_popen.return_value = MockSubProcess(output=mock_output)
output = hook.run_cli(hql=hql, hive_conf={'key': 'value'})
process_inputs = " ".join(mock_popen.call_args_list[0][0][0])
self.assertIn('value', process_inputs)
self.assertIn('test_dag_id', process_inputs)
self.assertIn('test_task_id', process_inputs)
self.assertIn('test_execution_date', process_inputs)
self.assertIn('test_dag_run_id', process_inputs)
self.assertIn('value', output)
self.assertIn('test_dag_id', output)
self.assertIn('test_task_id', output)
self.assertIn('test_execution_date', output)
self.assertIn('test_dag_run_id', output)