Redis distributed lock implementation for Python based on Pub/Sub messaging.

• Ensure atomicity by using the SETNX operation.
• Implements a Pub/Sub messaging system between the client attempting to acquire the lock and the one currently holding it.
• Includes a forced timeout mechanism to prevent infinite loops when attempting to acquire the lock.
• Supports asynchronous operations.

$> pip install redis-lock-py

• Python >= 3.9
• redis-py >= 5.2.0

import redis from redis_lock import RedisLock client = redis.Redis(host="127.0.0.1", port=6379) name = "foo" lock = RedisLock(client, name) if not lock.acquire(): raise Exception("Fail to acquire lock") print("Acquired lock successfully!") lock.release()

The redis-py library is required for Redis connection objects. After successfully acquiring the lock using RedisLock.acquire, ensure to release it by calling RedisLock.release to prevent lock retention.

import redis from redis_lock import RedisLock client = redis.Redis(host="127.0.0.1", port=6379) with RedisLock(client, name="foo", blocking_timeout=10): print("Acquired lock successfully!")

To avoid issues where the lock remains unreleased (potentially blocking other clients from acquiring it), you can use RedisLock with a context manager, which ensures that the lock is automatically released at the end of the with block. Both examples in sections 3.1 and 3.2 function in a same manner.

from redis.asyncio import Redis from redis_lock.asyncio import RedisLock client = Redis(host="127.0.0.1", port=6379) async with RedisLock(client, name="foo", blocking_timeout=10): print("Acquired lock successfully!")

redis-lock supports the asyncio platform.

import redis from redis_lock import RedisSpinLock client = redis.Redis(host="127.0.0.1", port=6379) lock = RedisSpinLock(client, name="foo") if not lock.acquire(blocking=True, sleep_time=0.1): raise Exception("Fail to acquire lock") print("Acquired lock successfully!") lock.release()

While a spin lock is available, it is not recommended unless there is a compelling reason to use it, as it is less efficient compared to the Pub/Sub messaging system.