ms3 is a mock server intended to replace the dependency on AWS S3 when testing a project.
Current features are:
- create/delete/list buckets
- enable/disable versioning in buckets
- get/put objects (with versions)
For more up to date features, please look at tests/test_s3_operations.py.
In order to get a virtualenv with all dependencies installed just run ./setup-development.sh
Setting up the connection (assuming a global level function that provides an S3 connection):
from boto.s3.connection import S3Connection, OrdinaryCallingFormat
from tornado.options import options
def current_s3_connection():
""" Return an AWS S3 connection """
if getattr(current_s3_connection, '_connection', None) is None:
if options.use_ms3:
_connection = S3Connection('X', 'Y',
is_secure=options.ms3_https,
port=options.ms3_port, host=options.ms3_host,
calling_format=OrdinaryCallingFormat())
else:
_connection = S3Connection(options.aws_access_key,
options.aws_secret_key)
current_s3_connection._connection = _connection
return current_s3_connection._connectionStarting with a test case:
from ms3.testing import MS3Server
class ExampleTestCase(TestCase):
def setUp(self):
self.datadir = helpers.tests_dir("s3")
MS3Server.start(self.datadir, port=9011, with_exec=True)
def tearDown(self):
MS3Server.stop()
def test_example():
s3_conn = current_s3_connection()
buckets = s3_conn.get_all_buckets()
...In order to get a ms3 server up and running (for development purposes), run python -m ms3.app.
For running the test: python tests/test_s3_operations.py
You can find out more details regarding the configuration options by typing: python -m ms3.app --help
ms3 is released under MIT licence (see LICENSE file).