Train's Rate Limiter Implementation

[TrainTravel]

In this repo, a rate limiter class is implemented using the simple "Token Bucket Algorithm".

Then, in the ResourceController, a concurrentHashMap to hold a mapping between client's IP address and its own bucket(limiter). ConcurrentHasMap is used to ensure that the code is thread-safe.

With each successful request, client can see how many tokens are left in the header "X-Rate-Limit-Remaining".

When the client is rate-limited, the header includes "X-Rate-Limit-Retry-After-Seconds"

Unit tests were written.

Further improvements can be done in taken into consideration of using the functional endpoints provided by Spring Webflux.

And also, error handling haven't been added to the REST endpoint.