Sign Up for Free

RunKit +

Try any Node.js package right in your browser

This is a playground to test code. It runs a full Node.js environment and already has all of npm’s 400,000 packages pre-installed, including addsearch-js-client with all npm packages installed. Try it out:

var addsearchJsClient = require("addsearch-js-client")

This service is provided by RunKit and is not affiliated with npm, Inc or the package authors.

addsearch-js-client v0.4.3

AddSearch API JavaScript client

AddSearch Search API Client for JavaScript

AddSearch is a Search-as-a-Service for all your search needs. This API Client lets you easily use the AddSearch Search API with JavaScript on web browsers or from Node.js.

Quick Start

The library is available on the global CDN jsDelivr:

<script src=""></script>

Or install the library locally to use it with Node.js:

npm install addsearch-js-client --save

After installation, add the library to your JS code

var AddSearchClient = require('addsearch-js-client');

Or use import in ES6

import AddSearchClient from 'addsearch-js-client';

Execute the first search query

// Create client with your 32-character SITEKEY
var client = new AddSearchClient('YOUR PUBLIC SITEKEY');

// Callback function
var cb = function(res) {
  // Print results to console

// Execute search. Callback function will be called with search results'keyword', cb);

Publicly accessible functions

The client provides the following functions.

Fetch search results

// Search with a specific keyword'keyword', callback);

// Search with the previously used keyword or execute a "match all" query;

// Search with the previously used keyword and callback (e.g. after modifying filters);

Fetch search suggestions

Search suggestions are keywords and search phrases that real users have used in your search. Configure Search suggestions on AddSearch Dashboard before using this function.

// Get suggestions starting with a specific prefix
client.suggestions('a', callback);

Set the number of search suggestions to fetch

// Number of search suggestions to fetch (default 10)

Custom field autocompletion

Custom fields autocomplete can be used for predictive search. For example, product names or categories can be suggested as the keyword is being typed in.

// Fetch custom field values starting with a specific prefix In this example, fetch records 
// starting with *a* from the *custom_fields.brand* field. Results could be "adidas, apple, azure"
client.autocomplete('custom_fields.brand', 'a', callback);

Set the number of custom field autocompletion results to fetch

// Number of autocompletion results to fetch (default 10)

Search with fuzzy matching

Fuzzy matching is used for typo tolerance. There are four options:

  • false: No typo tolerance
  • true: Exact matches and fuzzy matches are equal
  • "auto": Exact matches first, followed by fuzzy matches
  • "retry": Show exact matches only. If none were found, show fuzzy matches
// Control fuzzy matching used for typo-tolerance
// Possible values true/false/"auto"/"retry" (default: "auto")

Postfix wildcard

Enable or disable postfix wildcard. I.e. should keyword "add" match to "addsearch" or should it just match to the term add

// Possible values true/false (default: true)


Set page number, page size and sorting parameters. It's possible to order results by:

  • relevance (descending)
  • date (ascending or descending)
  • custom field value (ascending or descending. E.g. custom_fields.price)
// Defaults: page "1", pageSize "10", sortBy "relevance", sortOrder "desc"
client.setPaging(page, pageSize, sortBy, sortOrder);

Other functions.

// Next page (call search function to fetch results)

// Previous page


Define language filter

// Fetch documents in specific language (e.g. "en" or "de")

Define publishing date filter

// Documents published between specific date range
client.setDateFilter('2019-01-01', '2019-01-31');

Define price range filter

// Products in specific price range (in cents. e.g. 100,00 - 200,00)
client.setPriceRangeFilter('10000', '20000');

Define category filters

Filter by URL patterns, document types or addsearch-category meta tag values. See the full documentation.

// Only PDF files or products

Custom field filters

Filter by custom fields. Custon fields can be defined in meta tags or AddSearch crawler can pick them up from your HTML or JSON data. See the full documentation.

// Search by specific city (Berlin, Paris or Boston)

// Remove Paris (Berlin and Boston remaining)

// Remove all cities

Set filtering object

Set complex filtering object that can contain nested and, or, not, and range filters.

// Find results where brand is apple, color is not white, and price is between 200 and 500
var filter = {
     {'custom_fields.brand': 'apple'},
     {'not': {'custom_fields.color': 'white'}},
     {'range': {'custom_fields.price': {'gt': 200, 'lt':500}}}


Set result type

// By default, fetch all search results
// If "organic", Pinned results and Promotions are left out


// Declare fields for faceting. Number of hits found from
// these fields will be returned

By default, 10 facets with most hits are returned per field. Use the following function to get more or less facets.


Numerical range facets

Group numerical custom fields into range buckets.

// Define ranges. E.g. products with price $0-$100, $100-$200, and over $200.
// From value is inclusive, to value is exclusive
var ranges = [
  {'to': 100},
  {'from': 100, 'to': 200},
  {'from': 200}

// Parameters: field name, range array
client.addRangeFacet('custom_fields.price', ranges);

Field statistics

Get minimum, maximum, and average values of a numerical or date-based custom field. The information is handy for applications like range filtering.

// Search response will have a fieldStats element with information like 
// custom_fields.price: {min: 1230, max: 1590, avg: 1382}

Search analytics

Send search event to analytics

When search is executed, send the event to your AddSearch Analytics Dashboard.

// If the numberOfResults is 0, the search is shown in the list of "queries with no hits"
client.sendStatsEvent('search', keyword, {numberOfResults: n});

Send click event to analytics

When a search results is clicked, send the event to your AddSearch Analytics Dashboard. Click information is shown in your statistics and used by the self-learning search algorithm.

// documentId is the 32-character long id that is part of each hit in search results.
// position is the position of the document that was clicked, the first result being 1
client.sendStatsEvent('click', keyword, {documentId: id, position: n});

Set or get stats session ID

Control the search session ID manually. Search queries with the same ID are grouped on the Analytics Dashboard. For example, in a search-as-you-type implementation the final keyword of a given session is shown.


Collect search events automatically

Send search events automatically to the Analytics Dashboard. Not recommended in search-as-you-type implementations, as every keystroke would fire a statistics event

// Control whether search queries are sent to your AddSearch Analytics Dashboard automatically or not (default: true)


Set user token (for personalized search results)

// Add a user token to the search request (if personalization in use)

Send personalization events with search query

In personalized search, user events are typically sent to AddSearch via API and a user token is passed with the search query (see setUserToken function). An alternative way is to send user events needed for personalization with the search query.

// Events depend on the personalization strategy
// Contact AddSearch for more information
var events = [
  {favorite_genre: 'rock'},
  {favorite_band: 'Red Hot Chili Peppers'},
  {least_favorite_genre: 'country'}



Set JSON Web Token (for authentication)

// Add JWT to the search request (if protected search index)

Set API throttling

// Set API call throttle time in milliseconds. Default is 200.

Supported web browsers and node.js versions

The client is tested on

  • Chrome
  • Firefox
  • Edge
  • Safari 6.1+
  • Internet Explorer 10+
  • Node.js 4+


To modify this client library, clone this repository to your computer and execute following commands.

Install dependencies

npm install


Re-compile automatically when source files are changed

npm run watch

Run tests

npm test


npm run build

Built bundle is saved under the dist/ folder


Feel free to send any questions, ideas, and suggestions at or visit for more information.

RunKit is a free, in-browser JavaScript dev environment for prototyping Node.js code, with every npm package installed. Sign up to share your code.
Sign Up for Free