【Java多线程】之十：Timer and TimerTask

java.util.Timer is a utility class that can be used to schedule a thread to be executed at certain time in future. Java Timer class can be used to schedule a task to be run one-time or to be run at regular intervals.

java.util.TimerTask is an abstract class that implements Runnable interface and we need to extend this class to create our own TimerTask that can be scheduled using java Timer class.

Timer class is thread safe and multiple threads can share a single Timer object without need for external synchronization. Timer class uses java.util.TaskQueue to add tasks at given regular interval and at any time there can be only one thread running the TimerTask, for example if you are creating a Timer to run every 10 seconds but single thread execution takes 20 seconds, then Timer object will keep adding tasks to the queue and as soon as one thread is finished, it will notify the queue and another thread will start executing.

Timer class uses Object wait and notify methods to schedule the tasks.

Here is an example of Timer and TimerTask implementation.

MyTimerTask.java

package com.journaldev.threads;

import java.util.Date;
import java.util.Timer;
import java.util.TimerTask;

public class MyTimerTask extends TimerTask {

    @Override
    public void run() {
        System.out.println("Timer task started at:"+new Date());
        completeTask();
        System.out.println("Timer task finished at:"+new Date());
    }

    private void completeTask() {
        try {
            //assuming it takes 20 secs to complete the task
            Thread.sleep(20000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
    }

    public static void main(String args[]){
        TimerTask timerTask = new MyTimerTask();
        //running timer task as daemon thread
        Timer timer = new Timer(true);
        timer.scheduleAtFixedRate(timerTask, 0, 10*1000);
        System.out.println("TimerTask started");
        //cancel after sometime
        try {
            Thread.sleep(120000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
        timer.cancel();
        System.out.println("TimerTask cancelled");
        try {
            Thread.sleep(30000);
        } catch (InterruptedException e) {
            e.printStackTrace();
        }
    }

}

Notice that one thread execution will take 20 seconds but Timer object is scheduled to run the task every 10 seconds. Here is the output of the program:

TimerTask started
Timer task started at:Sun Aug 09 16:27:28 CST 2015
Timer task finished at:Sun Aug 09 16:27:48 CST 2015
Timer task started at:Sun Aug 09 16:27:48 CST 2015
Timer task finished at:Sun Aug 09 16:28:08 CST 2015
Timer task started at:Sun Aug 09 16:28:08 CST 2015
Timer task finished at:Sun Aug 09 16:28:28 CST 2015
Timer task started at:Sun Aug 09 16:28:28 CST 2015
Timer task finished at:Sun Aug 09 16:28:48 CST 2015
Timer task started at:Sun Aug 09 16:28:48 CST 2015
Timer task finished at:Sun Aug 09 16:29:08 CST 2015
Timer task started at:Sun Aug 09 16:29:08 CST 2015
TimerTask cancelled
Timer task finished at:Sun Aug 09 16:29:28 CST 2015

The output confirms that if a task is already executing, Timer will wait for it to finish and once finished, it will start again the next task from the queue.

Timer object can be created to run the associated tasks as a daemon thread. Timer cancel() method is used to terminate the timer and discard any scheduled tasks, however it doesn’t interfere with the currently executing task and let it finish. If the timer is run as daemon thread, whether we cancel it or not, it will terminate as soon as all the user threads are finished executing.

Timer class contains several schedule() methods to schedule a task to run once at given date or after some delay. There are several scheduleAtFixedRate() methods to run a task periodically with certain interval.

While scheduling tasks using Timer, you should make sure that time interval is more than normal thread execution, otherwise tasks queue size will keep growing and eventually task will be executing always.