Spring注解驱动之ApplicationListener异步处理事件说明

2024-04-02 19:04:59 858人浏览安东尼

Python 官方文档：入门教程 => 点击学习

摘要

目录概述示例自定义事件多波器测试用例ApplicationListener异步执行源码分析概述之前我们讲过简单使用ApplicationListener发布事件，处理事件，但是发现

概述

之前我们讲过简单使用ApplicationListener发布事件，处理事件，但是发现是同一个线程发送事件并自己处理事件的。

下面我们就来说下如何使用自定义的线程池来异步处理接收的事件。

示例

实现一个ApplicationListener用于处理事件

package com.atguigu.ext;

import org.springframework.context.ApplicationListener;
import org.springframework.context.PayloadApplicationEvent;
import org.springframework.stereotype.Component;

@Component
public class MyApplicationListener implements ApplicationListener<PayloadApplicationEvent> {

    public void onApplicationEvent(PayloadApplicationEvent applicationEvent) {
        System.out.println("exe thread start:" + Thread.currentThread().getName() + ", time:" + System.currentTimeMillis());
        System.out.println("收到事件：" + applicationEvent);
        System.out.println(applicationEvent.getPayload());
        try {
            Thread.sleep(3000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        System.out.println("exe thread end:" + Thread.currentThread().getName() + ", time:" + System.currentTimeMillis());
    }
}

自定义事件多波器

package com.atguigu.config;

import org.springframework.context.annotation.Bean;
import org.springframework.context.annotation.ComponentScan;
import org.springframework.context.annotation.Configuration;
import org.springframework.context.event.SimpleApplicationEventMulticaster;

import java.util.concurrent.BlockingQueue;
import java.util.concurrent.LinkedBlockingDeque;
import java.util.concurrent.ThreadPoolExecutor;
import java.util.concurrent.TimeUnit;

@Configuration
@ComponentScan("com.atguigu.ext")
public class ExtConfig {

    @Bean
    public SimpleApplicationEventMulticaster applicationEventMulticaster() {
        SimpleApplicationEventMulticaster simpleApplicationEventMulticaster = new SimpleApplicationEventMulticaster();
        BlockingQueue<Runnable> blockingQueue = new LinkedBlockingDeque<>(1000);
        ThreadPoolExecutor threadPoolExecutor = new ThreadPoolExecutor(5, 5, 10, TimeUnit.SECONDS, blockingQueue);
        simpleApplicationEventMulticaster.setTaskExecutor(threadPoolExecutor);
        return simpleApplicationEventMulticaster;
    }
}

之前看源码可以发现，在容器创建的refresh方法中的initApplicationEventMulticaster()方法执行时，先从容器中获取name为applicationEventMulticaster的组件，如果获取不到就好创建一个默认的applicationEventMulticaster组件，该组件默认是不会设置taskExecutor任务执行器的，所以这里我们自定义一个设置了TaskExecutor的多波器，当执行initApplicationEventMulticaster方法从beanFactory中获取applicationEventMulticaster组件时，走getBean逻辑。

BeanFactory.getBean()逻辑是先从容器查看是否有该组件，如果没有获取该组件的定义，如果有定义就会创建一个组件返回并把组件保存到容器中。

测试用例

package com.atguigu;

import com.atguigu.config.ExtConfig;
import org.springframework.context.annotation.AnnotationConfigApplicationContext;


public class AnnotationTest {
    public static void main(String[] args) {
        final AnnotationConfigApplicationContext applicationContext
                = new AnnotationConfigApplicationContext(ExtConfig.class);
        System.out.println("main thread start:" + Thread.currentThread().getName() + ", time:" + System.currentTimeMillis());
        applicationContext.publishEvent("发送事件");
        System.out.println("main thread end:" + Thread.currentThread().getName() + ", time:" + System.currentTimeMillis());
        applicationContext.close();
    }
}

测试结果

main thread start:main, time:1663499481185
main thread end:main, time:1663499481188
exe thread start:pool-1-thread-1, time:1663499481188
收到事件：org.springframework.context.PayloadApplicationEvent[source=org.springframework.context.annotation.AnnotationConfigApplicationContext@5ebec15: startup date [Sun Sep 18 19:11:20 CST 2022]; root of context hierarchy]
发送事件
九月 18, 2022 7:11:21 下午 org.springframework.context.annotation.AnnotationConfigApplicationContext doClose
信息: Closing org.springframework.context.annotation.AnnotationConfigApplicationContext@5ebec15: startup date [Sun Sep 18 19:11:20 CST 2022]; root of context hierarchy
exe thread end:pool-1-thread-1, time:1663499484198

通过测试结果可以看出，main线程很快就返回了，而实际处理事件的线程是pool-1-thread-1，等待了3s多才返回。