SAKA'S BLOG

Java中的char[]的输出为什么不是内存地址

Java中共有八种基本数据类型:byte,int,short,long,float,double,char,boolean。
计算机中的基础数据单位是bit, 1byte=8bit。

数据类型 存储大小 举例 注释 包装类
byte 1byte 3 字节 Byte
int 4byte 4 整数 Integer
short 2bytes 5 短整数 Short
long 8bytes 6 长整数 Long
float 4bytes 1.3 单精度浮点型 Float
double 8bytes 1.2 双精度浮点型 Double
char 2bytes ‘a’ 字符 Char
boolean 1bit true 布尔值 Boolean

这8种基本数据类型很简单,在示例中应用来看一下:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
public class Test {

public static void main(String[] args){
System.out.println("8种基本数据类型");
int a=5;
System.out.println(a);
char b='z';
System.out.println(b);
boolean d=false;
System.out.println(d);
byte e=3;
System.out.println(e);
short f=4;
System.out.println(f);
long g=32000000;
System.out.println(g);
float h=5;
System.out.println(h);
double i=6;
System.out.println(i);
}
}

一段简单的输出代码,看看打印结果:

1
2
3
4
5
6
7
8
9
8种基本数据类型
5
z
false
3
4
32000000
5.0
6.0

可以看到输出结果是没有问题的。

基本数据类型和对象引用

基本数据类型会一直在栈中创建,当声明基本类型时,不需要new。
int a=1;
栈的读取速度比堆快。基本类型一旦被声明,java将在栈上直接存储它,所以基本类型的变量表示的是数据本身。

假如调用基本类型的包装类来创建对象,那么将会在堆中创建。
Employee a=new Emploee(1.4);

  1. 等号右侧的new Employee()。这个new是在内存的堆中为对象开辟控件,保存对象的数据和方法。
  2. 等号左侧 Emploee a。a指代的是Emploee的一个对象,称为对象引用,这个对象引用是在栈中创建的。实际上a不是对象本身,它用来指向一个地址。
  3. 赋值=。这个就是把对象的地址赋给a。

此时输出a就是一个内存地址。有兴趣的同学自己试一试。

这个地方说明一个问题,假如你自定义的对象重写了.toString方法,此处就会显示你的自定义的重写方法的输出值。
在java的基本类型包装类中就重写了这个方法,所以调用print方法时会自动调用它的toString()方法。

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
public class Wrapper {
static class Employee{
static int age;
Employee(int a){
age=a;
}
}

static class Employer{
static int year;
Employer (int y){
year=y;
}

@Override
public String toString() {
return "Employer's year="+year;
}
}
public static void main(String[] args){
Employee e=new Employee(4);
System.out.println("e="+e);
Employer f=new Employer(5);
System.out.println("f="+f);
}
}

在上边的例子中Employee的toString()方法没有被重写,Employer的toString()方法被重写了。来看输出结果:

1
2
e=Wrapper$Employee@1b6d3586
f=Employer's year=5

前者仍然是内存地址,后者是我们重写的方法。

print方法在调用事,假如类中的toString()方法没有被重写,则会电泳String.valueof()方法(后边有讲),假如重写了就会调用toString方法。
所有的包装类(Integer,Boolean等)都已经重写了toString方法,所以不会输出内存地址,而是输出正确的值。
下面的是Double类中的方法:

1
2
3
4
private final double value;
public String toString() {
return toString(value);
}

整形数据类型取值范围

byte占据8位,则其取值范围应该是2的8次方,也就是-128~127,超过这个区间就会报错,例如:

byte a=128;
在编译器中会报错,提示不能将int转换为byte,因为128已经超出byte的范围了。

同样可以推得其他值的取值范围。

基本类型的数组输出值

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
public class TestOne {
public static void main(String[] args) {
int a=127;
System.out.println(a);
int[] b=new int[]{1,2,3};
System.out.println(b);
int[] c=new int[100];
System.out.println(c);
int[] d={1,2,3};
System.out.println(d);
boolean e=false;
System.out.println(e);
boolean[] f={false,false,true};
System.out.println(f);
char g='a';
System.out.println(g);
char[] h={'a','b','c'};
System.out.println(h);
char[] i=new char[]{'a','b','c'};
System.out.println(i);
float j=1.2f;
System.out.println(j);
float[] k={1.2f,1.3f,1.4f};
System.out.println(k);
}
}

看一下打印的结果:

1
2
3
4
5
6
7
8
9
10
11
127
[I@15db9742
[I@6d06d69c
[I@7852e922
false
[Z@4e25154f
a
abc
abc
1.2
[F@70dea4e

可以看到,在结果中,所有的基本类型都可以打印出来,数组类型只能打印出char数组,其他的都是内存地址。

来看一下源码,在print函数中

1
2
3
public void print(char c) {
write(String.valueOf(c));
}

这个char被转换为了String类型,然后进行wirte方法:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
private void write(String s) {
try {
synchronized (this) {
ensureOpen();
textOut.write(s);
textOut.flushBuffer();
charOut.flushBuffer();
if (autoFlush && (s.indexOf('\n') >= 0))
out.flush();
}
}
catch (InterruptedIOException x) {
Thread.currentThread().interrupt();
}
catch (IOException x) {
trouble = true;
}
}

这里会立即发送缓冲流输出。
对于所有的基础类型都会打印出具体的值,这个没有问题,但是对于数组为什么只有char的数组类型打印出了正确的结果而没有输出内存地址?
带着这个问题我们来了解一下:
对于int型数组,java调用的是下面的方法:

1
2
3
4
5
6
7
public void println(Object x) {
String s = String.valueOf(x);
synchronized (this) {
print(s);
newLine();
}
}

此处数组被认为是Object类型,调用的是

1
2
3
public static String valueOf(Object obj) {
return (obj == null) ? "null" : obj.toString();
}

此处的三目表达式用来判空,然后看一下obj.toString()方法:

1
2
3
public String toString() {
return getClass().getName() + "@" + Integer.toHexString(hashCode());
}

相信看到此处应该可以看出来为什么输出会是[I@1b6d3586了,I代表的类的名称。

那么对于char数组类型的调用呢,次数室友玄机的:

1
2
3
4
5
6
public void println(char x[]) {
synchronized (this) {
print(x);
newLine();
}
}

此处调用的是println(char x[])这个函数,那么这个char x[]是个什么鬼呢?
其实就是java中的数组初始化,相当于char[] x。

然后看看print(x)函数:

1
2
3
public void print(char s[]) {
write(s);
}

最后是write()函数:

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
private void write(char buf[]) {
try {
synchronized (this) {
ensureOpen();
textOut.write(buf);
textOut.flushBuffer();
charOut.flushBuffer();
if (autoFlush) {
for (int i = 0; i < buf.length; i++)
if (buf[i] == '\n')
out.flush();
}
}
}
catch (InterruptedIOException x) {
Thread.currentThread().interrupt();
}
catch (IOException x) {
trouble = true;
}
}

到了这大家知道为什么会有区别了么,因为其他类型的数组都被认为是Object类型了,所以会输出内存地址。而char[]调用的方法是输出char这个数组中的每一个值,所以不是内存地址了。