from sqlite3 import connect
Advanced sessions for Starlette and FastAPI frameworks
Install starsessions
package:
pip install starsessions
Use redis
extra for Redis support.
See the example application in examples/
directory of this repository.
- Add
starsessions.SessionMiddleware
to your application to enable session support, - Configure the session store and pass it to the middleware,
- Load the session in your view/middleware by calling
load_session(connection)
utility.
from starlette.applications import Starlette
from starlette.middleware import Middleware
from starlette.responses import JSONResponse
from starlette.routing import Route
from starsessions import CookieStore, load_session, SessionMiddleware
async def index_view(request):
await load_session(request)
session_data = request.session
return JSONResponse(session_data)
session_store = CookieStore(secret_key='TOP SECRET')
app = Starlette(
middleware=[
Middleware(SessionMiddleware, store=session_store, lifetime=3600 * 24 * 14),
],
routes=[
Route('/', index_view),
]
)
By default, the middleware uses strict defaults.
The cookie lifetime is limited to the browser session and sent via HTTPS protocol only.
You can change these defaults by changing cookie_https_only
and lifetime
arguments:
from starlette.middleware import Middleware
from starsessions import CookieStore, SessionMiddleware
session_store = CookieStore(secret_key='TOP SECRET')
middleware = [
Middleware(SessionMiddleware, store=session_store, cookie_https_only=False, lifetime=3600 * 24 * 14),
]
The example above will let session usage over insecure HTTP transport and the session lifetime will be set to 14 days.
The session data is not loaded by default. Call load_session
to load data from the store.
async def index_view(request):
await load_session(request)
request.session['key'] = 'value'
However, if you try to access an uninitialized session, SessionNotLoaded
exception will be raised.
async def index_view(request):
request.session['key'] = 'value' # raises SessionNotLoaded
You can automatically load a session by using SessionAutoloadMiddleware
middleware.
For performance reasons, the session is not autoloaded by default. Sometimes it is annoying to call load_session
too
often.
We provide SessionAutoloadMiddleware
class to reduce the boilerplate code by autoloading the session for you.
There are two options: always autoload or autoload for specific paths only. Here are examples:
from starlette.middleware import Middleware
from starsessions import CookieStore, SessionAutoloadMiddleware, SessionMiddleware
session_store = CookieStore(secret_key='TOP SECRET')
# Always autoload
middleware = [
Middleware(SessionMiddleware, store=session_store),
Middleware(SessionAutoloadMiddleware),
]
# Autoload session for selected paths
middleware = [
Middleware(SessionMiddleware, store=session_store),
Middleware(SessionAutoloadMiddleware, paths=['/admin', '/app']),
]
# regex patterns also supported
import re
admin_rx = re.compile('/admin*')
middleware = [
Middleware(SessionMiddleware, store=session_store),
Middleware(SessionAutoloadMiddleware, paths=[admin_rx]),
]
The default behavior of SessionMiddleware
is to expire the cookie after lifetime
seconds after it was set.
For example, if you create a session with lifetime=3600
, the session will be terminated exactly in 3600 seconds.
Sometimes this may not be what you need, so we provide an alternate expiration strategy - rolling sessions.
When rolling sessions are activated, the cookie expiration time will be extended by lifetime
value on every response.
Let's see how it works for example. First, on the first response you create a new session with lifetime=3600
,
then the user does another request, and the session gets extended by another 3600 seconds, and so on.
This approach is useful when you want to use short-timed sessions but don't want them to interrupt in the middle of
the user's operation. With the rolling strategy, a session cookie will expire only after some period of the user's
inactivity.
To enable the rolling strategy set rolling=True
.
from starlette.middleware import Middleware
from starsessions import SessionMiddleware
middleware = [
Middleware(SessionMiddleware, lifetime=300, rolling=True),
]
The snippet above demonstrates an example setup where the session will be dropped after 300 seconds (5 minutes) of inactivity, but will be automatically extended by another 5 minutes while the user is online.
You can pass cookie_path
argument to bind the session cookies to specific URLs. For example, to activate a session
cookie
only for the admin area, use cookie_path="/admin"
middleware argument.
from starlette.middleware import Middleware
from starsessions import SessionMiddleware
middleware = [
Middleware(SessionMiddleware, cookie_path='/admin'),
]
All other URLs not matching the value of cookie_path
will not receive cookies thus session will be unavailable.
You can also specify which hosts can receive a cookie by passing cookie_domain
argument to the middleware.
from starlette.middleware import Middleware
from starsessions import SessionMiddleware
middleware = [
Middleware(SessionMiddleware, cookie_domain='example.com'),
]
Note, this makes session cookies available for subdomains too. For example, when you set
cookie_domain=example.com
then session cookie will be available on subdomains likeapp.example.com
.
If you want the session cookie to be automatically removed from the browser when the tab closes set lifetime
to 0
.
Note, this depends on browser implementation!
from starlette.middleware import Middleware
from starsessions import SessionMiddleware
middleware = [
Middleware(SessionMiddleware, lifetime=0),
]
Class: starsessions.InMemoryStore
Simply stores data in memory. The data is cleared after the server restart. Mostly for use with unit tests.
Class: starsessions.CookieStore
Stores session data in a signed cookie on the client.
Class: starsessions.stores.redis.RedisStore
Stores session data in a Redis server. The store accepts either a connection URL or an instance of Redis
.
Requires redis-py, use
pip install starsessions[redis]
Note, redis-py requires explicit disconnect of connection. The library does not handle it for you at the moment. The recommended solution is to pass a Redis instance to the store and call
.close()
on application shutdown. For example, you can close the connection using lifespan handler. See more https://redis-py.readthedocs.io/en/latest/examples/asyncio_examples.html
from redis.asyncio import Redis
from starsessions.stores.redis import RedisStore
client = Redis.from_url('redis://localhost')
store = RedisStore(connection=client)
store = RedisStore(connection=client)
# close connection on shutdown
await client.close()
By default, all keys in Redis prefixed with starsessions.
. If you want to change this use prefix
argument.
from starsessions.stores.redis import RedisStore
store = RedisStore(url='redis://localhost', prefix='my_sessions')
Prefix can be a callable:
from starsessions.stores.redis import RedisStore
def make_prefix(key: str) -> str:
return 'my_sessions_' + key
store = RedisStore(url='redis://localhost', prefix=make_prefix)
The library automatically manages key expiration, usually you have nothing to do with it.
But for cases when lifetime=0
we don't know when the session will be over, and we have to heuristically calculate TTL,
otherwise the data will remain in Redis forever. At this moment, we just set 30 days TTL. You can change it by
setting gc_ttl
value on the store.
from starsessions.stores.redis import RedisStore
store = RedisStore(url='redis://localhost', gc_ttl=3600) # max 1 hour
Creating new stores is quite simple. All you need is to extend starsessions.SessionStore
class and implement abstract methods.
Here is an example of how we can create a memory-based session store. Note, that it is important that write
method
returns session ID as a string value.
from typing import Dict
from starsessions import SessionStore
# instance of class that manages session persistence
class InMemoryStore(SessionStore):
def __init__(self):
self._storage = {}
async def read(self, session_id: str, lifetime: int) -> bytes:
""" Read session data from a data source using session_id. """
return self._storage.get(session_id, {})
async def write(self, session_id: str, data: bytes, lifetime: int, ttl: int) -> str:
""" Write session data into the data source and return session ID. """
self._storage[session_id] = data
return session_id
async def remove(self, session_id: str):
""" Remove session data. """
del self._storage[session_id]
async def exists(self, session_id: str) -> bool:
return session_id in self._storage
The write
accepts two special arguments: lifetime
and ttl
.
The difference is that lifetime
is the total session duration (set by the middleware)
and ttl
is the remaining session time. After ttl
seconds the data can be safely deleted from the storage.
Your custom backend has to correctly handle cases when
lifetime = 0
. In such cases, you don't have an exact expiration value, and you would have to find a way to extend session TTL on the storage side, if any.
The library automatically serializes session data to string using JSON.
By default, we use starsessions.JsonSerializer
but you can implement your own by extending starsessions.Serializer
class.
import json
import typing
from starlette.middleware import Middleware
from starsessions import Serializer, SessionMiddleware
class MySerializer(Serializer):
def serialize(self, data: typing.Any) -> bytes:
return json.dumps(data).encode('utf-8')
def deserialize(self, data: bytes) -> typing.Dict[str, typing.Any]:
return json.loads(data)
middleware = [
Middleware(SessionMiddleware, serializer=MySerializer()),
]
The middleware will remove session data and cookies if the session has no data. Use request.session.clear
to empty
data.
Sometimes you need a new session ID to avoid session fixation attacks (for example, after successful signs-in).
For that, use starsessions.session.regenerate_session_id(connection)
utility.
from starsessions.session import regenerate_session_id
from starlette.responses import Response
def login(request):
regenerate_session_id(request)
return Response('successfully signed in')