Fingerprint is a device intelligence platform offering 99.5% accurate visitor identification
This library is a framework-agnostic wrapper around the Fingerprint Pro JavaScript Agent, adding multiple built-in caching mechanisms with recommended default settings.
[!NOTE] This library assumes you have a Fingerprint Pro subscription or trial, it is not compatible with the source-available FingerprintJS. See our documentation to learn more about the differences between Fingerprint Pro and FingerprintJS.
Using npm:
npm install @fingerprintjs/fingerprintjs-pro-spa
Using yarn:
yarn add @fingerprintjs/fingerprintjs-pro-spa
Using pnpm:
pnpm add @fingerprintjs/fingerprintjs-pro-spa
In order to identify visitors you'll need a Fingerprint Pro account (you can sign up for free).
Create a FpjsClient
instance before rendering or initializing your application. You should only have one instance of the client. You need to specify your public API key and other configuration options based on your chosen region and active integration.
import { FpjsClient, FingerprintJSPro } from '@fingerprintjs/fingerprintjs-pro-spa'
const fpjsClient = new FpjsClient({
// You can also pass these options later in `.init()` method
loadOptions: {
apiKey: '<PUBLIC_API_KEY>',
endpoint: [
//'https://metrics.yourwebsite.com',
FingerprintJSPro.defaultEndpoint,
],
scriptUrlPattern: [
// 'https://metrics.yourwebsite.com/web/v<version>/<apiKey>/loader_v<loaderVersion>.js',
FingerprintJSPro.defaultScriptUrlPattern,
],
// region: "eu"
},
})
[!NOTE] You must provide
loadOptions
containing your public API key either in the constructor or in theinit
method. If you don't, the SDK will throw an error. You can learn more about different load options here in the JS Agent documentation.
Before you start making identification requests to the Fingerprint Pro API, you need to initialize the JS Agent. This downloads the latest client-side logic from Fingerprint CDN. Call init()
before the getVisitorData()
method to avoid errors.
// with async/await
await fpjsClient.init()
const visitorData = await fpjsClient.getVisitorData()
// with promises
const visitorData = fpjsClient.init().then(() => {
return fpjsClient.getVisitorData()
})
You can also pass the loadOptions
into the init
method here. They will be merged with the options passed to the constructor.
await fpjsClient.init({
apiKey: '<PUBLIC_API_KEY>',
// endpoint: ["<CUSTOM_ENDPOINT>", FingerprintJSPro.defaultEndpoint],
// scriptUrlPattern: ["<CUSTOM_SCRIPT_URL>", FingerprintJSPro.defaultScriptUrlPattern],
// region: "eu"
})
The getVisitorData
method returns visitor identification data based on the request options.
Set ignoreCache
to true
to call the API even if the data is present in the cache.
// with async/await
const visitorData = await fpjsClient.getVisitorData({ extendedResult: true }, false)
// with promises
const visitorData = fpjsClient.getVisitorData({ extendedResult: true }).then((visitorData) => {
// use visitor data in your fraud prevention logic
checkIfFingerprintIsFraudulent(visitorData.visitorId) // this method is just an example, this SDK doesn't actually supply it
})
See the JS Agent API reference for more details.
The visitorId
provided by Fingerprint Identification is especially useful when combined with information you already know about your users, for example, account IDs, order IDs, etc. To learn more about various applications of the linkedId
and tag
, see Linking and tagging information.
const visitorData = await fpjsClient.getVisitorData({
linkedId: 'user_1234',
tag: {
userAction: 'login',
analyticsId: 'UA-5555-1111-1',
},
})
Fingerprint Pro usage is billed per API call. To avoid unnecessary API calls, it is a good practice to cache identification results. The SDK provides three ways to cache visitor data out of the box:
sessionStorage
localStorage
memory
nocache
You can specify the cacheLocation
option when creating the FpjsClient
:
const fpjsClient = new FpjsClient({
loadOptions: {
apiKey: 'your-fpjs-public-api-key',
},
cacheLocation: 'localstorage',
// You can also use the provided TypeScript enum
// cacheLocation: CacheLocation.LocalStorage
})
Cache keys are based on the combination of GetOptions. For example, API responses for calls with extendedResult: true
and extendedResult: false
are stored independently.
[!NOTE] If you use data from
extendedResult
, pay additional attention to your caching strategy. Some fields, for example,ip
orlastSeenAt
, might change over time for the same visitor.
You can ignore the cached result for a specific API call and using { ignoreCache: true }
:
const visitorData = await fpjsClient.getVisitorData({ ignoreCache: true })
Check if your response was retrieved from cache using the returned cacheHit
flag:
const { cacheHit, ...visitorData } = await fpjsClient.getVisitorData()
Use getVisitorDataFromCache
to directly retrieve responses from cache:
// Checks if request matching given options is present in cache
await fpjsClient.isInCache({ extendedResult: true })
// Returns cached visitor data based on the request options, or undefined if the data is not present in cache
const cachedResult = await fpjsClient.getVisitorDataFromCache({ extendedResult: true })
You can also use your custom cache implementation as described below.
The SDK can use a custom cache store implemented inside your application. This is useful when a different data store is more convenient in your environment, such as a hybrid mobile app.
You can provide an object to the cache
property of the SDK configuration that implements the following functions. All the functions can return a Promise or a static value.
Signature | Return type | Description |
---|---|---|
get(key) |
Promise | Returns the item from the cache with the specified key, or undefined if it was not found |
set(key: string, object: any) |
Promise |
Sets an item into the cache |
remove(key) |
Promise |
Removes a single item from the cache at the specified key, or no-op if the item was not found |
allKeys() |
Promise<string[]> or string [] | Returns the list of all keys. By default, the keys we use are prefixed with @fpjs@client@ but you can pass your own custom prefix as an option when you create the FpjsClient |
[!NOTE] The
cache
property takes priority overcacheLocation
if both are set. A warning is displayed in the console if that happens.
We export the internal InMemoryCache
, LocalStorageCache
, SessionStorageCache
, and CacheStub
implementations, so you can wrap your custom cache around these implementations if you wish.
Use the cacheTimeInSeconds
client constructor option to set a custom cache time. To ensure high identification accuracy we recommend not to cache visitors data for longer than 24 hours. If you pass a value higher than 86400 (60 x 60 x 24), the FpjsClient
constructor will throw an error.
To report problems, ask questions, or provide feedback, please use Issues. If you need private support, you can email us at oss-support@fingerprint.com.
This library uses Fingerprint Pro under the hood.
This project is licensed under the MIT license. See the LICENSE file for more information.