notion-sdk-py is a simple and easy to use client library for the official Notion API.

It is meant to be a Python version of the reference JavaScript SDK, so usage should be very similar between both. 😊 (If not, please open an issue or PR!)

📢 Announcement (28-12-2023) — Release 2.2.1 is out and fixes iteration helpers.

What's new in 2.2.0 (26-12-2023):

• Icons and covers can now be removed from pages.
• filter_properties has been added to notion.pages.retrieve.
• You can now pass your own start_cursor in the iteration helpers.

Installation

pip install notion-client

Usage

Use Notion's Getting Started Guide to get set up to use Notion's API.

Import and initialize a client using an integration token or an OAuth access token.

import os
from notion_client import Client

notion = Client(auth=os.environ["NOTION_TOKEN"])

In an asyncio environment, use the asynchronous client instead:

from notion_client import AsyncClient

notion = AsyncClient(auth=os.environ["NOTION_TOKEN"])

Make a request to any Notion API endpoint.

See the complete list of endpoints in the API reference.

from pprint import pprint

list_users_response = notion.users.list()
pprint(list_users_response)

or with the asynchronous client:

list_users_response = await notion.users.list()
pprint(list_users_response)

This would output something like:

{'results': [{'avatar_url': 'https://secure.notion-static.com/e6a352a8-8381-44d0-a1dc-9ed80e62b53d.jpg',
              'id': 'd40e767c-d7af-4b18-a86d-55c61f1e39a4',
              'name': 'Avocado Lovelace',
              'object': 'user',
              'person': {'email': 'avo@example.org'},
              'type': 'person'},
             ...]}

All API endpoints are available in both the synchronous and asynchronous clients.

Endpoint parameters are grouped into a single object. You don't need to remember which parameters go in the path, query, or body.

my_page = notion.databases.query(
    **{
        "database_id": "897e5a76-ae52-4b48-9fdf-e71f5945d1af",
        "filter": {
            "property": "Landmark",
            "rich_text": {
                "contains": "Bridge",
            },
        },
    }
)

Handling errors

If the API returns an unsuccessful response, an APIResponseError will be raised.

The error contains properties from the response, and the most helpful is code. You can compare code to the values in the APIErrorCode object to avoid misspelling error codes.

import logging
from notion_client import APIErrorCode, APIResponseError

try:
    my_page = notion.databases.query(
        **{
            "database_id": "897e5a76-ae52-4b48-9fdf-e71f5945d1af",
            "filter": {
                "property": "Landmark",
                "rich_text": {
                    "contains": "Bridge",
                },
            },
        }
    )
except APIResponseError as error:
    if error.code == APIErrorCode.ObjectNotFound:
        ...  # For example: handle by asking the user to select a different database
    else:
        # Other error handling code
        logging.error(error)

Logging

The client emits useful information to a logger. By default, it only emits warnings and errors.

If you're debugging an application, and would like the client to log request & response bodies, set the log_level option to logging.DEBUG.

notion = Client(
    auth=os.environ["NOTION_TOKEN"],
    log_level=logging.DEBUG,
)

You may also set a custom logger to emit logs to a destination other than stdout. Have a look at Python's logging cookbook if you want to create your own logger.

Client options

Client and AsyncClient both support the following options on initialization. These options are all keys in the single constructor parameter.

Full API responses

The following functions can distinguish between full and partial API responses.

from notion_client.helpers import is_full_page

full_or_partial_pages = await notion.databases.query(
    database_id="897e5a76-ae52-4b48-9fdf-e71f5945d1af"
)

for page in full_or_partial_pages["results"]:
    if not is_full_page_or_database(page):
        continue
    print(f"Created at: {page['created_time']}")

Utility functions

These functions can be helpful for dealing with any of the paginated APIs.

iterate_paginated_api(function, **kwargs) and its async version async_iterate_paginated_api(function, **kwargs) turn any paginated API into a generator.

The function parameter must accept a start_cursor argument. Example: notion.blocks.children.list.

from notion_client.helpers import iterate_paginated_api

for block in iterate_paginated_api(
    notion.databases.query, database_id="897e5a76-ae52-4b48-9fdf-e71f5945d1af"
):
    # Do something with block.
    ...

If you don't need a generator, collect_paginated_api(function, **kwargs) and its async version async_collect_paginated_api(function, **kwargs) have the same behavior than the previous functions, but return a list of all results from the paginated API.

from notion_client.helpers import collect_paginated_api

all_results = collect_paginated_api(
    notion.databases.query, database_id="897e5a76-ae52-4b48-9fdf-e71f5945d1af"
)

Testing

Run the tests with the pytest command. If you want to test against all Python versions, you can run tox instead.

The tests are using pytest-vcr's cassettes for simulating requests to the Notion API. To create new tests or run them without cassettes, you need to set up the environment variables NOTION_TOKEN and NOTION_TEST_PAGE_ID (a page where your integration has all the capabilities enabled).

The code will use the page at NOTION_TEST_PAGE_ID to generate a temporary environment with the Notion objects to be tested, which will be deleted at the end of the session.

Requirements

This package supports the following minimum versions:

• Python >= 3.7
• httpx >= 0.15.0

Earlier versions may still work, but we encourage people building new applications to upgrade to the current stable.

Getting help

If you want to submit a feature request for Notion's API, or are experiencing any issues with the API platform, please email developers@makenotion.com.

If you found a bug with the library, please submit an issue.