CH03-InputStream
层级结构
InputStream 抽象类
public abstract int read()
// 读取数据
public int read(byte b[])
// 将读取到的数据放在 byte 数组中,该方法实际上是根据下面的方法实现的,off 为 0,len 为数组的长度
public int read(byte b[], int off, int len)
// 从第 off 位置读取 len 长度字节的数据放到 byte 数组中,流是以 -1 来判断是否读取结束的
public long skip(long n)
// 跳过指定个数的字节不读取,想想看电影跳过片头片尾
public int available()
// 返回可读的字节数量
public void close()
// 读取完,关闭流,释放资源
public synchronized void mark(int readlimit)
// 标记读取位置,下次还可以从这里开始读取,使用前要看当前流是否支持,可以使用 markSupport() 方法判断
public synchronized void reset()
// 重置读取位置为上次 mark 标记的位置
public boolean markSupported()
// 判断当前流是否支持标记流,和上面两个方法配套使用
源码实现
InputStream
public abstract class InputStream implements Closeable {
private static final int SKIP_BUFFER_SIZE = 2048; //用于skip方法,和skipBuffer相关
private static byte[] skipBuffer; // skipBuffer is initialized in skip(long), if needed.
//从输入流中读取下一个字节,
//正常返回0-255,到达文件的末尾返回-1
//在流中还有数据,但是没有读到时该方法会阻塞(block)
//Java IO和New IO的区别就是阻塞流和非阻塞流
//抽象方法!不同的子类不同的实现!
public abstract int read() throws IOException;
//将流中的数据读入放在byte数组的第off个位置先后的len个位置中
//返回值为放入字节的个数。
//这个方法在利用抽象方法read,某种意义上简单的Templete模式。
public int read(byte b[], int off, int len) throws IOException {
//检查输入是否正常。一般情况下,检查输入是方法设计的第一步
if (b == null) {
throw new NullPointerException();
} else if (off < 0 || len < 0 || len > b.length - off) {
throw new IndexOutOfBoundsException();
} else if (len == 0) {
return 0;
}
//读取下一个字节
int c = read();
//到达文件的末端返回-1
if (c == -1) { return -1; }
//返回的字节downcast
b[off] = (byte)c;
//已经读取了一个字节
int i = 1;
try {
//最多读取len个字节,所以要循环len次
for (; i < len ; i++) {
//每次循环从流中读取一个字节
//由于read方法阻塞,
//所以read(byte[],int,int)也会阻塞
c = read();
//到达末尾,理所当然返回-1
if (c == -1) { break; }
//读到就放入byte数组中
b[off + i] = (byte)c;
}
} catch (IOException ee) { }
return i;
}
//利用上面的方法read(byte[] b)
public int read(byte b[]) throws IOException {
return read(b, 0, b.length);
}
//方法内部使用的、表示要跳过的字节数目,
public long skip(long n) throws IOException {
long remaining = n;
int nr;
if (skipBuffer == null)
//初始化一个跳转的缓存
skipBuffer = new byte[SKIP_BUFFER_SIZE];
//本地化的跳转缓存
byte[] localSkipBuffer = skipBuffer;
//检查输入参数,应该放在方法的开始
if (n <= 0) { return 0; }
//一共要跳过n个,每次跳过部分,循环
while (remaining > 0) {
nr = read(localSkipBuffer, 0, (int) Math.min(SKIP_BUFFER_SIZE, remaining));
//利用上面的read(byte[],int,int)方法尽量读取n个字节
//读到流的末端,则返回
if (nr < 0) { break; }
//没有完全读到需要的,则继续循环
remaining -= nr;
}
return n - remaining;//返回时要么全部读完,要么因为到达文件末端,读取了部分
}
//查询流中还有多少可以读取的字节
//该方法不会block。在java中抽象类方法的实现一般有以下几种方式:
//1.抛出异常(java.util);2.“弱”实现。像上面这种。子类在必要的时候覆盖它。
//3.“空”实现。
public int available() throws IOException {
return 0;
}
//关闭当前流、同时释放与此流相关的资源
//关闭当前流、同时释放与此流相关的资源
public void close() throws IOException {}
//markSupport可以查询当前流是否支持mark
public synchronized void mark(int readlimit) {}
//对mark过的流进行复位。只有当流支持mark时才可以使用此方法。
public synchronized void reset() throws IOException {
throw new IOException("mark/reset not supported");
}
//查询是否支持mark
//绝大部分不支持,因此提供默认实现,返回false。子类有需要可以覆盖。
public boolean markSupported() {
return false;
}
}
FilterInputStream
public class FilterInputStream extends InputStream {
//装饰器的代码特征: 被装饰的对象一般是装饰器的成员变量
protected volatile InputStream in; //将要被装饰的字节输入流
protected FilterInputStream(InputStream in) { //通过构造方法传入此被装饰的流
this.in = in;
}
//下面这些方法,完成最小的装饰――0装饰,只是调用被装饰流的方法而已
public int read() throws IOException {
return in.read();
}
public int read(byte b[]) throws IOException {
return read(b, 0, b.length);
}
public int read(byte b[], int off, int len) throws IOException {
return in.read(b, off, len);
}
public long skip(long n) throws IOException {
return in.skip(n);
}
public int available() throws IOException {
return in.available();
}
public void close() throws IOException {
in.close();
}
public synchronized void mark(int readlimit) {
in.mark(readlimit);
}
public synchronized void reset() throws IOException {
in.reset();
}
public boolean markSupported() {
return in.markSupported();
}
}
ByteArrayInputStream
public class ByteArrayInputStream extends InputStream {
protected byte buf[]; //内部的buffer,一般通过构造器输入
protected int pos; //当前位置的cursor。从0至byte数组的长度。
//byte[pos]就是read方法读取的字节
protected int mark = 0; //mark的位置。
protected int count; //流中字节的数目。
//构造器,从一个byte[]创建一个ByteArrayInputStream
public ByteArrayInputStream(byte buf[]) {
//初始化流中的各个成员变量
this.buf = buf;
this.pos = 0;
this.count = buf.length;
}
//构造器
public ByteArrayInputStream(byte buf[], int offset, int length) {
this.buf = buf;
this.pos = offset; //与上面不同
this.count = Math.min(offset + length, buf.length);
this.mark = offset; //与上面不同
}
//从流中读取下一个字节
public synchronized int read() {
//返回下一个位置的字节//流中没有数据则返回-1
return (pos < count) ? (buf[pos++] & 0xff) : -1;
}
// ByteArrayInputStream要覆盖InputStream中可以看出其提供了该方法的实现
//某些时候,父类不能完全实现子类的功能,父类的实现一般比较通用。
//当子类有更有效的方法时,我们会覆盖这些方法。
public synchronized int read(byte b[], int off, int len) {
//首先检查输入参数的状态是否正确
if(b==null){
throw new NullPointerException();
} else if (off < 0 || len < 0 || len > b.length - off) {
throw new IndexOutOfBoundsException();
}
if (pos >= count) { return -1; }
if (pos + len > count) { len = count - pos; }
if (len <= 0) { return 0; }
//java中提供数据复制的方法
//出于速度的原因!他们都用到System.arraycopy方法
System.arraycopy(buf, pos, b, off, len);
pos += len;
return len;
}
//下面这个方法,在InputStream中也已经实现了。
//但是当时是通过将字节读入一个buffer中实现的,好像效率低了一点。
//比InputStream中的方法简单、高效
public synchronized long skip(long n) {
//当前位置,可以跳跃的字节数目
if (pos + n > count) { n = count - pos; }
//小于0,则不可以跳跃
if (n < 0) { return 0; }
//跳跃后,当前位置变化
pos += n;
return n;
}
//查询流中还有多少字节没有读取。
public synchronized int available() {
return count - pos;
}
//ByteArrayInputStream支持mark所以返回true
public boolean markSupported() {
return true;
}
//在流中当前位置mark。
public void mark(int readAheadLimit) {
mark = pos;
}
//重置流。即回到mark的位置。
public synchronized void reset() {
pos = mark;
}
//关闭ByteArrayInputStream不会产生任何动作。
public void close() throws IOException { }
}
BufferedInputStream
public class BufferedInputStream extends FilterInputStream {
private static int defaultBufferSize = 8192; //默认缓存的大小
protected volatile byte buf[]; //内部的缓存
protected int count; //buffer的大小
protected int pos; //buffer中cursor的位置
protected int markpos = -1; //mark的位置
protected int marklimit; //mark的范围
//原子性更新。和一致性编程相关
private static final
AtomicReferenceFieldUpdater<BufferedInputStream, byte[]> bufUpdater =
AtomicReferenceFieldUpdater.newUpdater (BufferedInputStream.class, byte[].class,"buf");
//检查输入流是否关闭,同时返回被包装流
private InputStream getInIfOpen() throws IOException {
InputStream input = in;
if (input == null) throw new IOException("Stream closed");
return input;
}
//检查buffer的状态,同时返回缓存
private byte[] getBufIfOpen() throws IOException {
byte[] buffer = buf;
//不太可能发生的状态
if (buffer == null) throw new IOException("Stream closed");
return buffer;
}
//构造器
public BufferedInputStream(InputStream in) {
//指定默认长度的buffer
this(in, defaultBufferSize);
}
//构造器
public BufferedInputStream(InputStream in, int size) {
super(in);
//检查输入参数
if(size<=0){
throw new IllegalArgumentException("Buffer size <= 0");
}
//创建指定长度的buffer
buf = new byte[size];
}
//从流中读取数据,填充如缓存中。
private void fill() throws IOException {
//得到buffer
byte[] buffer = getBufIfOpen();
if (markpos < 0)
//mark位置小于0,此时pos为0
pos = 0;
//pos大于buffer的长度
else if (pos >= buffer.length)
if (markpos > 0) {
int sz = pos - markpos;
System.arraycopy(buffer, markpos, buffer, 0, sz);
pos = sz;
markpos = 0;
} else if (buffer.length >= marklimit) {
//buffer的长度大于marklimit时,mark失效
markpos = -1;
//丢弃buffer中的内容
pos = 0;
}else{
//buffer的长度小于marklimit时对buffer扩容
int nsz = pos * 2;
if (nsz > marklimit)
nsz = marklimit;//扩容为原来的2倍,太大则为marklimit大小
byte nbuf[] = new byte[nsz];
//将buffer中的字节拷贝如扩容后的buf中
System.arraycopy(buffer, 0, nbuf, 0, pos);
if (!bufUpdater.compareAndSet(this, buffer, nbuf)) {
//在buffer在被操作时,不能取代此buffer
throw new IOException("Stream closed");
}
//将新buf赋值给buffer
buffer = nbuf;
}
count = pos;
int n = getInIfOpen().read(buffer, pos, buffer.length - pos);
if (n > 0) count = n + pos;
}
//读取下一个字节
public synchronized int read() throws IOException {
//到达buffer的末端
if (pos >= count) {
//就从流中读取数据,填充buffer
fill();
//读过一次,没有数据则返回-1
if (pos >= count) return -1;
}
//返回buffer中下一个位置的字节
return getBufIfOpen()[pos++] & 0xff;
}
//将数据从流中读入buffer中
private int read1(byte[] b, int off, int len) throws IOException {
int avail = count - pos; //buffer中还剩的可读字符
//buffer中没有可以读取的数据时
if(avail<=0){
//将输入流中的字节读入b中
if (len >= getBufIfOpen().length && markpos < 0) {
return getInIfOpen().read(b, off, len);
}
fill();//填充
avail = count - pos;
if (avail <= 0) return -1;
}
//从流中读取后,检查可以读取的数目
int cnt = (avail < len) ? avail : len;
//将当前buffer中的字节放入b的末端
System.arraycopy(getBufIfOpen(), pos, b, off, cnt);
pos += cnt;
return cnt;
}
public synchronized int read(byte b[], int off, int len)throws IOException {
getBufIfOpen();
// 检查buffer是否open
//检查输入参数是否正确
if ((off | len | (off + len) | (b.length - (off + len))) < 0) {
throw new IndexOutOfBoundsException();
} else if (len == 0) {
return 0;
}
int n = 0;
for (;;) {
int nread = read1(b, off + n, len - n);
if (nread <= 0) return (n == 0) ? nread : n;
n += nread;
if (n >= len) return n;
InputStream input = in;
if (input != null && input.available() <= 0) return n;
}
}
public synchronized long skip(long n) throws IOException {
// 检查buffer是否关闭
getBufIfOpen();
//检查输入参数是否正确
if (n <= 0) { return 0; }
//buffered中可以读取字节的数目
long avail = count - pos;
//可以读取的小于0,则从流中读取
if (avail <= 0) {
//mark小于0,则mark在流中
if (markpos <0) return getInIfOpen().skip(n);
// 从流中读取数据,填充缓冲区。
fill();
//可以读的取字节为buffer的容量减当前位置
avail = count - pos;
if (avail <= 0) return 0;
}
long skipped = (avail < n) ? avail : n;
pos += skipped;
//当前位置改变
return skipped;
}
//该方法不会block!返回流中可以读取的字节的数目。
//该方法的返回值为缓存中的可读字节数目加流中可读字节数目的和
public synchronized int available() throws IOException {
return getInIfOpen().available() + (count - pos);
}
//当前位置处为mark位置
public synchronized void mark(int readlimit) {
marklimit = readlimit;
markpos = pos;
}
public synchronized void reset() throws IOException {
// 缓冲去关闭了,肯定就抛出异常!程序设计中经常的手段
getBufIfOpen();
if (markpos < 0) throw new IOException("Resetting to invalid mark");
pos = markpos;
}
//该流和ByteArrayInputStream一样都支持mark
public boolean markSupported() {
return true;
}
//关闭当前流同时释放相应的系统资源。
public void close() throws IOException {
byte[] buffer;
while ( (buffer = buf) != null) {
if (bufUpdater.compareAndSet(this, buffer, null)) {
InputStream input = in;
in = null;
if (input != null) input.close();
return;
}
// Else retry in case a new buf was CASed in fill()
}
}
}
PipedInputStream
public class PipedInputStream extends InputStream {
//标识有读取方或写入方关闭
boolean closedByWriter = false;
volatile boolean closedByReader = false;
//是否建立连接
boolean connected = false;
//标识哪个线程
Thread readSide;
Thread writeSide;
//缓冲区的默认大小
protected static final int PIPE_SIZE = 1024;
//缓冲区
protected byte buffer[] = new byte[PIPE_SIZE];
//下一个写入字节的位置。0代表空,in==out代表满
protected int in = -1;
//下一个读取字节的位置
protected int out = 0;
//给定源的输入流
public PipedInputStream(PipedOutputStream src) throws IOException {
connect(src);
}
//默认构造器,下部一定要connect源
public PipedInputStream() { }
//连接输入源
public void connect(PipedOutputStream src) throws IOException {
//调用源的connect方法连接当前对象
src.connect(this);
}
//只被PipedOuputStream调用
protected synchronized void receive(int b) throws IOException {
//检查状态,写入
checkStateForReceive();
//永远是PipedOuputStream
writeSide = Thread.currentThread();
//输入和输出相等,等待空间
if (in == out) awaitSpace();
if (in < 0) {
in = 0;
out = 0;
}
//放入buffer相应的位置
buffer[in++] = (byte)(b & 0xFF);
//in为0表示buffer已空
if (in >= buffer.length) { in = 0; }
}
synchronized void receive(byte b[], int off, int len) throws IOException {
checkStateForReceive();
//从PipedOutputStream可以看出
writeSide = Thread.currentThread();
int bytesToTransfer = len;
while (bytesToTransfer > 0) {
//满了,会通知读取的;空会通知写入
if (in == out) awaitSpace();
int nextTransferAmount = 0;
if (out < in) {
nextTransferAmount = buffer.length - in;
} else if (in < out) {
if (in == -1) {
in = out = 0;
nextTransferAmount = buffer.length - in;
} else {
nextTransferAmount = out - in;
}
}
if (nextTransferAmount > bytesToTransfer) nextTransferAmount = bytesToTransfer;
assert(nextTransferAmount > 0);
System.arraycopy(b, off, buffer, in, nextTransferAmount);
bytesToTransfer -= nextTransferAmount;
off += nextTransferAmount;
in += nextTransferAmount;
if (in >= buffer.length) { in = 0; }
}
}
//检查当前状态,等待输入
private void checkStateForReceive() throws IOException {
if (!connected) {
throw new IOException("Pipe not connected");
} else if (closedByWriter || closedByReader) {
throw new IOException("Pipe closed");
} else if (readSide != null && !readSide.isAlive()) {
throw new IOException("Read end dead");
}
}
//Buffer已满,等待一段时间
private void awaitSpace() throws IOException {
//in==out表示满了,没有空间
while (in == out) {
//检查接受端的状态
checkStateForReceive();
//通知读取端
notifyAll();
try {
wait(1000);
} catch (InterruptedException ex) {
throw new java.io.InterruptedIOException();
}
}
}
//通知所有等待的线程()已经接受到最后的字节
synchronized void receivedLast() {
closedByWriter = true; //
notifyAll();
}
public synchronized int read() throws IOException {
//检查一些内部状态
if (!connected) {
throw new IOException("Pipe not connected");
} else if (closedByReader) {
throw new IOException("Pipe closed");
} else if (writeSide != null && !writeSide.isAlive()&& !closedByWriter && (in < 0)) {
throw new IOException("Write end dead");
}
//当前线程读取
readSide = Thread.currentThread();
//重复两次? ? ?
int trials = 2;
while (in < 0) {
//输入断关闭返回-1
if (closedByWriter) { return -1; }
//状态错误
if ((writeSide != null) && (!writeSide.isAlive()) && (--trials < 0)) {
throw new IOException("Pipe broken");
}
notifyAll(); // 空了,通知写入端可以写入 try {
wait(1000);
} catch (InterruptedException ex) {
throw new java.io.InterruptedIOException();
}
}
int ret = buffer[out++] & 0xFF; if (out >= buffer.length) { out = 0; }
//没有任何字节
if (in == out) { in = -1; }
return ret;
}
public synchronized int read(byte b[], int off, int len) throws IOException {
//检查输入参数的正确性
if (b == null) {
throw new NullPointerException();
} else if (off < 0 || len < 0 || len > b.length - off) {
throw new IndexOutOfBoundsException();
} else if (len == 0) {
return 0;
}
//读取下一个
int c = read();
//已经到达末尾了,返回-1
if (c < 0) { return -1; }
//放入外部buffer中
b[off] = (byte) c;
//return-len
int rlen = 1;
//下一个in存在,且没有到达len
while ((in >= 0) && (--len > 0)) {
//依次放入外部buffer
b[off + rlen] = buffer[out++];
rlen++;
//读到buffer的末尾,返回头部
if (out >= buffer.length) { out = 0; }
//读、写位置一致时,表示没有数据
if (in == out) { in = -1; }
}
//返回填充的长度
return rlen;
}
//返回还有多少字节可以读取
public synchronized int available() throws IOException {
//到达末端,没有字节
if(in < 0)
return 0;
else if(in == out)
//写入的和读出的一致,表示满
return buffer.length;
else if (in > out)
//写入的大于读出
return in - out;
else
//写入的小于读出的
return in + buffer.length - out;
}
//关闭当前流,同时释放与其相关的资源
public void close() throws IOException {
//表示由输入流关闭
closedByReader = true;
//同步化当前对象,in为-1
synchronized (this) { in = -1; }
}
}
Feedback
Was this page helpful?
Glad to hear it! Please tell us how we can improve.
Sorry to hear that. Please tell us how we can improve.