A standalone API for JS and other clients to use with Schema. Extend basic endpoints with your own custom functionality, for example, validating user input special or scoping product queries for client usage.
Schema.io <==> NodeJS API (This Package) <==> Your App (React/Angular/Ember/Rails/Etc)
- Clone this repository
- Create
.envfile in the repository and setdevelopmentvalues (see example.envfile) - Run
nvm install(make sure you have nvm installed) - Run
npm install - Run
npm run watch
NODE_ENV=development
PORT=3001
FORCE_SSL=true # redirect all requests to https
SCHEMA_CLIENT_ID=my-client-id
SCHEMA_CLIENT_KEY=my-client-keySessions are used to track persistent data associated with the end user, such as Account and Cart records. The client should pass an HTTP header named X-Session in every request to identify the current user.
X-Session: a4916a9b-717b-4d68-bb78-e5fc8e51591f
Your session ID can be of any format and should be created and stored by the client, typically stored as a cookie in a browser.
It's not strictly required to pass the session header, but certain endpoints will throw an error without one.
GET /v1/accountSensitive fields may be removed from the response. See api/v1/account.js for details.
PUT /v1/account
{
"first_name": "Example",
"last_name": "Customer",
"email": "[email protected]"
}Only specific fields may be updated. See api/v1/account.js for details.
POST /v1/account
{
"first_name": "Example",
"last_name": "Customer",
"email": "[email protected]"
}Only specific fields may be created. See api/v1/account.js for details.
POST /v1/account/login
{
"email": "[email protected]",
"password": "example password"
}If email and password are correct, an account record will be returned. Otherwise null.
POST /v1/account/logoutThis will remove the previously logged in account_id from the session.
POST /v1/account/recover
{
"email": "[email protected]",
"reset_url": "https://mystore.com/account/recover/{key}"
}This will send an email to the account if one found, that contains the reset_url parameter and a dynamically generated key appended to it. Your recovery page must recognize this key parameter and use it in the following step.
POST /v1/account/recover
{
"reset_key": "iud287ebuf9uwf92fdi2uhef872h",
"password": "new password"
}This will reset an account's password if the reset_key is found. If successful, it will return an account record and automatically login using the current session. If the reset_key has expired or is not found, it will return an error.
GET /v1/account/orders
{
"limit": 10,
"page": 1
}Get a list of orders placed by the current logged in account.
GET /v1/account/orders/:idGet a single order placed by the current logged in account.
GET /v1/account/addressesGet a list of addresses stored by the current logged in account.
GET /v1/account/cardsGet a list of credit cards stored by the current logged in account.
GET /v1/account/reviewsGet a list of product reviews created by the current logged in account.
GET /v1/account/creditsGet a list of credits applied to the current logged in account.
GET /v1/cartSensitive fields may be removed from the response. See api/v1/cart.js for details.
PUT /v1/cart
{
"shipping": {
"name": "Example Customer",
"address1": "123 Example St",
"city": "Example City",
"state": "EX",
"zip": "90210",
"country": "US",
"service": "fedex_ground"
},
"billing": {
"name": "Example Customer",
"address1": "123 Example St",
"city": "Example City",
"state": "EX",
"zip": "90210",
"country": "US",
"card": {
"token": "tok_..."
}
},
"items": [
{
"product_id": "...",
"quantity": 3,
"options": [
{
"id": "optional",
"value": "example"
}
]
}
]
}Only specific fields may be updated. See api/v1/cart.js for details.
This endpoint supports incremental updates for multi-page checkout flows. For example, you might pass shipping details in one request, and billing details in another request. Also, you'll probably want to use /v1/cart/add-item and /v1/cart/remove-item for incremental item updates.
If the cart does not exist for the current session, it will be automatically created.
POST /v1/cartThis will create a cart for the current session, if one does not already exist. Note that it's not usually necessary to make this request, since other requests like /v1/cart/add-item will do the same automatically on demand.
POST /v1/cart/add-item
{
"product_id": "...",
"quantity": 3,
"options": [
{
"id": "optional",
"value": "example"
}
]
}This will add an item to the cart. If the cart does not exist for the current session, it will be automatically created.
If the same product and options already exist in the cart, then its quantity will combined into a single item.
POST /v1/cart/remove-item
{
"item_id": "..."
}This will remove an item from the cart matching item_id. You can get the item ID value from any of the previous calls to the cart endpoint.
POST /v1/cart/apply-coupon
{
"code": "SHIPFREE"
}This will apply a valid coupon code to the cart and affect all relevant prices. If the coupon code is not found or is not valid, the server will respond with status 400 and an error message.
To remove an applied coupon, make the same request with "code": null.
GET /v1/cart/shipment-ratingThis will return an object with shipping services and prices relevant to the current cart state. For example, if your setup is configured with FedEx Ground enabled and real-time pricing, then you will get an object containing that shipping method and others available to the current shipping address in the cart.
The typical flow is to update the cart with shipping information first, then make this request to get available shipping options and prices.
POST /v1/cart/checkoutThis will attempt to convert a cart to an order including all the cart's details such as items, shipping, and billing. You may optionally pass any of these details along with this request in order to consolidate API calls.
If successful, an order record will be returned. Otherwise an error.
GET /v1/categories/:slugSensitive fields may be removed from the response. See api/v1/categories.js for details.
GET /v1/categories/:slug/childrenGET /v1/categories/:slug/productsThis will return all products in a category, including all nested sub-categories. For example, if :id is "jellybeans" which contains sub-categories "red" and "blue", then the result will contain all red and blue jellybean products.
Note: If you want to get products in a single category only and ignore sub-categories, then see /v1/products section.
GET /v1/products/:idSensitive fields may be removed from the response. See api/v1/products.js for details.
GET /v1/products?category=:slugThis will return all products contain in a single category, ignoring sub-categories.
Note: If you want to get products including all sub-categories, then see /v1/categories section.
POST /v1/contacts/subscribe
{
"first_name": "Example",
"last_name": "Customer",
"email": "[email protected]",
"email_optin_lists": ["optional", "lists"]
}All fields are optional except email. You may use email_optin_lists for tracking your own custom email list segments.
POST /v1/contacts/unsubscribe
{
"email": "[email protected]",
"email_optin_lists": ["optional"]
}This will flip the email_optin field to false on the contact record. If email_optin_lists is passed, then it will remove those lists from the contact record.
GET /v1/pages/:idPages are used to store content such as About Us, Privacy Policy, etcetera.
GET /v1/pages/:id/articlesArticles are useful for different things depending on the page itself. For example, you might think of articles on a Gallery page as images, or articles in a knowledge base. Use your imagination.
GET /v1/pages/:id/articles/:article_idGET /v1/sessionThis will return current session data, including account_id and cart_id, along with any other arbitrary fields stored by your client.
PUT /v1/session
{
"arbitrary_field": "example"
}Update the current session with any fields that might be useful to your client. Specific fields such as account_id and cart_id are restricted and will result in an error if passed.
npm testAs a best practice, you should write tests for all new or modified endpoints.
The test setup includes a Schema client that should be used to stub itself with expected requests and results. This allows you to easily test your own API code, without calling out to the Schema.io API itself. It makes test faster and more reliable.
Here's an example using this test client:
const schema = Test.schemaClient();
const api = Test.apiClient(schema);
describe('/v1/account', () => {
describe('PUT /v1/account', () => {
it('throws an error when logged out', () => {
return api.put('/v1/account').then(result => {
assert.ok(result && result.error);
});
});
describe('when logged in', () => {
beforeEach(() => {
schema.reset();
schema.expectLoggedIn();
});
it('sets account fields which are allowed', () => {
schema.expects([
{
method: 'put',
url: '/accounts/{id}',
result: {
first_name: 'foo',
last_name: 'bar',
}
}
]);
return api.put('/v1/account', {
first_name: 'foo',
last_name: 'bar',
}).then(result => {
assert.ok(result && result.first_name);
});
});
it('throws an error when updating restricted fields', () => {
return api.put('/v1/account', {
bad_field: 'foo',
}).then(result => {
assert.ok(result && result.error);
assert.equal(result.error, 'Error: Query may not specify `bad_field` on this resource');
});
});
});
});
});Need help with this package? Visit us in Slack at https://slack.schema.io
Pull requests are welcome.
MIT