def test_exec_failure(self, db_mock_class):
"""
Test the execute function in case where the run failed.
"""
run = {
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=run)
db_mock = db_mock_class.return_value
db_mock.submit_run.return_value = 1
db_mock.get_run_state.return_value = RunState('TERMINATED', 'FAILED', '')
with self.assertRaises(AirflowException):
op.execute(None)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID,
})
db_mock_class.assert_called_once_with(
DEFAULT_CONN_ID,
retry_limit=op.databricks_retry_limit,
retry_delay=op.databricks_retry_delay)
db_mock.submit_run.assert_called_once_with(expected)
db_mock.get_run_page_url.assert_called_once_with(RUN_ID)
db_mock.get_run_state.assert_called_once_with(RUN_ID)
self.assertEquals(RUN_ID, op.run_id)
def test_exec_failure(self, db_mock_class):
"""
Test the execute function in case where the run failed.
"""
run = {
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=run)
db_mock = db_mock_class.return_value
db_mock.submit_run.return_value = 1
db_mock.get_run_state.return_value = RunState('TERMINATED', 'FAILED',
'')
with self.assertRaises(AirflowException):
op.execute(None)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID,
})
db_mock_class.assert_called_once_with(
DEFAULT_CONN_ID, retry_limit=op.databricks_retry_limit)
db_mock.submit_run.assert_called_once_with(expected)
db_mock.get_run_page_url.assert_called_once_with(RUN_ID)
db_mock.get_run_state.assert_called_once_with(RUN_ID)
self.assertEquals(RUN_ID, op.run_id)
def test_init_with_named_parameters(self):
"""
Test the initializer with the named parameters.
"""
op = DatabricksSubmitRunOperator(task_id=TASK_ID, new_cluster=NEW_CLUSTER, notebook_task=NOTEBOOK_TASK)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID
})
self.assertDictEqual(expected, op.json)
def test_init_with_named_parameters(self):
"""
Test the initializer with the named parameters.
"""
op = DatabricksSubmitRunOperator(task_id=TASK_ID, new_cluster=NEW_CLUSTER, notebook_task=NOTEBOOK_TASK)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID
})
self.assertDictEqual(expected, op.json)
def test_init_with_json(self):
"""
Test the initializer with json data.
"""
json = {'new_cluster': NEW_CLUSTER, 'notebook_task': NOTEBOOK_TASK}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=json)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID
})
self.assertDictEqual(expected, op.json)
def test_init_with_templating(self):
json = {
'new_cluster': NEW_CLUSTER,
'notebook_task': TEMPLATED_NOTEBOOK_TASK,
}
dag = DAG('test', start_date=datetime.now())
op = DatabricksSubmitRunOperator(dag=dag, task_id=TASK_ID, json=json)
op.json = op.render_template('json', op.json, {'ds': DATE})
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': RENDERED_TEMPLATED_NOTEBOOK_TASK,
'run_name': TASK_ID,
})
self.assertDictEqual(expected, op.json)
def test_init_with_templating(self):
json = {
'new_cluster': NEW_CLUSTER,
'notebook_task': TEMPLATED_NOTEBOOK_TASK,
}
dag = DAG('test', start_date=datetime.now())
op = DatabricksSubmitRunOperator(dag=dag, task_id=TASK_ID, json=json)
op.json = op.render_template('json', op.json, {'ds': DATE})
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': RENDERED_TEMPLATED_NOTEBOOK_TASK,
'run_name': TASK_ID,
})
self.assertDictEqual(expected, op.json)
def test_init_with_json(self):
"""
Test the initializer with json data.
"""
json = {
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK
}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=json)
expected = op._deep_string_coerce({
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID
})
self.assertDictEqual(expected, op.json)
def test_deep_string_coerce(self):
op = DatabricksSubmitRunOperator(task_id='test')
test_json = {
'test_int': 1,
'test_float': 1.0,
'test_dict': {'key': 'value'},
'test_list': [1, 1.0, 'a', 'b'],
'test_tuple': (1, 1.0, 'a', 'b')
}
expected = {
'test_int': '1',
'test_float': '1.0',
'test_dict': {'key': 'value'},
'test_list': ['1', '1.0', 'a', 'b'],
'test_tuple': ['1', '1.0', 'a', 'b']
}
self.assertDictEqual(op._deep_string_coerce(test_json), expected)
def test_deep_string_coerce(self):
op = DatabricksSubmitRunOperator(task_id='test')
test_json = {
'test_int': 1,
'test_float': 1.0,
'test_dict': {'key': 'value'},
'test_list': [1, 1.0, 'a', 'b'],
'test_tuple': (1, 1.0, 'a', 'b')
}
expected = {
'test_int': '1',
'test_float': '1.0',
'test_dict': {'key': 'value'},
'test_list': ['1', '1.0', 'a', 'b'],
'test_tuple': ['1', '1.0', 'a', 'b']
}
self.assertDictEqual(op._deep_string_coerce(test_json), expected)
def test_init_with_merging(self):
"""
Test the initializer when json and other named parameters are both
provided. The named parameters should override top level keys in the
json dict.
"""
override_new_cluster = {'workers': 999}
json = {
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=json, new_cluster=override_new_cluster)
expected = op._deep_string_coerce({
'new_cluster': override_new_cluster,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID,
})
self.assertDictEqual(expected, op.json)
def test_init_with_merging(self):
"""
Test the initializer when json and other named parameters are both
provided. The named parameters should override top level keys in the
json dict.
"""
override_new_cluster = {'workers': 999}
json = {
'new_cluster': NEW_CLUSTER,
'notebook_task': NOTEBOOK_TASK,
}
op = DatabricksSubmitRunOperator(task_id=TASK_ID, json=json, new_cluster=override_new_cluster)
expected = op._deep_string_coerce({
'new_cluster': override_new_cluster,
'notebook_task': NOTEBOOK_TASK,
'run_name': TASK_ID,
})
self.assertDictEqual(expected, op.json)
