def test_file_splitting(self, gcs_hook_mock_class, mssql_hook_mock_class):
"""Test that ndjson is split by approx_max_file_size_bytes param."""
mssql_hook_mock = mssql_hook_mock_class.return_value
mssql_hook_mock.get_conn().cursor().__iter__.return_value = iter(ROWS)
mssql_hook_mock.get_conn().cursor().description = CURSOR_DESCRIPTION
gcs_hook_mock = gcs_hook_mock_class.return_value
expected_upload = {
JSON_FILENAME.format(0): b''.join(NDJSON_LINES[:2]),
JSON_FILENAME.format(1): NDJSON_LINES[2],
}
def _assert_upload(bucket, obj, tmp_filename, mime_type=None, gzip=False):
assert BUCKET == bucket
assert 'application/json' == mime_type
assert GZIP == gzip
with open(tmp_filename, 'rb') as file:
assert expected_upload[obj] == file.read()
gcs_hook_mock.upload.side_effect = _assert_upload
op = MSSQLToGCSOperator(
task_id=TASK_ID,
sql=SQL,
bucket=BUCKET,
filename=JSON_FILENAME,
approx_max_file_size_bytes=len(expected_upload[JSON_FILENAME.format(0)]),
)
op.execute(None)
def test_exec_success_json(self, gcs_hook_mock_class, mssql_hook_mock_class):
"""Test successful run of execute function for JSON"""
op = MSSQLToGCSOperator(
task_id=TASK_ID, mssql_conn_id=MSSQL_CONN_ID, sql=SQL, bucket=BUCKET, filename=JSON_FILENAME
)
mssql_hook_mock = mssql_hook_mock_class.return_value
mssql_hook_mock.get_conn().cursor().__iter__.return_value = iter(ROWS)
mssql_hook_mock.get_conn().cursor().description = CURSOR_DESCRIPTION
gcs_hook_mock = gcs_hook_mock_class.return_value
def _assert_upload(bucket, obj, tmp_filename, mime_type=None, gzip=False):
assert BUCKET == bucket
assert JSON_FILENAME.format(0) == obj
assert 'application/json' == mime_type
assert GZIP == gzip
with open(tmp_filename, 'rb') as file:
assert b''.join(NDJSON_LINES) == file.read()
gcs_hook_mock.upload.side_effect = _assert_upload
op.execute(None)
mssql_hook_mock_class.assert_called_once_with(mssql_conn_id=MSSQL_CONN_ID)
mssql_hook_mock.get_conn().cursor().execute.assert_called_once_with(SQL)
def test_schema_file(self, gcs_hook_mock_class, mssql_hook_mock_class):
"""Test writing schema files."""
mssql_hook_mock = mssql_hook_mock_class.return_value
mssql_hook_mock.get_conn().cursor().__iter__.return_value = iter(ROWS)
mssql_hook_mock.get_conn().cursor().description = CURSOR_DESCRIPTION
gcs_hook_mock = gcs_hook_mock_class.return_value
def _assert_upload(bucket, obj, tmp_filename, mime_type, gzip): # pylint: disable=unused-argument
if obj == SCHEMA_FILENAME:
with open(tmp_filename, 'rb') as file:
self.assertEqual(b''.join(SCHEMA_JSON), file.read())
gcs_hook_mock.upload.side_effect = _assert_upload
op = MSSQLToGCSOperator(
task_id=TASK_ID,
sql=SQL,
bucket=BUCKET,
filename=JSON_FILENAME,
schema_filename=SCHEMA_FILENAME)
op.execute(None)
# once for the file and once for the schema
self.assertEqual(2, gcs_hook_mock.upload.call_count)
def test_init(self):
"""Test MySqlToGoogleCloudStorageOperator instance is properly initialized."""
op = MSSQLToGCSOperator(task_id=TASK_ID, sql=SQL, bucket=BUCKET, filename=JSON_FILENAME)
assert op.task_id == TASK_ID
assert op.sql == SQL
assert op.bucket == BUCKET
assert op.filename == JSON_FILENAME
def test_init(self):
"""Test MySqlToGoogleCloudStorageOperator instance is properly initialized."""
op = MSSQLToGCSOperator(task_id=TASK_ID, sql=SQL, bucket=BUCKET, filename=JSON_FILENAME)
self.assertEqual(op.task_id, TASK_ID)
self.assertEqual(op.sql, SQL)
self.assertEqual(op.bucket, BUCKET)
self.assertEqual(op.filename, JSON_FILENAME)
# specific language governing permissions and limitations
# under the License.
import os
from datetime import datetime
from airflow import models
from airflow.providers.google.cloud.transfers.mssql_to_gcs import MSSQLToGCSOperator
GCS_BUCKET = os.environ.get("GCP_GCS_BUCKET", "example-airflow")
FILENAME = 'test_file'
SQL_QUERY = "USE airflow SELECT * FROM Country;"
with models.DAG(
'example_mssql_to_gcs',
schedule_interval='@once',
start_date=datetime(2021, 12, 1),
catchup=False,
tags=['example'],
) as dag:
# [START howto_operator_mssql_to_gcs]
upload = MSSQLToGCSOperator(
task_id='mssql_to_gcs',
mssql_conn_id='airflow_mssql',
sql=SQL_QUERY,
bucket=GCS_BUCKET,
filename=FILENAME,
export_format='csv',
)
# [END howto_operator_mssql_to_gcs]
