49 lines
1.4 KiB
Python
49 lines
1.4 KiB
Python
|
from __future__ import annotations
|
||
|
|
|
||
|
|
import time
|
||
|
|
from collections import OrderedDict
|
||
|
|
from threading import Lock
|
||
|
|
|
||
|
|
|
||
|
|
class MessageDeduplicator:
|
||
|
|
TTL_SECONDS = 60.0
|
||
|
|
|
||
|
|
def __init__(self, max_size: int = 1000):
|
||
|
|
self._max_size = max_size
|
||
|
|
self._cache: OrderedDict[str, float] = OrderedDict()
|
||
|
|
self._lock = Lock()
|
||
|
|
|
||
|
|
def is_duplicate(self, message_id: str) -> bool:
|
||
|
|
now = time.monotonic()
|
||
|
|
with self._lock:
|
||
|
|
if message_id in self._cache:
|
||
|
|
ts = self._cache[message_id]
|
||
|
|
if now - ts < self.TTL_SECONDS:
|
||
|
|
return True
|
||
|
|
del self._cache[message_id]
|
||
|
|
|
||
|
|
self._cache[message_id] = now
|
||
|
|
self._cache.move_to_end(message_id)
|
||
|
|
|
||
|
|
if len(self._cache) > self._max_size:
|
||
|
|
self._cleanup_expired(now)
|
||
|
|
|
||
|
|
return False
|
||
|
|
|
||
|
|
def mark_seen(self, message_id: str) -> None:
|
||
|
|
with self._lock:
|
||
|
|
self._cache[message_id] = time.monotonic()
|
||
|
|
self._cache.move_to_end(message_id)
|
||
|
|
|
||
|
|
if len(self._cache) > self._max_size:
|
||
|
|
self._cleanup_expired(time.monotonic())
|
||
|
|
|
||
|
|
def _cleanup_expired(self, now: float) -> None:
|
||
|
|
expired = [k for k, ts in self._cache.items() if now - ts >= self.TTL_SECONDS]
|
||
|
|
for k in expired:
|
||
|
|
del self._cache[k]
|
||
|
|
|
||
|
|
def __len__(self) -> int:
|
||
|
|
with self._lock:
|
||
|
|
return len(self._cache)
|