def test_process_file_successful(patch_storage, patch_rabbit, tmp_path):
schema = {'header_1': [], 'header_2': []}
header = ','.join(key for key in schema.keys())
mock_processor = setup_mock_processor(schema, None)
mock_processor.build_event_messages.side_effect = lambda row: [row]
bulk_processor = BulkProcessor(mock_processor)
bulk_processor.working_dir = tmp_path
bulk_processor.rabbit = patch_rabbit
test_file = RESOURCE_PATH.joinpath('bulk_test_file_success.csv')
success_file, error_file, error_detail_file = bulk_processor.initialise_results_files(
test_file.name)
success_count, failure_count = bulk_processor.process_file(
test_file, success_file, error_file, error_detail_file)
assert not failure_count, 'Should have no processing errors'
assert success_count == 1, 'Should successfully process one row'
assert success_file.read_text() == test_file.read_text()
assert error_file.read_text() == header + '\n'
assert error_detail_file.read_text() == HEADER_IS_VALID
patch_rabbit.publish_message.assert_called_once_with(
message=json.dumps({
'header_1': 'foo',
'header_2': 'bar'
}),
content_type='application/json',
headers=None,
exchange=mock_processor.exchange,
routing_key=mock_processor.routing_key)
def test_process_file_encoding_failure(patch_storage, patch_rabbit, tmp_path):
schema = {'header_1': [], 'header_2': []}
header = ','.join(key for key in schema.keys())
mock_processor = setup_mock_processor(schema, None)
bulk_processor = BulkProcessor(mock_processor)
bulk_processor.working_dir = tmp_path
test_file = RESOURCE_PATH.joinpath('bulk_test_file_encoding_failure.csv')
success_file, error_file, error_detail_file = bulk_processor.initialise_results_files(
test_file.name)
success_count, failure_count = bulk_processor.process_file(
test_file, success_file, error_file, error_detail_file)
assert failure_count == 1, 'Should have one failure when it tries to decode the file'
assert not success_count, 'Should not successfully process any rows'
assert success_file.read_text() == header + '\n'
assert 'Invalid file encoding, requires utf-8' in error_detail_file.read_text(
)
patch_rabbit.publish_message.assert_not_called()
def test_process_file_success_failure_mix(patch_storage, patch_rabbit,
tmp_path):
error_message_description = 'tests value invalid failure message'
error_detail_message = '[Column: header_1, Error: tests value invalid failure message]'
schema = {
'header_1': [no_invalid_validator(message=error_message_description)],
'header_2': []
}
header = ','.join(key for key in schema.keys())
mock_processor = setup_mock_processor(schema, None)
mock_processor.build_event_messages.side_effect = lambda row: [row]
bulk_processor = BulkProcessor(mock_processor)
bulk_processor.working_dir = tmp_path
bulk_processor.rabbit = patch_rabbit
test_file = RESOURCE_PATH.joinpath(
'bulk_test_file_success_failure_mix.csv')
success_file, error_file, error_detail_file = bulk_processor.initialise_results_files(
test_file.name)
success_count, failure_count = bulk_processor.process_file(
test_file, success_file, error_file, error_detail_file)
assert failure_count == 1, 'Should fail to process one row'
assert success_count == 1, 'Should successfully process one row'
assert success_file.read_text() == header + '\n' + 'foo,bar' + '\n'
assert error_file.read_text() == header + '\n' + 'invalid,bar' + '\n'
assert error_detail_file.read_text(
) == HEADER_IS_VALID + error_detail_message + '\n'
patch_rabbit.publish_message.assert_called_once_with(
message=json.dumps({
'header_1': 'foo',
'header_2': 'bar'
}),
content_type='application/json',
headers=None,
exchange=mock_processor.exchange,
routing_key=mock_processor.routing_key)