congestion.limits — listora/ring-congestion 0.1.3

RateLimit clj protocol

A RateLimit has to define a key, a quota, and a time-to-live.

The key is used to store the rate limit counter in storage. The returned key needs to be unique for each counter, eg an IP-based rate limit would include the IP-address of the client in the returned key so that each IP would have their own counter in storage.

The quota represents the number of requests that can be made withint the provided ttl.

The ttl (ie time-to-live) describes the timespan after which the rate limit is reset. Eg a rate limit with a ttl of 1s would be reset every second.

Note: all three fns take the incoming request as an argument. This allows the return values to be generated dynamically from the request. Eg an IP-based rate limit can lookup the :remote-addr from the request to generate the key. The main utility of this is in procuducing the key from the request, but it is also possible to use the request to dynamically generate quotas or ttls.

get-key clj

(get-key self req)

get-quota clj

(get-quota self req)

get-ttl clj

(get-ttl self req)