DataStax Enterprise Node.js Driver

This driver is built on top of Node.js driver for Apache Cassandra and provides the following additions for DataStax Enterprise:

  • Authenticator implementations that use the authentication scheme negotiation in the server-side DseAuthenticator;
  • encoders for geospatial types which integrate seamlessly with the driver;
  • DSE graph integration.

The DataStax Enterprise Node.js Driver can be used solely with DataStax Enterprise. Please consult the license.


npm install dse-driver

Build Status


Getting Help

You can use the project mailing list or create a ticket on the Jira issue tracker.

Getting Started

Client inherits from the CQL driver counterpart Client.

const dse = require('dse-driver');
const client = new dse.Client({ contactPoints: ['host1', 'host2'] });

const query = 'SELECT name, email FROM users WHERE key = ?';
client.execute(query, [ 'someone' ])
  .then(result => console.log('User with email %s', result.rows[0].email));

Along with the rest of asynchronous execution methods in the driver, execute() returns a Promise that can be chained using then() method. On modern JavaScript engines, promises can be awaited upon using the await keyword within async functions.

Alternatively, you can use the callback-based execution for all asynchronous methods of the API by providing a callback as the last parameter.

client.execute(query, [ 'someone' ], function(err, result) {
  console.log('User with email %s', result.rows[0].email);

In order to have concise code examples in this documentation, we will use the promise-based API of the driver along with the await keyword.

The same submodules structure in the Node.js driver for Apache Cassandra is available in the dse-driver, for example:

const dse = require('dse-driver');
const Uuid = dse.types.Uuid;


For clients connecting to a DSE cluster secured with DseAuthenticator, two authentication providers are included:

  • DsePlainTextAuthProvider: Plain-text authentication;
  • DseGSSAPIAuthProvider: GSSAPI authentication;

To configure a provider, pass it when initializing a cluster:

const dse = require('dse-driver');
const client = new dse.Client({
  contactPoints: ['h1', 'h2'], 
  keyspace: 'ks1',
  authProvider: new dse.auth.DseGssapiAuthProvider()

See the jsdoc of each implementation for more details.


Client includes the executeGraph() method to execute graph queries:

const client = new dse.Client({
  contactPoints: ['host1', 'host2'],
  profiles: [
    new ExecutionProfile('default', {
      graphOptions: { name: 'demo' }
// executeGraph() method returns a Promise
const result = await client.executeGraph('g.V()');
const vertex = result.first();

Graph Options

You can set graph options in execution profiles when initializing Client. Also, to avoid providing the graph name option in each executeGraph() call, you can set the graph options in the default execution profile:

const client = new dse.Client({
  contactPoints: ['host1', 'host2'],
  profiles: [
    new ExecutionProfile('default', {
      graphOptions: { name: 'demo' }
    new ExecutionProfile('demo2-profile', {
      graphOptions: { name: 'demo2' }
// Execute a traversal on the 'demo' graph
const result = await client.executeGraph(query, params);

If needed, you can specify an execution profile different from the default one:

// Execute a traversal on the 'demo2' graph
client.executeGraph(query, params, { executionProfile: 'demo2-profile'});

Additionally, you can also set the default graph options without using execution profiles (not recommended).

const client = new dse.Client({
  contactPoints: ['host1', 'host2'],
  graphOptions: { name: 'demo' }

Handling Results

Graph queries return a GraphResultSet, which is an iterable of rows. The format of the data returned is dependent on the data requested. For example, the payload representing edges will be different than those that represent vertices using the ‘modern’ graph:

// Creating the 'modern' graph
const query =
  'Vertex marko = graph.addVertex(label, "person", "name", "marko", "age", 29);\n' +
  'Vertex vadas = graph.addVertex(label, "person", "name", "vadas", "age", 27);\n' +
  'Vertex lop = graph.addVertex(label, "software", "name", "lop", "lang", "java");\n' +
  'Vertex josh = graph.addVertex(label, "person", "name", "josh", "age", 32);\n' +
  'Vertex ripple = graph.addVertex(label, "software", "name", "ripple", "lang", "java");\n' +
  'Vertex peter = graph.addVertex(label, "person", "name", "peter", "age", 35);\n' +
  'marko.addEdge("knows", vadas, "weight", 0.5f);\n' +
  'marko.addEdge("knows", josh, "weight", 1.0f);\n' +
  'marko.addEdge("created", lop, "weight", 0.4f);\n' +
  'josh.addEdge("created", ripple, "weight", 1.0f);\n' +
  'josh.addEdge("created", lop, "weight", 0.4f);\n' +
  'peter.addEdge("created", lop, "weight", 0.2f);';

await client.executeGraph(query);
// Handling Edges
const result = await client.executeGraph('g.E()');
result.forEach(function (edge) {
  console.log(; // [an internal id representing the edge]
  console.log(edge.type); // edge
  console.log(edge.label); // created
  console.log(; // 0.4
  console.log(edge.outVLabel); // person
  console.log(edge.outV); // [an id representing the outgoing vertex]
  console.log(edge.inVLabel); // software
  console.log(edge.inV); // [an id representing the incoming vertex]
// Using ES6 for...of
const result = await client.executeGraph('g.E()');
for (let edge of result) {
  console.log(edge.label); // created
  // ...
// Handling Vertices
const result = await client.executeGraph('g.V().hasLabel("person")');
result.forEach(function(vertex) {
  console.log(; // [an internal id representing the vertex]
  console.log(vertex.type); // vertex
  console.log(vertex.label); // person
  console.log([0].value); // marko
  console.log([0].value); // 29


Unlike CQL queries which support both positional and named parameters, graph queries only support named parameters. As a result of this, parameters must be passed in as an object:

const query = 'g.addV(label, vertexLabel, "name", username)';
const result = await client.executeGraph(query, { vertexLabel: 'person', username: 'marko' });
const vertex = result.first();
// ...

Parameters are encoded in json, thus will ultimately use their json representation (toJSON if present, otherwise object representation).

You can use results from previous queries as parameters to subsequent queries. For example, if you want to use the id of a vertex returned in a previous query for making a subsequent query:

let result = await client.executeGraph('g.V().hasLabel("person").has("name", "marko")');
const vertex = result.first();
result = await client.executeGraph('g.V(vertexId).out("knows").values("name")', { vertexId: });
const names = result.toArray();
console.log(names); // [ 'vadas', 'josh' ]

Prepared graph statements

Prepared graph statements are not supported by DSE Graph yet (they will be added in the near future).

Geospatial types

DSE 5.0 comes with a set of additional CQL types to represent geospatial data: PointType, LineStringType and PolygonType.

cqlsh> CREATE TABLE points_of_interest(name text PRIMARY KEY, coords 'PointType');
cqlsh> INSERT INTO points_of_interest (name, coords) VALUES ('Eiffel Tower', 'POINT(48.8582 2.2945)');

The DSE driver includes encoders and representations of these types in the geometry module that can be used directly as parameters in queries:

const dse = require('dse-driver');
const Point = dse.geometry.Point;
const insertQuery = 'INSERT INTO points_of_interest (name, coords) VALUES (?, ?)';
const selectQuery = 'SELECT coords FROM points_of_interest WHERE name = ?';

await client.execute(insertQuery, [ 'Eiffel Tower', new Point(48.8582, 2.2945) ], { prepare: true });
const result = await client.execute(selectQuery, ['Eiffel Tower'], { prepare: true });
const row = result.first();
const point = row['coords'];
console.log(point instanceof Point); // true
console.log('x: %d, y: %d', point.x, point.y); // x: 48.8582, y: 2.2945


Copyright 2016-2017 DataStax