Algolia Search is a hosted full-text, numerical, and faceted search engine capable of delivering realtime results from the first keystroke. The Algolia Search API Client for JavaScript lets you easily use the Algolia Search REST API from your JavaScript code.
The JavaScript client works both on the frontend (browsers) or on the backend (Node.js) with the same API.
The backend (Node.js) API can be used to index your data using your Algolia admin API keys.
Our JavaScript library is UMD compatible, you can use it with any module loader.
When not using any module loader, it will export an algoliasearch function in the window object.
You can find the full reference on Algolia's website.
You can either use a package manager like npm or include a <script> tag.
We are browserifyable and webpack friendly.
npm install algoliasearch --saveFor Typescript typings, we provide the definition file via typings
npm install --save @types/algoliasearchtns plugin add nativescript-algoliabower install algoliasearch -SjsDelivr is a global CDN delivery for JavaScript libraries.
To include the latest releases and all upcoming features and patches, use this:
<script src="https://cdn.jsdelivr.net/algoliasearch/3/algoliasearch.min.js"></script>We recommend using jsDelivr, but algoliasearch is also available at:
We have a lightweight build available that can only do searches. Use it when filesize is important to you or if you like to include only what you need.
Find it on jsDelivr:
<script src="https://cdn.jsdelivr.net/algoliasearch/3/algoliasearchLite.min.js"></script>In 30 seconds, this quick start tutorial will show you how to index and search objects.
You first need to initialize the client. For that you need your Application ID and API Key. You can find both of them on your Algolia account.
// var algoliasearch = require('algoliasearch');
// var algoliasearch = require('algoliasearch/reactnative');
// var algoliasearch = require('algoliasearch/lite');
// or just use algoliasearch if you are using a <script> tag
// if you are using AMD module loader, algoliasearch will not be defined in window,
// but in the AMD modules of the page
var client = algoliasearch('applicationID', 'apiKey');Without any prior configuration, you can start indexing 500 contacts in the contacts index using the following code:
for this use: admin API key in client instantiation
var index = client.initIndex('contacts');
var contactsJSON = require('./contacts.json');
index.addObjects(contactsJSON, function(err, content) {
if (err) {
console.error(err);
}
});You can now search for contacts using firstname, lastname, company, etc. (even with typos):
for this use: search only API key in client instantiation
// firstname
index.search('jimmie', function(err, content) {
console.log(content.hits);
});
// firstname with typo
index.search('jimie', function(err, content) {
console.log(content.hits);
});
// a company
index.search('california paint', function(err, content) {
console.log(content.hits);
});
// a firstname & company
index.search('jimmie paint', function(err, content) {
console.log(content.hits);
});Settings can be customized to tune the search behavior. For example, you can add a custom sort by number of followers to the already great built-in relevance:
for this use: admin API key in client instantiation
index.setSettings({
'customRanking': ['desc(followers)']
}, function(err, content) {
console.log(content);
});You can also configure the list of attributes you want to index by order of importance (first = most important):
Note: Since the engine is designed to suggest results as you type, you'll generally search by prefix. In this case the order of attributes is very important to decide which hit is the best:
index.setSettings({
'searchableAttributes': [
'lastname',
'firstname',
'company',
'email',
'city',
'address'
]
}, function(err, content) {
console.log(content);
});In most situations, there is no need to tune the options. We provide this list to be transparent with our users.
timeout(Number) timeout for requests to our servers, in milliseconds- in Node.js this is an inactivity timeout. Defaults to 15s
- in the browser, this is a global timeout. Defaults to 2s (incremental)
protocol(String) protocol to use when communicating with algolia- in the browser, we use the page protocol by default
- in Node.js it's https by default
- possible values: 'http:', 'https:'
hosts.read([String]) array of read hosts to use to call Algolia servers, computed automaticallyhosts.write([String]) array of write hosts to use to call Algolia servers, computed automaticallyhttpAgent(HttpAgent) node-only Node.js httpAgent instance to use when communicating with Algolia servers.dsn([Boolean=true]) enable or disable DSN (defaults to true). Disable this when you're in a backend situation, but still have worldwide DSN for another (frontend) search.
To pass an option, use:
var client = algoliasearch(applicationId, apiKey, {
timeout: 4000
})Every API call takes a callback as last parameter. This callback will then be called with two arguments:
- error: null or an Error object. More info on the error can be find in
error.message. - content: the object containing the answer from the server, it's a JavaScript object
If you do not provide a callback, you will get a promise (but never both).
Promises are the native Promise implementation.
We use jakearchibald/es6-promise as a polyfill when needed.
The request strategy used by the JavaScript client includes:
- On the browser:
- CORS for modern browsers
- XDomainRequest for IE <= 10
- JSONP in any situation where Ajax requests are unavailable or blocked.
- Node.js:
- native
httpmodule
- native
Connections are always keep-alive.
Browser only
To avoid performing the same API calls twice search results will be stored
in a cache that will be tied to your JavaScript client and index objects.
Whenever a call for a specific query (and filters) is made, we store the results
in a local cache. If you ever call the exact same query again, we read the
results from the cache instead of doing an API call.
This is particularly useful when your users are deleting characters from their current query, to avoid useless API calls. Because it is stored as a simple JavaScript object in memory, the cache is automatically reset whenever you reload the page.
It is never automatically purged, nor can it be completely disabled. Instead, we
provide the index.clearCache() (or client.clearCache() if you're using the
Search multiple indices method that you can call to reset it.
Node.js only
If you are behind a proxy, just set HTTP_PROXY or HTTPS_PROXY environment variables before starting your Node.js program.
HTTP_PROXY=http://someproxy.com:9320 node main.jsNode.js only
Keep-alive is activated by default.
Because of the nature of keepalive connections, your process will hang even if you do not do any more command using the client.
To fix this, we expose a client.destroy() method that will terminate all remaining alive connections.
You should call this method when you are finished working with the AlgoliaSearch API. So that your process will exit gently.
Note: keep-alive is still always activated in browsers, this is a native behavior of browsers.
The client will send you errors when a method call fails for some reasons.
You can get detailed debugging information:
index.search('something', function searchDone(err) {
if (err) {
console.log(err.message);
console.log(err.debugData);
return;
}
});err.debugData contains the array of requests parameters that were used to issue requests.
- Need help? Ask a question to the Algolia Community or on Stack Overflow.
- Found a bug? You can open a GitHub issue.