当前位置：网站首页>ByteArrayInputStream class source code analysis

ByteArrayInputStream class source code analysis

2022-07-30 18:01:00 【51CTO】

这是《水煮 JDK 源码》系列的第3篇文章,计划撰写100篇关于JDK源码相关的文章

ByteArrayInputStream 类位于 java.io 包下,继承于 InputStream 类,Represents a byte array input stream,It creates a byte array buffer in memory,Then save all the data read from the input stream in the buffer,其 UML 类图如下：

ByteArrayInputStream 类源码分析_字节数组输入流

1、成员变量

在 ByteArrayInputStream 中定义了4个成员变量,如下：

/** Byte array buffer to hold data */
protected byte buf[];
/** The position index of the next character read from the character array buffer */
protected int pos;
/** The position of the marker in the current stream,初始化时为0 */
protected int mark = 0;
/** The number of bytes valid in the byte array buffer */
protected int count;

     1.
2.
3.
4.
5.
6.
7.
8.

2、构造方法

创建 ByteArrayInputStream There are two main ways to input stream of byte array：

public ByteArrayInputStream(byte buf[]) {
    this.buf = buf;
    // The first character position index to read is 0
    this.pos = 0;
    this.count = buf.length;
}

public ByteArrayInputStream(byte buf[], int offset, int length) {
    this.buf = buf;
    // The location index of the first read is  offset
    this.pos = offset;
    this.count = Math.min(offset + length, buf.length);
    // The marker position is also  offset
    this.mark = offset;
}

     1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.

3、读取字节方法

ByteArrayInputStream Two methods for reading bytes are provided,如下：

public synchronized int read() {
    // If the next character position read is greater than the length of the byte array,直接返回 -1,Indicates that the array has been read at this time
    // After each byte is read,posposition automatically +1
    return (pos < count) ? (buf[pos++] & 0xff) : -1;
}

/** Input a byte array to the stream from off The most after the position starts len length of bytes of data,Read into parameter array b[] 中,What is returned is the actual length of the data read */
public synchronized int read(byte b[], int off, int len) {
    // If the parameter array is null,则直接抛出空指针异常
    if (b == null) {
        throw new NullPointerException();
    } else if (off < 0 || len < 0 || len > b.length - off) {
        // 数组越界检查
        throw new IndexOutOfBoundsException();
    }

    // If there is no more readable data in the input stream,直接返回-1
    if (pos >= count) {
        return -1;
    }

    // The length of data that can be read from the input stream avail
    int avail = count - pos;
    // Length of data if desired to be read from the stream len Greater than the length that can be read in the stream avail,
    // then at most will be read from the stream avail 个字节数据
    if (len > avail) {
        len = avail;
    }
    if (len <= 0) {
        return 0;
    }
    // 使用 System.arraycopy 方法从 buf[] Copy the byte data into the parameter b[] 中
    System.arraycopy(buf, pos, b, off, len);
    // 读取完后,改变 pos 的位置索引
    pos += len;
    return len;
}

     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.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.
37.

上面的两个 read() 方法都使用了 synchronized 关键字,That is, it is represented as a synchronous method,It can be seen from the specific implementation,ByteArrayInputStream The data in can not be read repeatedly,Each read a byte of data,The index of the next readable byte position is automatic+1,If the last element of the byte array has been read,can no longer be read.

4、其他方法

ByteArrayInputStream The class provides some other methods,如下：

public synchronized long skip(long n)：Skip from the input stream n 个字节;
public synchronized int available()：Gets the length in bytes that can also be read from the input stream;
public boolean markSupported()：Determines whether the input stream supports the tagging function,默认返回 true,即 ByteArrayInputStream 是支持的;
public void mark(int readAheadLimit)：Sets the marker position in the input stream to the current readable position;
public synchronized void reset()：Resets the next readable position in the input stream to the marked position;
public void close()：关闭输入流;

下面分别看看这些方法的具体实现.

public synchronized long skip(long n) {
    // Get the length of data that can be read in the current input stream k
    long k = count - pos;
    // Determine the length to be skipped n Is it less than the readable length k
    if (n < k) {
        k = n < 0 ? 0 : n;
    }
	// Adjust the index of the next readable byte position,直接偏移 k 个长度
    pos += k;
    // What is returned is the actual skipped byte data length,不一定等于 n
    return k;
}

public synchronized int available() {
    // Gets the length in bytes that can also be read from the input stream
    return count - pos;
}

public boolean markSupported() {
    // Determines whether the input stream supports the tagging function,默认返回 true
    return true;
}

public void mark(int readAheadLimit) {
    // Sets the marker position in the input stream to the current readable position,其中参数 readAheadLimit 没有任何用途
    mark = pos;
}

public synchronized void reset() {
    // Resets the next readable position in the input stream to the marker position
    pos = mark;
}

public void close() throws IOException {
    // 没有任何代码实现
}

     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.
27.
28.
29.
30.
31.
32.
33.
34.
35.
36.

和 ByteArrayOutputStream 一样,ByteArrayInputStream 类中的 close() methods are also implemented without any code,即使被调用,也没有任何的作用.

5、示例代码

5.1 Simple data read

上面分析了 ByteArrayInputStream 类的相关源码,Let's take a look at it with a simple example ByteArrayInputStream 该如何使用.

package com.magic.test;

import java.io.ByteArrayInputStream;

public class ByteArrayInputStreamTest {

    public static void main(String[] args) {
        byte[] bytes = "01234".getBytes();
        ByteArrayInputStream inputStream = new ByteArrayInputStream(bytes);
        int c;
        
        while ((c = inputStream.read()) != -1) {
            System.out.println(Character.toUpperCase((char) c));
        }
    }
}

     1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.

运行后,输出的结果如下：

5.2 验证不可重复读

Let's verify that the input stream cannot be read repeatedly.

package com.magic.test;

import java.io.ByteArrayInputStream;

public class ByteArrayInputStreamTest {

    public static void main(String[] args) {
        byte[] bytes = "01234".getBytes();
        ByteArrayInputStream inputStream = new ByteArrayInputStream(bytes);
        int c;

        System.out.println("第1次读取");
        while ((c = inputStream.read()) != -1) {
            System.out.println(Character.toUpperCase((char) c));
        }
        
        System.out.println("第2次读取");
        while ((c = inputStream.read()) != -1) {
            System.out.println(Character.toUpperCase((char) c));
        }
    }
}


     1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.
14.
15.
16.
17.
18.
19.
20.
21.
22.
23.

运行后,输出结果如下：

5.3 Repeatable read implementation

从上面的结果可以看出,第2次读取时,未读取到任何数据,If you want to be able to read again,那么又该如何处理呢？可以在第1after reading,使用 reset() 方法进行重置.

package com.magic.test;

import java.io.ByteArrayInputStream;

public class ByteArrayInputStreamTest {

    public static void main(String[] args) {
        byte[] bytes = "01234".getBytes();
        ByteArrayInputStream inputStream = new ByteArrayInputStream(bytes);
        int c;

        System.out.println("第1次读取");
        while ((c = inputStream.read()) != -1) {
            System.out.println(Character.toUpperCase((char) c));
        }

        // 使用 reset 方法进行重置,从上面对 reset() 方法源码可以看出
        // 重置后 pos = mark,而 mark 的默认值为 0,此时 pos = 0
        // Then again from the index position 0 Start reading byte data
        inputStream.reset();
        System.out.println("第2次读取");
        while ((c = inputStream.read()) != -1) {
            System.out.println(Character.toUpperCase((char) c));
        }
    }
}

     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.