87 lines
4.3 KiB
Python
87 lines
4.3 KiB
Python
import asyncio
|
||
import aio_pika
|
||
from config import RABBITMQ_URI
|
||
|
||
|
||
async def direct_consumer(queue_name: str, consumer_label: str):
|
||
"""
|
||
Direct Exchange 消费者：监听指定队列，处理对应路由键的消息
|
||
:param queue_name: 要监听的队列名称（与生产者中定义的队列一致）
|
||
:param consumer_label: 消费者标签（用于区分不同类型的消息处理器）
|
||
"""
|
||
# 1. 建立与 RabbitMQ 的连接（robust 模式自动重连，提升稳定性）
|
||
connection = await aio_pika.connect_robust(RABBITMQ_URI)
|
||
# 2. 创建通信信道（所有操作通过信道执行，减少 TCP 连接开销）
|
||
channel = await connection.channel()
|
||
|
||
# 3. 开启公平调度：确保消费者处理完 1 条消息后，再接收下 1 条
|
||
# 避免「快消费者空闲、慢消费者堆积」的不均衡问题
|
||
await channel.set_qos(prefetch_count=1)
|
||
|
||
# 4. 声明要监听的队列（与生产者中 queue_bindings 定义的队列完全一致）
|
||
# 若队列不存在（未执行 setup_direct_exchange），会报错提醒初始化
|
||
queue = await channel.declare_queue(
|
||
queue_name,
|
||
durable=True, # 与生产者一致：队列持久化（重启不丢失）
|
||
auto_delete=False # 队列不自动删除（即使无消费者也保留）
|
||
)
|
||
|
||
# 5. 定义消息处理逻辑（核心：根据队列类型处理对应级别消息）
|
||
async def on_message_received(message: aio_pika.IncomingMessage):
|
||
# async with 上下文：自动完成消息确认（处理完后告知 RabbitMQ 删除消息）
|
||
# 若处理过程中崩溃，消息会重新回到队列，避免丢失
|
||
async with message.process():
|
||
# 解码消息体（生产者用 utf-8 编码，此处对应解码）
|
||
message_content = message.body.decode("utf-8")
|
||
# Print key information (for debugging and log tracking)
|
||
print(f"[{consumer_label} Consumer] Received message:")
|
||
print(f" Queue name: {queue_name}")
|
||
print(f" Message content: {message_content}")
|
||
print(f" Message routing key: {message.routing_key}") # Verify routing key match
|
||
print(f" Processing time: {asyncio.get_running_loop().time():.2f}s\n")
|
||
|
||
# 模拟不同级别消息的处理耗时（业务场景可替换为实际逻辑）
|
||
if "error" in queue_name:
|
||
# 错误消息：可能需要重试、告警，耗时更长
|
||
await asyncio.sleep(2)
|
||
elif "warning" in queue_name:
|
||
# 警告消息：可能需要记录日志、轻量处理
|
||
await asyncio.sleep(1)
|
||
elif "info" in queue_name:
|
||
# 信息/调试消息：快速处理，仅记录
|
||
await asyncio.sleep(0.5)
|
||
|
||
# 6. 启动队列监听：将消息处理函数绑定到队列
|
||
consumer_tag = f"direct_{consumer_label.lower().replace(' ', '_')}_{queue_name}"
|
||
await queue.consume(on_message_received, consumer_tag=consumer_tag)
|
||
# Print startup log, indicating consumer is ready
|
||
print(f"[{consumer_label} Consumer] Started, listening to queue: {queue_name} (tag: {consumer_tag})\n")
|
||
|
||
# 7. 保持消费者运行（无限期阻塞，直到手动停止程序）
|
||
# 若不阻塞，协程会立即结束，消费者会断开连接
|
||
await asyncio.Future()
|
||
|
||
|
||
async def start_all_direct_consumers(queue_prefix="demo.direct.queue-"):
|
||
"""
|
||
启动所有 Direct Exchange 对应的消费者
|
||
与生产者 setup_direct_exchange 中的 queue_bindings 完全对应
|
||
"""
|
||
# Define list of consumers to start (queue name + consumer label)
|
||
consumers = [
|
||
# Error queue: handles messages with routing key "error"
|
||
direct_consumer(f"{queue_prefix}error", "Error Level"),
|
||
# Warning queue: handles messages with routing key "warning"
|
||
direct_consumer(f"{queue_prefix}warning", "Warning Level"),
|
||
# Info queue: handles messages with routing keys "info" and "debug"
|
||
direct_consumer(f"{queue_prefix}info", "Info/Debug Level")
|
||
]
|
||
|
||
# 同时启动所有消费者（并发运行，互不阻塞）
|
||
await asyncio.gather(*consumers)
|
||
|
||
|
||
if __name__ == "__main__":
|
||
# 启动所有消费者（需先执行 setup_direct_exchange 初始化队列）
|
||
asyncio.run(start_all_direct_consumers())
|