Speedbump

Flexible and extensible rate limiting with batteries included implementations.

Concepts

The crate is built around two (user-definable) traits.

• Store: Defines how the rate limiting state (e.g., counters, timestamps) is persisted. Implementations could include in-memory, Redis, or other storage backends.
• Strategy: Defines the rate limiting algorithm (e.g., fixed window, token bucket) and how it uses the persisted state to determine if a request should be allowed or rate-limited.

These traits are used within a Limiter struct, which orchestrates the retrieval of state from the Store and the application of the Strategy to decide whether a request should be rejected or allowed. This Limiter can then be integrated into middleware.

Todo

• Documentation
• More examples
• More strategies:
  • Sliding window
  • Token bucket
  • Leaky bucket
• More stores:
  • Redis
  • Memcached
• Middleware implementations:
  • Tower
  • Actix