def run_beam_job(
job_name: Optional[str] = None,
job_class: Optional[Type[base_jobs.JobBase]] = None
) -> beam_job_domain.BeamJobRun:
"""Starts a new Apache Beam job and returns metadata about its execution.
Args:
job_name: str. The name of the job to run. If not provided, then
job_class must not be None.
job_class: type(JobBase). A subclass of JobBase to begin running. This
value takes precedence over job_name.
Returns:
BeamJobRun. Metadata about the run's execution.
"""
if job_class is None and job_name is None:
raise ValueError('Must specify the job class or name to run')
if job_class is None:
# MyPy is wrong. We know job_name is not None in this branch because if
# it were, the ValueError above would have been raised.
job_class = jobs_registry.get_job_class_by_name(job_name) # type: ignore[arg-type]
run_synchronously = constants.EMULATOR_MODE
run_model = jobs_manager.run_job(job_class, run_synchronously)
return get_beam_job_run_from_model(run_model)
def test_failing_sync_job(self) -> None:
run = jobs_manager.run_job(FailingJob, True, namespace=self.namespace)
self.assertEqual(run.latest_job_state, 'FAILED')
run_model = beam_job_models.BeamJobRunModel.get(run.id)
self.assertEqual(run, run_model)
self.assertIn('uh-oh',
beam_job_services.get_beam_job_run_result(run.id).stderr)
def test_working_sync_job(self) -> None:
run = jobs_manager.run_job(WorkingJob, True, namespace=self.namespace)
self.assertEqual(run.latest_job_state, 'DONE')
run_model = beam_job_models.BeamJobRunModel.get(run.id)
self.assertEqual(run, run_model)
self.assertEqual(
beam_job_services.get_beam_job_run_result(run.id).to_dict(), {
'stdout': 'o',
'stderr': 'e'
})
def test_async_job(self) -> None:
mock_run_result = mock.Mock()
mock_run_result.has_job = True
mock_run_result.job_id.return_value = '123'
mock_run_result.state = 'PENDING'
pipeline = beam.Pipeline(
runner=runners.DirectRunner(),
options=job_options.JobOptions(namespace=self.namespace))
with self.swap_to_always_return(pipeline, 'run',
value=mock_run_result):
run = jobs_manager.run_job(WorkingJob, False, pipeline=pipeline)
self.assertEqual(run.dataflow_job_id, '123')
self.assertEqual(run.latest_job_state, 'PENDING')
def test_async_job_that_does_not_start(self) -> None:
mock_run_result = mock.Mock()
mock_run_result.has_job = False
mock_run_result.job_id.return_value = None
mock_run_result.state = 'UNKNOWN'
pipeline = beam.Pipeline(
runner=runners.DirectRunner(),
options=job_options.JobOptions(namespace=self.namespace))
with self.swap_to_always_return(pipeline, 'run',
value=mock_run_result):
run = jobs_manager.run_job(WorkingJob, False, pipeline=pipeline)
self.assertIsNone(run.dataflow_job_id)
self.assertEqual(run.latest_job_state, 'FAILED')
result = beam_job_services.get_beam_job_run_result(run.id)
self.assertIn('Failed to deploy WorkingJob', result.stderr)
