aiohttp

Rate-limit middleware using client IP with aiohttp's @web.middleware decorator.

pip install aiohttp rate-limiter

Request Flow

sequenceDiagram
    participant C as Client
    participant M as rate_limit_middleware
    participant RL as RateLimiter
    participant H as Handler
    C->>M: HTTP Request
    M->>RL: allow(request.remote)
    alt allowed
        RL-->>M: true
        M->>H: handler(request)
        H-->>M: response
        M-->>C: 200 OK
    else denied
        RL-->>M: false
        M-->>C: 429 Rate limit exceeded
    end

Code

from aiohttp import web

from rate_limiter import RateLimiter, TokenBucketStrategy

limiter = RateLimiter(lambda: TokenBucketStrategy(capacity=10, rate=2))


@web.middleware
async def rate_limit_middleware(request: web.Request, handler):
    key = request.remote or "unknown"
    if not await limiter.allow(key):
        return web.Response(text="Rate limit exceeded", status=429)
    return await handler(request)


async def index(request: web.Request) -> web.Response:
    return web.json_response({"message": "Hello, World!"})


app = web.Application(middlewares=[rate_limit_middleware])
app.router.add_get("/", index)

if __name__ == "__main__":
    web.run_app(app, port=8080)

aiohttp is natively async, so the middleware awaits the limiter directly — no asyncio.run() bridge needed.

View Source on GitHub