Spring Batch 作业监听器（Job Listener）
每个作业都有生命周期。Spring Batch 允许在生命周期的不同阶段通过提供的钩子（ Hook ）来添加额外的逻辑。
JobExecutionListener 接口提供了两个回调方法：beforeJob() 和 afterJob() 。
创建 Spring Batch 监听器有两种方式：
- 实现上面的
JobExecutionListener接口； - 使用
@BeforeJob和@AfterJob注解，并通过JobListenerFactoryBean提供的getListener()将其代理为JobExecutionListener。
1. 实现 JobExecutionListener 接口
继承 JobExecutionListener 并实现其 beforeJob() 和 afterJob() 方法即可。
参数均为 JobExecution 实例，可以通过其获取作业的执行信息。
示例代码如下：
java
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.batch.core.JobExecution;
import org.springframework.batch.core.JobExecutionListener;
import org.springframework.stereotype.Component;
/**
* @author 佳佳
*/
@Component
public class JobLoggerListener implements JobExecutionListener {
private static final Logger LOGGER = LoggerFactory.getLogger(JobLoggerListener.class);
@Override
public void beforeJob(JobExecution jobExecution) {
LOGGER.info("{} is beginning execution",
jobExecution.getJobInstance().getJobName());
}
@Override
public void afterJob(JobExecution jobExecution) {
LOGGER.info("{} has completed with the status {}",
jobExecution.getJobInstance().getJobName(),
jobExecution.getStatus());
}
}调用 JobBuilder 的 listener() 方法添加监听器（可以调用多次以添加多个监听器）到作业：
java
@Bean
public Job listenerJob(@Qualifier("step1") Step step1,
JobLoggerListener jobLoggerListener) {
return this.jobBuilderFactory.get("JobListenerJob")
.start(step1)
.listener(jobLoggerListener)
.incrementer(new RunIdIncrementer())
.build();
}运行日志如下：
java
2022-11-30 11:45:17.436 INFO 14896 --- [ main] o.s.b.c.l.support.SimpleJobLauncher : Job: [SimpleJob: [name=JobListenerJob]] launched with the following parameters: [{run.id=1}]
2022-11-30 11:45:17.563 INFO 14896 --- [ main] m.l.b.c.job.listener.JobLoggerListener : JobListenerJob is beginning execution
2022-11-30 11:45:17.709 INFO 14896 --- [ main] o.s.batch.core.job.SimpleStepHandler : Executing step: [step1]
Hello, World!
2022-11-30 11:45:17.893 INFO 14896 --- [ main] o.s.batch.core.step.AbstractStep : Step: [step1] executed in 184ms
2022-11-30 11:45:17.980 INFO 14896 --- [ main] m.l.b.c.job.listener.JobLoggerListener : JobListenerJob has completed with the status COMPLETED
2022-11-30 11:45:18.051 INFO 14896 --- [ main] o.s.b.c.l.support.SimpleJobLauncher : Job: [SimpleJob: [name=JobListenerJob]] completed with the following parameters: [{run.id=1}] and the following status: [COMPLETED] in 486ms可以看到 JobLoggerListener 在作业执行前后分别打印了一条日志。
2. 使用 @BeforeJob 和 @AfterJob 注解
使用注解方式的代码示例如下：
java
import org.slf4j.Logger;
import org.slf4j.LoggerFactory;
import org.springframework.batch.core.JobExecution;
import org.springframework.batch.core.annotation.AfterJob;
import org.springframework.batch.core.annotation.BeforeJob;
import org.springframework.stereotype.Component;
/**
* @author 佳佳
*/
@Component
public class JobLoggerAnnotationListener {
private static final Logger LOGGER = LoggerFactory.getLogger(JobLoggerAnnotationListener.class);
@BeforeJob
public void beforeJob(JobExecution jobExecution) {
LOGGER.info("{} is beginning execution",
jobExecution.getJobInstance().getJobName());
}
@AfterJob
public void afterJob(JobExecution jobExecution) {
LOGGER.info("{} has completed with the status {}",
jobExecution.getJobInstance().getJobName(),
jobExecution.getStatus());
}
}此时添加监听器的方式有两种：
通过
JobListenerFactoryBean.getListener()方法将其代理为JobExecutionListener。java@Bean public Job listenerJob(@Qualifier("step1") Step step1, JobLoggerAnnotationListener jobLoggerAnnotationListener) { return this.jobBuilderFactory.get("JobListenerJob") .start(step1) .listener(JobListenerFactoryBean.getListener(jobLoggerAnnotationListener)) .incrementer(new RunIdIncrementer()) .build(); }直接将监听器实例传给
listener(Object listener)重载方法。实际上是将
JobListenerFactoryBean.getListener()处理封装到了这个listener()方法里。相对于上一种，这种方式比较简单，而且这种方式添加了是否包含
@BeforeJob和@AfterJob注解的校验 -- 至少要有一个注解才会添加。java@Bean public Job listenerJob(@Qualifier("step1") Step step1, JobLoggerAnnotationListener jobLoggerAnnotationListener) { return this.jobBuilderFactory.get("JobListenerJob") .start(step1) .listener(jobLoggerAnnotationListener) .incrementer(new RunIdIncrementer()) .build(); }
运行结果如下：
java
2022-11-30 16:01:14.819 INFO 22460 --- [ main] o.s.b.c.l.support.SimpleJobLauncher : Job: [SimpleJob: [name=JobListenerJob]] launched with the following parameters: [{run.id=7}]
2022-11-30 16:01:14.937 INFO 22460 --- [ main] m.l.b.c.j.l.JobLoggerAnnotationListener : JobListenerJob is beginning execution
2022-11-30 16:01:15.076 INFO 22460 --- [ main] o.s.batch.core.job.SimpleStepHandler : Executing step: [step1]
Hello, World!
2022-11-30 16:01:15.221 INFO 22460 --- [ main] o.s.batch.core.step.AbstractStep : Step: [step1] executed in 144ms
2022-11-30 16:01:15.271 INFO 22460 --- [ main] m.l.b.c.j.l.JobLoggerAnnotationListener : JobListenerJob has completed with the status COMPLETED
2022-11-30 16:01:15.310 INFO 22460 --- [ main] o.s.b.c.l.support.SimpleJobLauncher : Job: [SimpleJob: [name=JobListenerJob]] completed with the following parameters: [{run.id=7}] and the following status: [COMPLETED] in 389ms执行顺序
可以通过继承 org.springframework.core.Ordered 接口或添加 @Order 注解来指定监听器执行的顺序。
- 同时使用
Ordered接口和@Order注解时，Ordered接口的优先级高。 - 所有未指定顺序的监听器始终排在最后执行。
- 使用上面
@BeforeJob、@AfterJob注解的方式添加监听器时，@Order注解的排序不生效（Ordered接口的方式是可以的）。原因在于由于使用了代理，获取不到被代理类的注解信息。 afterJob()方法的执行顺序和beforeJob()方法是相反的。beforeJob()按照指定的顺序执行，afterJob()按照相反的顺序执行（类似于堆栈 LIFO ）。
附 1. JobExecutionListener 接口源码
下面是在 spring-batch-core:4.3.7 中 JobExecutionListener 接口的源码：
java
/*
* Copyright 2006-2013 the original author or authors.
*
* Licensed under the Apache License, Version 2.0 (the "License");
* you may not use this file except in compliance with the License.
* You may obtain a copy of the License at
*
* https://www.apache.org/licenses/LICENSE-2.0
*
* Unless required by applicable law or agreed to in writing, software
* distributed under the License is distributed on an "AS IS" BASIS,
* WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
* See the License for the specific language governing permissions and
* limitations under the License.
*/
package org.springframework.batch.core;
/**
* Provide callbacks at specific points in the lifecycle of a {@link Job}.
* Implementations can be stateful if they are careful to either ensure thread
* safety, or to use one instance of a listener per job, assuming that job
* instances themselves are not used by more than one thread.
*
* @author Dave Syer
*
*/
public interface JobExecutionListener {
/**
* Callback before a job executes.
*
* @param jobExecution the current {@link JobExecution}
*/
void beforeJob(JobExecution jobExecution);
/**
* Callback after completion of a job. Called after both both successful and
* failed executions. To perform logic on a particular status, use
* "if (jobExecution.getStatus() == BatchStatus.X)".
*
* @param jobExecution the current {@link JobExecution}
*/
void afterJob(JobExecution jobExecution);
}参考
[1]:《Spring Batch 权威指南》 -- [美] 迈克尔·T.米内拉（Michael，T.，Minella）著