java中多线程 - 多线程中的基本方法

介绍一下线程中基本的方法使用

线程睡眠sleep()

Thread.sleep(毫秒);我们可以通过sleep方法设置让线程睡眠。可以看到sleep是个静态方法

public static native void sleep(long var0) throws InterruptedException;


try {
System.out.println(new Date().getSeconds());
Thread.sleep(5000);
System.out.println(new Date().getSeconds());
} catch (InterruptedException e) {
e.printStackTrace();
}

setDaemon守护线程

非守护线程停止，那么守护线程自动退出

public static void main(String[] args) {
Thread thread1 = new Thread() {
@Override
public void run() {
super.run();
for(int i = 0; i < 5; i ++) {
System.out.println("非守护线程");
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for(int i = 0; i < 200; i ++) {
System.out.println("守护线程");
}
}
}; thread2.setDaemon(true);
thread1.start();
thread2.start();
}

可以很明显的看到thread2本应该执行200次输出，但是这里只输出了几行。因为当thread1执行完毕后，thread2作为守护线程就自动停止了。

多线程join

如果执行了join方法，那么停止当前线程，先跑执行了join()的线程。相当于插队执行。如下，在执行thread2线程的时候，如果i==20的时候，则让thread1插队先执行

public static void main(String[] args) {
final Thread thread1 = new Thread() {
@Override
public void run() {
super.run();
for(int i = 0; i < 500; i ++) {
System.out.println("thread1---" + i);
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for(int i = 0; i < 200; i ++) {
if (i == 20) {
try {
//插队执行
thread1.join();
} catch (InterruptedException e) {
e.printStackTrace();
}
}
System.out.println(i);
}
}
};
thread1.start();
thread2.start();
}


join()方法也可以传参数long 毫秒 join(毫秒)

表示让执行join的线程，插队执行XXX毫秒，过了时间后，两个线程交替执行

public static void main(String[] args) {
final Thread thread1 = new Thread() {
@Override
public void run() {
super.run();
for(int i = 0; i < 500; i ++) {
System.out.println("thread1---" + i);
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for(int i = 0; i < 200; i ++) {
if (i == 20) {
try {
//插队执行１毫秒
thread1.join(1);
} catch (InterruptedException e) {
e.printStackTrace();
}
}
System.out.println(i);
}
}
};
thread1.start();
thread2.start();
}


yeild 礼让线程

yeild会让出cpu,让其他线程执行

public static void main(String[] args) {
final Thread thread1 = new Thread() {
@Override
public void run() {
super.run();
for(int i = 0; i < 500; i ++) {
System.out.println( getName() + "---" + i);
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for(int i = 0; i < 200; i ++) {
if (i % 5 == 0) {
Thread.yield();
}
System.out.println(getName() + "---" + i);
}
}
};
thread1.start();
thread2.start();
}


setPriority给线程设置优先级

默认优先级是5 最小１，最大10

越大优先级越高


public static void main(String[] args) {
final Thread thread1 = new Thread() {
@Override
public void run() {
super.run();
for(int i = 0; i < 500; i ++) {
System.out.println( getName() + "---" + i);
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for(int i = 0; i < 500; i ++) { System.out.println(getName() + "---" + i);
}
}
};
//设置最大的线程优先级最大为10
thread1.setPriority(Thread.MIN_PRIORITY);
//设置最小的线程优先级，最小为１
thread2.setPriority(Thread.MAX_PRIORITY);
thread1.start();
thread2.start();
}


synchronized

同步代码块

当多线程并发，多段代码同时执行的时候。希望在执行其中代码的时候，cpu不切换线程

不用synchronized的情况

我们来看一下不用synchronized的情况会发生什么

public class ThreadSynchronied { public static void main(String[] args) {
final Say say = new Say(); Thread thread1 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say();
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say1();
}
}
};
//设置最大的线程优先级最大为10
thread1.setPriority(Thread.MIN_PRIORITY);
//设置最小的线程优先级，最小为１
thread2.setPriority(Thread.MAX_PRIORITY);
thread1.start();
thread2.start();
}
}class Say {
void say() {
System.out.print("s ");
System.out.print("a ");
System.out.print("y ");
System.out.print("h ");
System.out.print("e ");
System.out.print("l ");
System.out.print("l ");
System.out.println("o");
} void say1() {
System.out.print("1 ");
System.out.print("2 ");
System.out.print("3 ");
System.out.print("4 ");
System.out.print("5 ");
System.out.print("6 ");
System.out.print("7 ");
System.out.println("8");
}
}


我们发现有些输出并没有打印全，在执行线程thread1的过程中，cpu被thread2抢占。这种情况下，肯定是不符合我们的业务逻辑的。所以我们要保证线程执行了一个完整的方法后，cpu才会被其他线程抢占

使用synchronized

public class ThreadSynchronied { public static void main(String[] args) {
final Say say = new Say(); Thread thread1 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say();
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say1();
}
}
};
//设置最大的线程优先级最大为10
thread1.setPriority(Thread.MIN_PRIORITY);
//设置最小的线程优先级，最小为１
thread2.setPriority(Thread.MAX_PRIORITY);
thread1.start();
thread2.start();
}
}class Say {
String s = "hahaah"; void say() {
synchronized (s) {
System.out.print("s ");
System.out.print("a ");
System.out.print("y ");
System.out.print("h ");
System.out.print("e ");
System.out.print("l ");
System.out.print("l ");
System.out.println("o");
}
} void say1() {
synchronized (s) {
System.out.print("1 ");
System.out.print("2 ");
System.out.print("3 ");
System.out.print("4 ");
System.out.print("5 ");
System.out.print("6 ");
System.out.print("7 ");
System.out.println("8");
}
}
}

使用synchronized同步代码块后，就发现不会出现上述情况了

同步方法

public class ThreadSynchroniedMethod { public static void main(String[] args) {
final Say say = new Say(); Thread thread1 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say();
}
}
}; Thread thread2 = new Thread() {
@Override
public void run() {
for (int i = 0 ; i < 10000 ; i ++) {
say.say1();
}
}
};
//设置最大的线程优先级最大为10
thread1.setPriority(Thread.MIN_PRIORITY);
//设置最小的线程优先级，最小为１
thread2.setPriority(Thread.MAX_PRIORITY);
thread1.start();
thread2.start();
}
}class Say {
//在方法上加锁
static synchronized void say() {
System.out.print("s ");
System.out.print("a ");
System.out.print("y ");
System.out.print("h ");
System.out.print("e ");
System.out.print("l ");
System.out.print("l ");
System.out.println("o"); } static void say1() {
synchronized (Say.class) {
System.out.print("1 ");
System.out.print("2 ");
System.out.print("3 ");
System.out.print("4 ");
System.out.print("5 ");
System.out.print("6 ");
System.out.print("7 ");
System.out.println("8");
}
}
}

同步方法指的就是在方法上加锁

静态同步方法的所对象是该类的字节码对象

非静态的同步方法锁对象是this

多个线程使用同一资源锁，容易造成死锁

什么是死锁?

死锁是指两个或两个以上的进程在执行过程中，由于竞争资源或者由于彼此通信而造成的一种阻塞的现象，若无外力作用，它们都将无法推进下去。此时称系统处于死锁状态或系统产生了死锁，这些永远在互相等待的进程称为死锁进程。


线程安全类

Vector

StringBuffer

HashTable

线程不安全

ArrayList

StringBuilder

HashSet

java.util.Collections中有synchronizedList等方法，支持我们把线程不安全的集合转成线程安全的

学习笔记

多次启动一个线程是非法的