The Stripe Node library provides convenient access to the Stripe API from applications written in server-side JavaScript.
Please keep in mind that this package is for use with server-side Node that uses Stripe secret keys. To maintain PCI compliance, tokenization of credit card information should always be done with Stripe.js on the client side. This package should not be used for that purpose.
See the Node API docs.
Install the package with:
npm install stripe --save
The package needs to be configured with your account's secret key which is available in your Stripe Dashboard. Require it with the key's value:
const stripe = require('stripe')('sk_test_...');
const customer = await stripe.customers.create({
email: '[email protected]'
});
Or with versions of Node.js prior to v7.9:
var stripe = require('stripe')('sk_test_...');
stripe.customers.create(
{ email: '[email protected]' },
function(err, customer) {
err; // null if no error occurred
customer; // the created customer object
}
);
Or using ES modules, this looks more like:
import Stripe from 'stripe';
const stripe = Stripe('sk_test_...');
//…
Stripe does not currently maintain typings for this package, but there are community typings available from DefinitelyTyped.
To install:
npm install --dev @types/stripe
To use:
// Note `* as` and `new Stripe` for TypeScript:
import * as Stripe from 'stripe';
const stripe = new Stripe('sk_test_...');
const customer: Promise<Stripe.customers.ICustomer> = stripe.customers.create(/* ... */);
Every method returns a chainable promise which can be used instead of a regular callback:
// Create a new customer and then a new charge for that customer:
stripe.customers.create({
email: '[email protected]'
}).then((customer) => {
return stripe.customers.createSource(customer.id, {
source: 'tok_visa'
});
}).then((source) => {
return stripe.charges.create({
amount: 1600,
currency: 'usd',
customer: source.customer
});
}).then((charge) => {
// New charge created on a new customer
}).catch((err) => {
// Deal with an error
});
Request timeout is configurable (the default is Node's default of 120 seconds):
stripe.setTimeout(20000); // in ms (this is 20 seconds)
A per-request Stripe-Account
header for use with Stripe Connect
can be added to any method:
// Retrieve the balance for a connected account:
stripe.balance.retrieve({
stripe_account: 'acct_foo'
}).then((balance) => {
// The balance object for the connected account
}).catch((err) => {
// Error
});
An https-proxy-agent can be configured with
setHttpAgent
.
To use stripe behind a proxy you can pass to sdk:
if (process.env.http_proxy) {
const ProxyAgent = require('https-proxy-agent');
stripe.setHttpAgent(new ProxyAgent(process.env.http_proxy));
}
Automatic network retries can be enabled with setMaxNetworkRetries
. This will retry requests n
times with exponential backoff if they fail due to an intermittent network problem. Idempotency keys are added where appropriate to prevent duplication.
// Retry a request once before giving up
stripe.setMaxNetworkRetries(1);
Some information about the response which generated a resource is available
with the lastResponse
property:
charge.lastResponse.requestId // see: https://stripe.com/docs/api/node#request_ids
charge.lastResponse.statusCode
The Stripe object emits request
and response
events. You can use them like this:
const stripe = require('stripe')('sk_test_...');
const onRequest = (request) => {
// Do something.
}
// Add the event handler function:
stripe.on('request', onRequest);
// Remove the event handler function:
stripe.off('request', onRequest);
{
api_version: 'latest',
account: 'acct_TEST', // Only present if provided
idempotency_key: 'abc123', // Only present if provided
method: 'POST',
path: '/v1/charges'
}
{
api_version: 'latest',
account: 'acct_TEST', // Only present if provided
idempotency_key: 'abc123', // Only present if provided
method: 'POST',
path: '/v1/charges',
status: 402,
request_id: 'req_Ghc9r26ts73DRf',
elapsed: 445 // Elapsed time in milliseconds
}
Stripe can optionally sign the webhook events it sends to your endpoint, allowing you to validate that they were not sent by a third-party. You can read more about it here.
Please note that you must pass the raw request body, exactly as received from Stripe, to the constructEvent()
function; this will not work with a parsed (i.e., JSON) request body.
You can find an example of how to use this with Express in the examples/webhook-signing
folder, but here's what it looks like:
const event = stripe.webhooks.constructEvent(
webhookRawBody,
webhookStripeSignatureHeader,
webhookSecret
);
If you're writing a plugin that uses the library, we'd appreciate it if you identified using stripe.setAppInfo()
:
stripe.setAppInfo({
name: 'MyAwesomePlugin',
version: '1.2.34', // Optional
url: 'https://myawesomeplugin.info', // Optional
});
This information is passed along when the library makes calls to the Stripe API.
As of stripe-node 6.11.0, you may auto-paginate list methods. We provide a few different APIs for this to aid with a variety of node versions and styles.
If you are in a Node environment that has support for async iteration, such as Node 10+ or babel, the following will auto-paginate:
for await (const customer of stripe.customers.list()) {
doSomething(customer);
if (shouldStop()) {
break;
}
}
If you are in a Node environment that has support for await
, such as Node 7.9 and greater,
you may pass an async function to .autoPagingEach
:
await stripe.customers.list().autoPagingEach(async (customer) => {
await doSomething(customer);
if (shouldBreak()) {
return false;
}
})
console.log('Done iterating.');
Equivalently, without await
, you may return a Promise, which can resolve to false
to break:
stripe.customers.list().autoPagingEach((customer) => {
return doSomething(customer).then(() => {
if (shouldBreak()) {
return false;
}
});
}).then(() => {
console.log('Done iterating.');
}).catch(handleError);
If you prefer callbacks to promises, you may also use a next
callback and a second onDone
callback:
stripe.customers.list().autoPagingEach(
function onItem(customer, next) {
doSomething(customer, function(err, result) {
if (shouldStop(result)) {
next(false); // Passing `false` breaks out of the loop.
} else {
next();
}
});
},
function onDone(err) {
if (err) {
console.error(err);
} else {
console.log('Done iterating.');
}
}
)
If your onItem
function does not accept a next
callback parameter or return a Promise,
the return value is used to decide whether to continue (false
breaks, anything else continues).
This is a convenience for cases where you expect the number of items
to be relatively small; accordingly, you must pass a limit
option
to prevent runaway list growth from consuming too much memory.
Returns a promise of an array of all items across pages for a list request.
const allNewCustomers = await stripe.customers.list({created: {gt: lastMonth}})
.autoPagingToArray({limit: 10000});
Run all tests:
$ npm install
$ npm test
Run a single test suite:
$ npm run mocha -- test/Error.spec.js
Run a single test (case sensitive):
$ npm run mocha -- test/Error.spec.js --grep 'Populates with type'
If you wish, you may run tests using your Stripe Test API key by setting the
environment variable STRIPE_TEST_API_KEY
before running the tests:
$ export STRIPE_TEST_API_KEY='sk_test....'
$ npm test