Overview
The RateLimiter interface is the core abstraction for evaluating whether a request should be allowed based on a rate limit policy. Implementations typically use a backing store (like Redis) to track request counts.
Package: io.github.v4runsharma.ratelimiter.core
Source: RateLimiter.java:7
Methods
evaluate
RateLimitDecision evaluate(String key, RateLimitPolicy policy)
The rate limit bucket key identifying the caller or resource. Should be stable and unique per bucket (e.g., “user:123”, “ip:192.168.1.1”).
The rate limit policy to enforce, containing the limit, window duration, and scope.
A decision object indicating whether the request is allowed, along with retry timing information.
Usage example
Direct usage
import io.github.v4runsharma.ratelimiter.core.RateLimiter;
import io.github.v4runsharma.ratelimiter.model.RateLimitPolicy;
import io.github.v4runsharma.ratelimiter.model.RateLimitDecision;
import java.time.Duration;
public class RateLimitService {
private final RateLimiter rateLimiter;
public RateLimitService(RateLimiter rateLimiter) {
this.rateLimiter = rateLimiter;
}
public boolean checkLimit(String userId) {
// Create a policy: 100 requests per minute
RateLimitPolicy policy = new RateLimitPolicy(
100,
Duration.ofMinutes(1),
"user"
);
// Evaluate the rate limit
String key = "user:" + userId;
RateLimitDecision decision = rateLimiter.evaluate(key, policy);
if (decision.isAllowed()) {
System.out.println("Request allowed");
return true;
} else {
System.out.println("Rate limit exceeded. Retry after: "
+ decision.getRetryAfterMillis() + "ms");
return false;
}
}
}
Custom implementation
import io.github.v4runsharma.ratelimiter.core.RateLimiter;
import io.github.v4runsharma.ratelimiter.model.RateLimitDecision;
import io.github.v4runsharma.ratelimiter.model.RateLimitPolicy;
public class InMemoryRateLimiter implements RateLimiter {
private final Map<String, AtomicInteger> buckets = new ConcurrentHashMap<>();
@Override
public RateLimitDecision evaluate(String key, RateLimitPolicy policy) {
AtomicInteger counter = buckets.computeIfAbsent(
key,
k -> new AtomicInteger(0)
);
int current = counter.incrementAndGet();
boolean allowed = current <= policy.getLimit();
if (allowed) {
return new RateLimitDecision(true, 0, null, null);
} else {
long retryAfter = policy.getWindow().toMillis();
return new RateLimitDecision(
false,
retryAfter,
policy.getWindow(),
policy.getWindow()
);
}
}
}
Implementations
The library provides the following implementations:
RedisRateLimiter
Production-ready implementation using Redis with sliding window algorithm:
import io.github.v4runsharma.ratelimiter.redis.RedisRateLimiter;
import org.springframework.data.redis.core.StringRedisTemplate;
RedisRateLimiter rateLimiter = new RedisRateLimiter(redisTemplate);