Skip to main content

Full stack authentication library for ASGI.

Project description

Imia

Imia (belarussian for "a name") is an authentication library for Starlette and FastAPI (python 3.8+).

PyPI GitHub Workflow Status GitHub Libraries.io dependency status for latest release PyPI - Downloads GitHub Release Date Lines of code

Production status

The library is considered in "beta" state thus may contain bugs or security issues, but I actively use it in production.

Installation

Install imia using PIP or poetry:

pip install imia
# or
poetry add imia

Features

  • Login/logout flows
  • Pluggable authenticators:
    • WWW-Basic
    • session
    • token
    • bearer token
    • any token (customizable)
    • API key
  • Database agnostic user storage
  • Authentication middleware
    • with fallback strategies:
      • redirect to an URL
      • raise an exception
      • do nothing
    • with optional URL protection
    • with option URL exclusion from protection
  • User Impersonation (stateless and stateful)
  • SQLAlchemy 1.4 (async mode) integration

TODO

  • remember me

A very quick start

If you are too lazy to read this doc, take a look into examples/ directory. There you will find several files demoing various parts of this library.

How it works?

Here are all moving parts:

  1. UserLike object, aka "user model" - is an arbitrary class that implements imia.UserLike protocol.
  2. a user provider - an adapter that loads user model (UserLike object) from the storage (a database).
  3. an authenticator - a class that loads user using the user provider from the request (eg. session)
  4. an authentication middleware that accepts an HTTP request and calls authenticators for a user model. The middleware always populates request.auth with UserToken.
  5. user token is a class that holds authentication state

When a HTTP request reaches your application, an imia.AuthenticationMiddleware will start handling it. The middleware iterates over configured authenticators and stops on the first one that returns non-None value. At this point the request is considered authenticated. If no authenticators return user model then the middleware will create anonymous user token. The user token available in request.auth property. Use user_token.is_authenticated token property to make sure that user is authenticated.

User authentication quick start

  1. Create a user model and implement methods defined by imia.UserLike protocol.
  2. Create an instance of imia.UserProvider that corresponds to your user storage. Feel free to create your own.
  3. Setup one or more authenticators and pass them to the middleware
  4. Add imia.AuthenticationMiddleware to your Starlette application

At this point you are done.

Here is a brief example that uses in-memory provider for demo purpose. For production environment you should use database backed providers like SQLAlchemyORMUserProvider or SQLAlchemyCoreUserProvider. Also, for simplicity reason we will not implement login/logout flow and will authenticate requests using API keys.

from dataclasses import dataclass, field

from starlette.applications import Starlette
from starlette.middleware import Middleware
from starlette.requests import Request
from starlette.responses import JSONResponse
from starlette.routing import Route

from imia import APIKeyAuthenticator, AuthenticationMiddleware, InMemoryProvider


@dataclass
class User:
    """This is our user model. It may be an ORM model, or any python class, the library does not care of it,
    it only expects that the class has methods defined by the UserLike protocol."""

    id: str
    password: str = 'password'
    scopes: list[str] = field(default_factory=list)

    def get_display_name(self) -> str:
        return self.id.split('@')[0].title()

    def get_id(self) -> str:
        return self.id

    def get_hashed_password(self) -> str:
        return self.password

    def get_scopes(self) -> list:
        return self.scopes


async def whoami_view(request: Request) -> JSONResponse:
    return JSONResponse({
        'id': request.auth.user_id,
        'name': request.auth.display_name,
    })


user_provider = InMemoryProvider({
    'user1@example.com': User(id='user1@example.com'),
    'user2@example.com': User(id='user2@example.com'),
})

authenticators = [
    APIKeyAuthenticator(user_provider=user_provider),
]

routes = [
    Route('/', whoami_view),
]

middleware = [
    Middleware(AuthenticationMiddleware, authenticators=authenticators)
]

app = Starlette(routes=routes, middleware=middleware)

Now save the file to myapp.py and run it with uvicorn application server:

uvicorn myapp:app

Open http://127.0.0.1:8000/ and see that your request is not authenticated and user is anonymous. Let's pass API key via query parameters to make the configured APIKeyAuthenticator to load user. This time open http://127.0.0.1:8000/?apikey=user1@example.com in your browser. Now the request is fully authenticated as User1 user.

For more details refer to the doc sections below.

Docs

  1. UserLike protocol (a user model)
  2. Load user from databases using User Providers
  3. Request authentication
  4. Built-in authenticators
  5. User token
  6. Passwords
  7. Login/Logout flow
  8. User impersontation

Usage

See examples/ directory.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

imia-0.5.3.tar.gz (14.3 kB view details)

Uploaded Source

Built Distribution

imia-0.5.3-py3-none-any.whl (13.8 kB view details)

Uploaded Python 3

File details

Details for the file imia-0.5.3.tar.gz.

File metadata

  • Download URL: imia-0.5.3.tar.gz
  • Upload date:
  • Size: 14.3 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.14 CPython/3.10.5 Linux/5.13.0-1031-azure

File hashes

Hashes for imia-0.5.3.tar.gz
Algorithm Hash digest
SHA256 e02cdebceef1828f076f52477bf78612dabf282ac9334855ffb491576c26bb1f
MD5 e561667ad5fce0e75f2bf2c9bb8b1b5e
BLAKE2b-256 7c675d7cc27a1c8b5d66f09736ad1ca513ccbf3dd9fa2f9a6d8b96124715fe00

See more details on using hashes here.

File details

Details for the file imia-0.5.3-py3-none-any.whl.

File metadata

  • Download URL: imia-0.5.3-py3-none-any.whl
  • Upload date:
  • Size: 13.8 kB
  • Tags: Python 3
  • Uploaded using Trusted Publishing? No
  • Uploaded via: poetry/1.1.14 CPython/3.10.5 Linux/5.13.0-1031-azure

File hashes

Hashes for imia-0.5.3-py3-none-any.whl
Algorithm Hash digest
SHA256 f8edce6c04c4eb4542aee0fe90126899858c3466636134dbfb414edb0a475273
MD5 2c52eb75d3c2909f1e13121f68876245
BLAKE2b-256 71657bb38c17d10016fe28a07921c5561a0e0045d72771fe8a9ab9bbcd267119

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page