The Shopify Admin API Python Library
You should be signed up as a partner on the Shopify Partners Dashboard so that you can create and manage shopify applications.
To easily install or upgrade to the latest release, use pip.
pip install --upgrade ShopifyAPI
- Usage
- Using Development Version
- Relative Cursor Pagination
- Set up pre-commit locally [OPTIONAL]
- Limitations
- Additional Resources
-
First create a new application in the Partners Dashboard, and retrieve your API Key and API Secret Key.
-
We then need to supply these keys to the Shopify Session Class so that it knows how to authenticate.
import shopify shopify.Session.setup(api_key=API_KEY, secret=API_SECRET)
-
In order to access a shop's data, apps need an access token from that specific shop. We need to authenticate with that shop using OAuth, which we can start in the following way:
shop_url = "SHOP_NAME.myshopify.com" api_version = '2024-07' state = binascii.b2a_hex(os.urandom(15)).decode("utf-8") redirect_uri = "http://myapp.com/auth/shopify/callback" scopes = ['read_products', 'read_orders'] newSession = shopify.Session(shop_url, api_version) auth_url = newSession.create_permission_url(scopes, redirect_uri, state) # redirect to auth_url
-
Once the merchant accepts, the shop redirects the owner to the
redirect_uri
of your application with a parameter named 'code'. This is a temporary token that the app can exchange for a permanent access token. You should compare the state you provided above with the one you received back to ensure the request is correct. Now we can exchange the code for an access_token when you get the request from shopify in your callback handler:session = shopify.Session(shop_url, api_version) access_token = session.request_token(request_params) # request_token will validate hmac and timing attacks # you should save the access token now for future use.
-
Now you're ready to make authorized API requests to your shop!:
session = shopify.Session(shop_url, api_version, access_token) shopify.ShopifyResource.activate_session(session) shop = shopify.Shop.current() # Get the current shop product = shopify.Product.find(179761209) # Get a specific product # execute a graphQL call shopify.GraphQL().execute("{ shop { name id } }")
Alternatively, you can use temp to initialize a Session and execute a command:
with shopify.Session.temp(shop_url, api_version, token): product = shopify.Product.find()
-
It is best practice to clear your session when you're done. A temporary session does this automatically:
shopify.ShopifyResource.clear_session()
Private apps are a bit quicker to use because OAuth is not needed. You can create the private app in the Shopify Merchant Admin. You can use the Private App password as your access_token
:
session = shopify.Session(shop_url, api_version, private_app_password)
shopify.ShopifyResource.activate_session(session)
# ...
shopify.ShopifyResource.clear_session()
with shopify.Session.temp(shop_url, api_version, private_app_password):
shopify.GraphQL().execute("{ shop { name id } }")
Note: Your application must be public to test the billing process. To test on a development store use the 'test': True
flag
- Create charge after session has been activated
application_charge = shopify.ApplicationCharge.create({ 'name': 'My public app', 'price': 123, 'test': True, 'return_url': 'https://domain.com/approve' }) # Redirect user to application_charge.confirmation_url so they can approve the charge
- After approving the charge, the user is redirected to
return_url
withcharge_id
parameter (Note: This action is no longer necessary if the charge is created with API version 2021-01 or later)charge = shopify.ApplicationCharge.find(charge_id) shopify.ApplicationCharge.activate(charge)
- Check that
activated_charge
status isactive
activated_charge = shopify.ApplicationCharge.find(charge_id) has_been_billed = activated_charge.status == 'active'
It is recommended to have at least a basic grasp on the principles of the pyactiveresource library, which is a port of rails/ActiveResource to Python and upon which this package relies heavily.
Instances of pyactiveresource
resources map to RESTful resources in the Shopify API.
pyactiveresource
exposes life cycle methods for creating, finding, updating, and deleting resources which are equivalent to the POST
, GET
, PUT
, and DELETE
HTTP verbs.
product = shopify.Product()
product.title = "Shopify Logo T-Shirt"
product.id # => 292082188312
product.save() # => True
shopify.Product.exists(product.id) # => True
product = shopify.Product.find(292082188312)
# Resource holding our newly created Product object
# Inspect attributes with product.attributes
product.price = 19.99
product.save() # => True
product.destroy()
# Delete the resource from the remote server (i.e. Shopify)
Here is another example to retrieve a list of open orders using certain parameters:
new_orders = shopify.Order.find(status="open", limit="50")
Some resources such as Fulfillment
are prefixed by a parent resource in the Shopify API (e.g. orders/450789469/fulfillments/255858046
). In order to interact with these resources, you must specify the identifier of the parent resource in your request.
shopify.Fulfillment.find(255858046, order_id=450789469)
This package also includes the shopify_api.py
script to make it easy to open an interactive console to use the API with a shop.
- Obtain a private API key and password to use with your shop (step 2 in "Getting Started")
- Save your default credentials:
shopify_api.py add yourshopname
- Start the console for the connection:
shopify_api.py console
- To see the full list of commands, type:
shopify_api.py help
This library also supports Shopify's new GraphQL API. The authentication process is identical. Once your session is activated, simply construct a new graphql client and use execute
to execute the query.
result = shopify.GraphQL().execute('{ shop { name id } }')
You can perform more complex operations using the variables
and operation_name
parameters of execute
.
For example, this GraphQL document uses a fragment to construct two named queries - one for a single order, and one for multiple orders:
# ./order_queries.graphql
fragment OrderInfo on Order {
id
name
createdAt
}
query GetOneOrder($order_id: ID!){
node(id: $order_id){
...OrderInfo
}
}
query GetManyOrders($order_ids: [ID]!){
nodes(ids: $order_ids){
...OrderInfo
}
}
Now you can choose which operation to execute:
# Load the document with both queries
document = Path("./order_queries.graphql").read_text()
# Specify the named operation to execute, and the parameters for the query
result = shopify.GraphQL().execute(
query=document,
variables={"order_id": "gid://shopify/Order/12345"},
operation_name="GetOneOrder",
)
python setup.py sdist
pip install --upgrade dist/ShopifyAPI-*.tar.gz
Note Use the bin/shopify_api.py
script when running from the source tree. It will add the lib directory to start of sys.path, so the installed version won't be used.
pip install setuptools --upgrade
python setup.py test
Cursor based pagination support has been added in 6.0.0.
import shopify
page1 = shopify.Product.find()
if page1.has_next_page():
page2 = page1.next_page()
# to persist across requests you can use next_page_url and previous_page_url
next_url = page1.next_page_url
page2 = shopify.Product.find(from_=next_url)
Pre-commit is set up as a GitHub action that runs on pull requests and pushes to the main
branch. If you want to run pre-commit locally, install it and set up the git hook scripts
pip install -r requirements.txt
pre-commit install
Currently there is no support for:
- asynchronous requests
- persistent connections