Quartz是实现了序列化接口的,包括接口,所以可以使用标准方式序列化到数据库。
而Spring2.5.6在集成Quartz时却未能考虑持久化问题。
Spring对JobDetail进行了封装,却未实现序列化接口,所以持久化的时候会产生NotSerializable问题,这也是网上一直在那边叫嚣为什么不能持久化到数据库问题,哥今天看了下Spring源码,发现Spring对Quartz持久化的问题.
1. 不知道Spring未来会不会对持久化的支持,不过我们可以有如下解决方案,比如改写
Spring的代码,实现序列化接口.
2. 不使用Spring的Fatory,自己实现任务的初始化.
既然Spring不支持持久化,那么持久化任务还是自己编写实现吧,否则每次都需要打包发布,麻烦,自己编写的类与Quartz完全兼容.
注意:为什么Spring不支持外配置任务,可能也是考虑到这方面问题所以才不提供这些任务的执行化支持.[配置文件配置与数据库配置重复]
直接使用Quartz是支持序列化功能,比如直接使用页面配置Quartz界面,设置任务执行时间等属性。
通过配置实现的是不应该初始化到数据库,否则直接在数据库中配置了。不过也是可以配置的,通过改写JobDetailBean.代码如下:
-
package org.frame.auth.service; import java.util.Map; import org.quartz.Job; import org.quartz.JobDetail; import org.quartz.Scheduler; import org.springframework.beans.factory.BeanNameAware; import org.springframework.beans.factory.InitializingBean; import org.springframework.scheduling.quartz.DelegatingJob; import org.springframework.scheduling.quartz.SchedulerFactoryBean; public class PersistentJobDetailBean extends JobDetail implements BeanNameAware, InitializingBean { private static final long serialVersionUID = -4389885435844732405L; private Class actualJobClass; private String beanName; /** * Overridden to support any job class, to allow a custom JobFactory * to adapt the given job class to the Quartz Job interface. * @see SchedulerFactoryBean#setJobFactory */ public void setJobClass(Class jobClass) { if (jobClass != null && !Job.class.isAssignableFrom(jobClass)) { super.setJobClass(DelegatingJob.class); this.actualJobClass = jobClass; } else { super.setJobClass(jobClass); } } /** * Overridden to support any job class, to allow a custom JobFactory * to adapt the given job class to the Quartz Job interface. */ public Class getJobClass() { return (this.actualJobClass != null ? this.actualJobClass : super.getJobClass()); } /** * Register objects in the JobDataMap via a given Map. * <p>These objects will be available to this Job only, * in contrast to objects in the SchedulerContext. * <p>Note: When using persistent Jobs whose JobDetail will be kept in the * database, do not put Spring-managed beans or an ApplicationContext * reference into the JobDataMap but rather into the SchedulerContext. * @param jobDataAsMap Map with String keys and any objects as values * (for example Spring-managed beans) * @see SchedulerFactoryBean#setSchedulerContextAsMap */ public void setJobDataAsMap(Map jobDataAsMap) { getJobDataMap().putAll(jobDataAsMap); } /** * Set a list of JobListener names for this job, referring to * non-global JobListeners registered with the Scheduler. * <p>A JobListener name always refers to the name returned * by the JobListener implementation. * @see SchedulerFactoryBean#setJobListeners * @see org.quartz.JobListener#getName */ public void setJobListenerNames(String[] names) { for (int i = 0; i < names.length; i++) { addJobListener(names[i]); } } public void setBeanName(String beanName) { this.beanName = beanName; } public void afterPropertiesSet() { if (getName() == null) { setName(this.beanName); } if (getGroup() == null) { setGroup(Scheduler.DEFAULT_GROUP); } } }
这里把Spring的ApplicationContext去掉了,因为这个属性没有实现序列化接口。其他配置与原告一致:
-
<?xml version="1.0" encoding="UTF-8"?> <!DOCTYPE beans PUBLIC "-//SPRING//DTD BEAN//EN" " http://www.springframework.org/dtd/spring-beans.dtd "> <beans default-autowire="byName"> <bean id="dataSource" class="org.springframework.jdbc.datasource.DriverManagerDataSource" destroy-method="close"> <property name="driverClassName" value="com.mysql.jdbc.Driver"/> <property name="url" > <value><![CDATA[jdbc:mysql://localhost:3306/txl?connectTimeout=1000&useUnicode=true&characterEncoding=utf-8]]></value> </property> <property name="username" value="root"/> <property name="password" value=""/> </bean> <bean id="jobDetail" class = "org.frame.auth.service.PersistentJobDetailBean"> <property name="jobClass" value="org.frame.auth.service.PersistentJob"></property> </bean> <!-- <bean id="trigger" class="org.springframework.scheduling.quartz.SimpleTriggerBean" >--> <!-- <property name="jobDetail" ref="jobDetail"></property>--> <!-- <property name="startDelay" value="1000"></property>--> <!-- <property name="repeatInterval" value="3000"></property>--> <!-- <property name="jobDataAsMap">--> <!-- <map>--> <!-- <entry key="message" value="this is trigger"></entry>--> <!-- </map>--> <!-- </property>--> <!-- </bean>--> <bean id="cronTrigger" class="org.springframework.scheduling.quartz.CronTriggerBean" > <property name="jobDetail" ref="jobDetail"/> <property name="cronExpression"> <value>0/10 * * * * ?</value> </property> </bean> <bean id="schedulerFactory" class="org.springframework.scheduling.quartz.SchedulerFactoryBean"> <property name="dataSource" ref="dataSource"></property> <property name="applicationContextSchedulerContextKey" value="applicationContextKey" /> <property name="configLocation" value="classpath:quartz.properties"/> </bean> </beans>
org.frame.auth.service.PersistentJob这个类很简单,如下:
-
package org.frame.auth.service; import org.quartz.Job; import org.quartz.JobExecutionContext; import org.quartz.JobExecutionException; public class PersistentJob implements Job { @Override public void execute(JobExecutionContext context) throws JobExecutionException { System.out.println("spring quartz!"); } }
有人可能会说,你这种任务调度持久化就没有意义了,是的,一般持久化到数据库的代码如下:
-
package org.frame.auth.service; import java.util.Map; import org.quartz.JobExecutionContext; import org.quartz.JobExecutionException; import org.quartz.StatefulJob; public class PersistentJob implements StatefulJob { @Override public void execute(JobExecutionContext context) throws JobExecutionException { // TODO Auto-generated method stub Map map = context.getJobDetail().getJobDataMap(); System.out.println("["+context.getJobDetail().getName()+"]"+map.get("message")); map.put("message", "updated Message"); } }
这样的话,信息message就会持久化到数据库中了.可以建立系统的连锁调度,这根据你的业务需求了.
在Spring中配置的任务通过我这种修改是可以运行,不过每次运行都需要把原先的任务删除,否则会提示任务已经存在,Quartz的优势是就算服务器停止,下次重启能够恢复原先的任务并继续执行.