from __future__ import annotations

import threading
import time
import unittest
from collections.abc import Callable


class TokenBucket:
    """Thread-safe token-bucket rate limiter."""

    def __init__(
        self,
        capacity: float,
        refill_rate: float,
        *,
        clock: Callable[[], float] = time.monotonic,
    ) -> None:
        if capacity <= 0:
            raise ValueError("capacity must be positive")
        if refill_rate <= 0:
            raise ValueError("refill_rate must be positive")

        self.capacity = float(capacity)
        self.refill_rate = float(refill_rate)
        self._tokens = self.capacity
        self._clock = clock
        self._updated_at = clock()
        self._lock = threading.Lock()

    def _refill(self) -> None:
        now = self._clock()
        elapsed = now - self._updated_at
        if elapsed > 0:
            self._tokens = min(
                self.capacity,
                self._tokens + elapsed * self.refill_rate,
            )
            self._updated_at = now

    def consume(self, tokens: float = 1.0) -> bool:
        """Consume tokens immediately, returning False when unavailable."""
        if tokens <= 0:
            raise ValueError("tokens must be positive")
        if tokens > self.capacity:
            return False

        with self._lock:
            self._refill()
            if self._tokens < tokens:
                return False
            self._tokens -= tokens
            return True

    def retry_after(self, tokens: float = 1.0) -> float:
        """Return seconds until the requested tokens should be available."""
        if tokens <= 0:
            raise ValueError("tokens must be positive")
        if tokens > self.capacity:
            return float("inf")

        with self._lock:
            self._refill()
            return max(0.0, (tokens - self._tokens) / self.refill_rate)


class FakeClock:
    def __init__(self) -> None:
        self.now = 0.0

    def __call__(self) -> float:
        return self.now

    def advance(self, seconds: float) -> None:
        self.now += seconds


class TokenBucketTests(unittest.TestCase):
    def test_consumption_and_refill(self) -> None:
        clock = FakeClock()
        bucket = TokenBucket(2, 1, clock=clock)

        self.assertTrue(bucket.consume(2))
        self.assertFalse(bucket.consume())
        clock.advance(1)
        self.assertTrue(bucket.consume())

    def test_retry_after(self) -> None:
        clock = FakeClock()
        bucket = TokenBucket(5, 2, clock=clock)

        self.assertTrue(bucket.consume(5))
        self.assertEqual(bucket.retry_after(3), 1.5)
        clock.advance(1)
        self.assertEqual(bucket.retry_after(3), 0.5)


if __name__ == "__main__":
    unittest.main()