Flask-CQLAlchemy handles connections to Cassandra clusters and gives a unified easier way to declare models and their columns.
$ pip install flask-cqlalchemy
As Flask-CQLAlchemy depends only on the cassandra-driver
. It is assumed that you already have Flask
installed.
Flask-CQLAlchemy has been tested with all versions of the cassandra-driver>=3.22.0
and Cassandra 3.0.25, 3.11.11,
4.x. All previous versions and configurations are deprecated. Used to be reported that plugin worked with
cassandra-driver>=2.5
, we can not guarantee proper work of older configurations so use on your own. Some versions of
cassandra-driver
can be incompatible with some versions of Cassandra itself either.
If you have problems using the plugin, try updating to the latest patch version of the minor version you are using.
# example_app.py
import uuid
from flask import Flask
from flask_cqlalchemy import CQLAlchemy
app = Flask(__name__)
app.config['CASSANDRA_HOSTS'] = ['127.0.0.1']
app.config['CASSANDRA_KEYSPACE'] = "cqlengine"
db = CQLAlchemy(app)
class User(db.Model):
uid = db.columns.UUID(primary_key=True, default=uuid.uuid4)
username = db.columns.Text(required=False)
# example_app_udt.py
from flask import Flask
from flask_cqlalchemy import CQLAlchemy
app = Flask(__name__)
app.config['CASSANDRA_HOSTS'] = ['127.0.0.1']
app.config['CASSANDRA_KEYSPACE'] = "cqlengine"
app.config['CASSANDRA_SETUP_KWARGS'] = {'protocol_version': 3}
db = CQLAlchemy(app)
class Address(db.UserType):
street = db.columns.Text()
zipcode = db.columns.Integer()
class Users(db.Model):
__keyspace__ = 'cqlengine'
name = db.columns.Text(primary_key=True)
addr = db.columns.UserDefinedType(Address)
Enter in Python Interpreter:
>>> from example_app import db, User
>>> db.sync_db()
>>> user1 = User.create(username='John Doe')
>>> user1
User(example_id=UUID('f94b6156-2964-4d46-919c-d6e4abcb9ef1'), username='John Doe')
>>> from example_app_udt import db, Address, Users
>>> db.sync_db()
>>> user_address = Address(street="Easy Street, 12", zipcode=12345)
>>> user = Users(name='John Appleseed', addr=user_address)
>>> user
Users(name='John Appleseed', addr=<example_app_udt.Address object at 0x10fe56070>)
>>> user.addr
<example_app_udt.Address object at 0x10fe56070>
>>> user.addr.street
'Easy Street, 12'
>>> user.addr.zipcode
12345
For a complete list of available methods refer to the cassandra.cqlengine.models documentation.
CQLAlchemy
object provides following the options available for the
cqlengine connection.setup()
:
CASSANDRA_HOSTS
— Alist
of hostsCASSANDRA_KEYSPACE
— The default keyspace name to useCASSANDRA_CONSISTENCY
— The global defaultConsistencyLevel
, default is the driver'sSession.default_consistency_level
CASSANDRA_LAZY_CONNECT
—True
if should not connect until first use, default isFalse
CASSANDRA_RETRY_CONNECT
—True
if we should retry to connect even if there was a connection failure initially, default isFalse
CASSANDRA_SETUP_KWARGS
— Pass-through keyword arguments forCluster()
CQLAlchemy
object provides some helper methods for Cassandra database management:
sync_db()
— Creates/Syncs all the tables corresponding to the models declared in the application.set_keyspace()
— Sets the keyspace for a session. Keyspaces once set will remain the default keyspace for the duration of the session. If the change is temporary, it must be reverted back to the default keyspace explicitly.
Found a bug? Need a feature? Open it in issues, or even better, open a PR. Please include tests in the PR.