Module rate_limit 

Token-bucket rate limiting middleware — rate_limit::RateLimitLayer. Per-IP, per-header, or global. Returns 429 with Retry-After when exhausted. Token-bucket rate limiting middleware for axum routers.

Configurable per-IP or per-user limits with burst allowance. Returns 429 Too Many Requests when the bucket is exhausted.

Quick start

use rustango::rate_limit::{RateLimitLayer, RateLimitRouterExt};
use std::time::Duration;

let app = Router::new()
    .route("/api/login", post(login))
    .rate_limit(RateLimitLayer::per_ip(5, Duration::from_secs(60))); // 5 req/min

Strategy

• Token bucket: each key (IP or user id) gets capacity tokens.
• On each request, one token is removed. If empty, return 429.
• Tokens refill at capacity / refill_period per second.
• Buckets live in an in-process tokio::sync::Mutex<HashMap>. Process-local — for distributed enforcement, integrate with the cache layer (future slice).

Structs

RateLimitLayer

Rate-limit configuration.

Enums

KeyBy

Strategy for picking the bucket key per request.

Traits

RateLimitRouterExt

Extension trait — apply a rate-limit layer to a router.