Initialize the JavaScript (Node) SDK

Use the createInstance method to initialize the JavaScript (Node) SDK and instantiate an instance of the Optimizely client class that exposes API methods like the Decide methods. Each client corresponds to the datafile representing the state of a project for a certain environment.

Version

SDK v5.0.0 and higher

Description

The createInstance method accepts a configuration object to configure Optimizely Feature Experimentation.

Some parameters are optional because the SDK provides a default implementation, but you may want to override these for your production environments. For example, you may want override these to set up an error handler and logger to catch issues, an event dispatcher to manage network calls, and a User Profile Service to ensure sticky bucketing.

Parameters

The table below lists the required and optional properties of the config object:

Returns

Instantiates an instance of the Optimizely Feature Experimentation class.

Examples

In the JavaScript (Node) SDK, you can provide a sdkKey or datafile or both.

• When initializing with just the SDK key – The SDK will poll for datafile changes in the background at regular intervals.
• When initializing with just the datafile – The SDK will NOT poll for datafile changes in the background.
• When initializing with both the SDK key and datafile – The SDK will use the given datafile and start polling for datafile changes in the background.

Instantiate using SDK Key

In the JavaScript (Node) SDK, you only need to pass the SDK key value to instantiate a client. Whenever the experiment configuration changes, the SDK handles the change for you.

Include sdkKey as a string property in the options object you pass to the createInstance method.

const { createInstance } = require('@optimizely/optimizely-sdk');

const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>', // Provide the sdkKey of your desired environment here
});

When you provide the sdkKey, the SDK instance downloads the datafile associated with that sdkKey. When the download completes, the SDK instance updates itself to use the downloaded datafile. You can use the onReady method to wait for the datafile to be downloaded before using the instance.

const { createInstance } = require('@optimizely/optimizely-sdk');
const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>', // Provide the sdkKey of your desired environment here
});

if (!optimizely) {
  // there was an error creating the instance, handle error
} else {
  // Use optimizelyClient
  optimizely.onReady().then(({ success, reason }) => {
    if (success) {
      // optimizelyClientInstance is ready to use, with datafile downloaded from the
      // Optimizely CDN
    } else {
      // this optimizely client instance cannot be used
      console.log(`client initialization unsuccessful, reason: ${reason}`);
    }
  });
}

Instantiate using datafile

To instantiate using datafile, first you must get a copy of the datafile from our server. In the example below, we demonstrate fetching the datafile using the node-fetch@2 library.

// Minimal client
const { createInstance } = require('@optimizely/optimizely-sdk');
const fetch = require('node-fetch');

// Replace <YOUR_SDK_KEY> to get the datafile
const sdkKey = '<YOUR_SDK_KEY>';
const DATAFILE_URL = `https://cdn.optimizely.com/datafiles/${sdkKey}.json`;
const response = await fetch(DATAFILE_URL);
const datafile = await response.json();

const optimizely = createInstance({
  datafile
});

if (!optimizely) {
  // there was an error creating the instance, handle error
} else {
  // Use optimizelyClient to run experiments
}

If you do not pass in an SDK key, the Optimizely Client will not automatically sync newer datafile versions. Any time you retrieve an updated datafile, just re-instantiate the same client.

For simple applications, all you need to provide to instantiate a client is a datafile specifying the project configuration for a given environment. For most advanced implementations, you'll want to customize the logger or error handler for your specific requirements.

Notes

Customize datafile management behavior

To customize datafile management behavior, provide a datafileOptions object property inside the options object passed to createInstance. The table lists the supported customizable options.

The following example shows how to customize datafile management behavior:

const { createInstance } = require('@optimizely/optimizely-sdk');

const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>',
  datafileOptions: {
    autoUpdate: true,
    updateInterval: 600000, // 10 minutes in milliseconds
    urlTemplate: 'http://localhost:5000/datafiles/%s.json',
  },
});

onReady details

Use the onReady method to wait until the download is complete and the SDK is ready to use.

The onReady method returns a Promise representing the initialization process.
onReady accepts an optional timeout argument (defined in milliseconds) that controls the maximum duration that the returned Promise will remain in the pending state. If timeout is not provided, it defaults to 30 seconds.

const { createInstance } = require('@optimizely/optimizely-sdk');
const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>',
});

if (!optimizely) {
  // there was an error creating the instance, handle error
} else {
  // Use optimizelyClient
  optimizely.onReady().then(result => {
    // Returned Promise is fulfilled with a result object
    console.log(result.success); // true if the instance fetched a datafile and is now ready to use
    console.log(result.reason); // If success is false, reason contains an error message
  });

  // Provide a timeout in milliseconds - promise will resolve if the datafile still is not available after the timeout
  optimizely.onReady({ timeout: 5000 }).then(result => {
    // Returned Promise is fulfilled with a result object
    console.log(result.success); // true if the instance fetched a datafile and is now ready to use
    console.log(result.reason); // If success is false, reason contains an error message
  });
};

The Promise returned from the onReady method is fulfilled with a result object containing a boolean success property.

When the success property is true, the instance is ready to use with a valid datafile. When the success property is false, there is also a reasonstring property describing the failure. Failure can be caused by expiration of the timeout, network error, unsuccessful HTTP response, datafile validation error, or the instance'sclose method being called.

Set a fallback datafile

If you provide a sdkKey and a static fallback datafile for initialization, the SDK uses the fallback datafile immediately if it is valid while simultaneously downloading the datafile associated with the sdkKey. After the download completes, if the downloaded datafile is valid and has a more recent revision than the fallback datafile, the SDK updates the instance to use the downloaded datafile.

const { createInstance } = require('@optimizely/optimizely-sdk');
const datafile = '{"version": "4", "rollouts": [], "typedAudiences": [], "anonymizeIP": false, "projectId": "12345", "variables": [], "featureFlags": [], "experiments": [], "audiences": [], "groups": [], "attributes": [], "botFiltering": false, "accountId": "12345", "events": [], "revision": "1"}'; 
const optimizely = createInstance({
  sdkKey: '<Your_SDK_Key>',
  datafile,
});
// optimizelyClientInstance can be used immediately with the given datafile, but
// will download the latest datafile and update itself

Dispose of the client

For effective resource management with the Optimizely JavaScript (Node) SDK, you must properly close the Optimizely client instance when it is no longer needed. This is done by calling optimizely.close().

The .close() method ensures that the processes and queues associated with the instance are properly released. This is essential for preventing memory leaks and ensuring that the application runs efficiently, especially in environments where resources are limited or in applications that create and dispose of many instances over their lifecycle.

See Close Optimizely Feature Experimentation JavaScript (Node) SDK on application exit.

OdpManager

OdpManager contains the logic supporting Real-Time Segments for Feature Experimentation-related features, including audience segments and sending ODP events.

The JavaScript (Node) SDK enables the Real-Time Segments for Feature Experimentation methods by default. But, the methods do nothing unless you have enabled and configured Real-Time Segments for Feature Experimentation.

If necessary, to disable Real-Time Segments for Feature Experimentation altogether, set disabled: true. See the following sample code for information.

Initialize the JavaScript (Node) SDK with Real-Time Segments for Feature Experimentation disabled:

const { createInstance } = require('@optimizely/optimizely-sdk');

const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>',
  odpOptions: {
    disabled: true,
  },
});

The following settings are optionally configurable when the JavaScript (Node) SDK is initialized:

• ODP SegmentsCache size – segmentsCacheSize
  • Default – 10,000
  • Set to 0 to disable caching.
• ODP SegmentsCache timeout (in seconds) – segmentsCacheTimeout
  • Default – 600 secs (10 minutes)
  • Set to 0 to disable timeout (never expires).
• ODP enable – disabled
  • Default – false (enabled)
  • The JavaScript (Node) SDK returns or logs an odpNotEnabled error when ODP is disabled and its features are requested.

Initialize the JavaScript (Node) SDK with Real-Time Segments for Feature Experimentation and custom settings:

const { createInstance } = require('@optimizely/optimizely-sdk');

const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>',
  odpOptions: {
    // disabled: false,  // change to true to disable
    eventApiTimeout: 1000,
    eventQueueSize: 1,
    // eventRequestHandler: new BrowserRequestHandler(),
    // eventManager: new OdpEventManager(),
    segmentsApiTimeout: 1000,
    segmentsCacheSize: 10,
    segmentsCacheTimeout: 1000,
    // segmentsCache: new ICache<string, string[]>(),
    // segmentsRequestHandler: new BrowserRequestHandler(),
    // segmentManager: new OdpSegmentManager(),
    
  },
});

Customize OdpManager

Using the parameters in odpOptions, you can customize the behavior of the ODP event and segment managers, even going as far as being able to replace them with your own implementations if desired.

Example odpOptions with a custom eventManager:

odpOptions: {  
  // ...
  eventManager: new OdpEventManager({  
    odpConfig, // new OdpConfig()  
    apiManager, // new OdpEventApiManager()  
    logger, // new LogHandler()  
    clientEngine: 'javascript-sdk',  
    clientVersion: '5.0.0',  
    batchSize: 10, 
    flushInterval: 2000,
  }),  
},

Customizing the OdpEventApiManager

export interface IOdpEventApiManager {  
  sendEvents(apiKey: string, apiHost: string, events: OdpEvent[]): Promise<boolean>;  
}

Example odpOptions with a custom segmentManager:

odpOptions: {  
  // ...
  segmentManager: new OdpSegmentManager(  
    odpConfig, // new OdpConfig()  
    new BrowserLRUCache<string, string[]>({  
      maxSize: 2,  
      timeout: 4000,  
    }), // new ICache<string, string[]>()  
    segmentApiManager, // new IOdpSegmentApiManager()  
  ),  
},

Custom cache

You can provide a custom cache, which is used to store the fetch_qualified_segments results. To provide a custom cache, you should implement the following interface:

Segments Cache Interface

export interface ICache<K, V> {
  lookup(key: K): V | null;
  save({ key, value }: { key: K; value: V }): void;
  reset(): void;
}

Custom Segments Cache Implementation example from LRUCache implementation

export class LRUCache<K, V> implements ICache<K, V> {
  private _map: Map<K, CacheElement<V>> = new Map();
  private _maxSize; // Defines maximum size of _map
  private _timeout; // Milliseconds each entry has before it becomes stale

  get map(): Map<K, CacheElement<V>> {
    return this._map;
  }

  get maxSize(): number {
    return this._maxSize;
  }

  get timeout(): number {
    return this._timeout;
  }

  constructor({ maxSize, timeout }: LRUCacheConfig) {
    const logger = getLogger();

    logger.debug(`Provisioning cache with maxSize of ${maxSize}`);
    logger.debug(`Provisioning cache with timeout of ${timeout}`);

    this._maxSize = maxSize;
    this._timeout = timeout;
  }

  public lookup(key: K): V | null {
    if (this._maxSize <= 0) {
      return null;
    }

    const element: CacheElement<V> | undefined = this._map.get(key);

    if (!element) return null;

    if (element.is_stale(this._timeout)) {
      this._map.delete(key);
      return null;
    }

    this._map.delete(key);
    this._map.set(key, element);

    return element.value;
  }

  public save({ key, value }: { key: K; value: V }): void {
    if (this._maxSize <= 0) return;

    const element: CacheElement<V> | undefined = this._map.get(key);
    if (element) this._map.delete(key);
    this._map.set(key, new CacheElement(value));

    if (this._map.size > this._maxSize) {
      const firstMapEntryKey = this._map.keys().next().value;
      this._map.delete(firstMapEntryKey);
    }
  }

  public reset(): void {
    if (this._maxSize <= 0) return;

    this._map.clear();
  }

  public peek(key: K): V | null {
    if (this._maxSize <= 0) return null;

    const element: CacheElement<V> | undefined = this._map.get(key);

    return element?.value ?? null;
  }
}

Use authenticated datafile in secure environment

You can fetch the Optimizely datafile from an authenticated endpoint using a server-side (only) Optimizely Feature Experimentation SDK, like the JavaScript (Node) SDK.

To use an authenticated datafile, download your Optimizely environment's access token from the Optimizely app at Settings>Environments. Select your secure environment, and copy the Datafile access token. The example below shows how to initialize the Optimizely client using an access token and sdk_key, enabling the client to fetch the authenticated datafile and complete initialization.

import { createInstance } from '@optimizely/optimizely-sdk';

// fetch the datafile from an authenticated endpoint
const optimizely = createInstance({
  sdkKey: '<YOUR_SDK_KEY>',
  datafileOptions: {
    datafileAccessToken: '<YOUR_DATAFILE_ACCESS_TOKEN>',
  },
});

Source files

The source code files containing the implementation for the JavaScript (Node) SDK are available on GitHub.