Node.js Driver for Apache Cassandra

A modern, feature-rich and highly tunable Node.js client library for Apache Cassandra (1.2+) using exclusively Cassandra's binary protocol and Cassandra Query Language v3. Use the DSE Node.js driver for better compatibility and support for DataStax Enterprise.

Installation

1$ npm install cassandra-driver

Features

• Simple, Prepared, and Batch statements
• Asynchronous IO, parallel execution, request pipelining
• Connection pooling
• Auto node discovery
• Automatic reconnection
• Configurable load balancing and retry policies
• Works with any cluster size
• Both promise and callback-based API
• Row streaming and pipes

Documentation

• Documentation index
• CQL types to JavaScript types
• API docs
• FAQ

Getting Help

You can use the project mailing list or create a ticket on the Jira issue tracker. Additionally, you can use the #datastax-drivers channel in the DataStax Academy Slack.

Basic usage

1const cassandra = require('cassandra-driver');
2const client = new cassandra.Client({ contactPoints: ['h1', 'h2'], localDataCenter: 'datacenter1', keyspace: 'ks1' });
3
4const query = 'SELECT name, email FROM users WHERE key = ?';
5client.execute(query, [ 'someone' ])
6  .then(result => console.log('User with email %s', result.rows[0].email));

Alternatively, you can use the callback-based execution for all asynchronous methods of the API.

1client.execute(query, [ 'someone' ], function(err, result) {
2  assert.ifError(err);
3  console.log('User with email %s', result.rows[0].email);
4});

Prepare your queries

Using prepared statements provides multiple benefits. Prepared statements are parsed and prepared on the Cassandra nodes and are ready for future execution. Also, when preparing, the driver retrieves information about the parameter types which allows an accurate mapping between a JavaScript type and a Cassandra type.

The driver will prepare the query once on each host and execute the statement with the bound parameters.

1// Use query markers (?) and parameters
2const query = 'UPDATE users SET birth = ? WHERE key=?'; 
3const params = [ new Date(1942, 10, 1), 'jimi-hendrix' ];
4// Set the prepare flag in the query options
5client.execute(query, params, { prepare: true })
6  .then(result => console.log('Row updated on the cluster'));

Row streaming and pipes

When using #eachRow() and #stream() methods, the driver parses each row as soon as it is received, yielding rows without buffering them.

1// Reducing a large result
2client.eachRow('SELECT time, val FROM temperature WHERE station_id=', ['abc'],
3  function(n, row) {
4    // The callback will be invoked per each row as soon as they are received
5    minTemperature = Math.min(row.val, minTemperature);
6  },
7  function (err) {
8    assert.ifError(err);
9  }
10);

The #stream() method works in the same way but instead of callback it returns a Readable Streams2 object in objectMode that emits instances of Row. It can be piped downstream and provides automatic pause/resume logic (it buffers when not read).

1client.stream('SELECT time, val FROM temperature WHERE station_id=', [ 'abc' ])
2  .on('readable', function () {
3    // 'readable' is emitted as soon a row is received and parsed
4    let row;
5    while (row = this.read()) {
6      console.log('time %s and value %s', row.time, row.val);
7    }
8  })
9  .on('end', function () {
10    // Stream ended, there aren't any more rows
11  })
12  .on('error', function (err) {
13    // Something went wrong: err is a response error from Cassandra
14  });

User defined types

User defined types (UDT) are represented as JavaScript objects.

For example: Consider the following UDT and table

1CREATE TYPE address (
2  street text,
3  city text,
4  state text,
5  zip int,
6  phones set<text>
7);
8CREATE TABLE users (
9  name text PRIMARY KEY,
10  email text,
11  address frozen<address>
12);

You can retrieve the user address details as a regular JavaScript object.

1const query = 'SELECT name, address FROM users WHERE key = ?';
2client.execute(query, [ key ], { prepare: true })
3  .then(result => {
4    const row = result.first();
5    const address = row.address;
6    console.log('User lives in %s, %s - %s', address.street, address.city, address.state); 
7  });

Read more information about using UDTs with the Node.js Driver.

Paging

All driver methods use a default fetchSize of 5000 rows, retrieving only first page of results up to a maximum of 5000 rows to shield an application against accidentally large result sets. To retrieve the following records you can use the autoPage flag in the query options of #eachRow() and #stream() methods.

1//Imagine a column family with millions of rows
2const query = 'SELECT * FROM largetable';
3client.eachRow(query, [], { autoPage: true }, function (n, row) {
4  // This function will be invoked per each of the rows in all the table
5}, endCallback);

Batch multiple statements

You can execute multiple statements in a batch to update/insert several rows atomically even in different column families.

1const queries = [
2  {
3    query: 'UPDATE user_profiles SET email=? WHERE key=?',
4    params: [ emailAddress, 'hendrix' ]
5  },
6  {
7    query: 'INSERT INTO user_track (key, text, date) VALUES (?, ?, ?)',
8    params: [ 'hendrix', 'Changed email', new Date() ]
9  }
10];
11client.batch(queries, { prepare: true })
12  .then(result => console.log('Data updated on cluster'));

---

Data types

There are few data types defined in the ECMAScript specification, this usually represents a problem when you are trying to deal with data types that come from other systems in JavaScript.

The driver supports all the CQL data types in Apache Cassandra (3.0 and below) even for types with no built-in JavaScript representation, like decimal, varint and bigint. Check the documentation on working with numerical values, uuids and collections.

Logging

Instances of Client() are EventEmitter and emit log events:

1client.on('log', function(level, className, message, furtherInfo) {
2  console.log('log event: %s -- %s', level, message);
3});

The level being passed to the listener can be verbose, info, warning or error.

Compatibility

• Apache Cassandra versions 2.0 and above.
• DataStax Enterprise versions 4.5 and above.
• Node.js versions 4 and above.

Note: DataStax products do not support big-endian systems.

Feedback Requested

Help us focus our efforts! Provide your input on the Platform and Runtime Survey (we kept it short).

Credits

This driver is based on the original work of Jorge Bay on node-cassandra-cql and adds a series of advanced features that are common across all other DataStax drivers for Apache Cassandra.

The development effort to provide an up to date, high performance, fully featured Node.js Driver for Apache Cassandra will continue on this project, while node-cassandra-cql will be discontinued.

License

Copyright 2018, YugaByte, Inc.

Licensed under the Apache License, Version 2.0 (the "License"); you may not use this file except in compliance with the License. You may obtain a copy of the License at

http://www.apache.org/licenses/LICENSE-2.0

Unless required by applicable law or agreed to in writing, software distributed under the License is distributed on an "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. See the License for the specific language governing permissions and limitations under the License.