【Java学习笔记】管道流

作者:gnuhpc
出处:http://www.cnblogs.com/gnuhpc/

1.引言

Java I/O系统是建立在数据流概念之上的,而在UNIX/Linux中有一个类似的概念,就是管道,它具有将一个程序的输出当作另一个程序的输入的能力。在Java中,可以使用管道流进行线程之间的通信,输入流和输出流必须相连接,这样的通信有别于一般的Shared Data通信,其不需要一个共享的数据空间。

 

2.相关类及其关系

1)字节流:

分为管道输出流(PipedOutputStream)和管道输入流(PipedInputStream),利用 java.io.PipedOutputStream和java.io.PipedInputStream可以实现线程之间的二进制信息传输。如果要进行管道输出,则必须把输出流连在输入流上。 java.io.PipedOutputStream是java.io.OutputStream的直接子类,而java.io. PipedInputStream是java.io.InputStream的直接子类。PipedOutputStream和 PipedInputStream往往成对出现、配合使用。举例说明:

 

TestPipe.Java

import java.io.IOException;

public class TestPipe {

    public static void main(String[] args) {

        Send s = new Send();
        Receive r = new Receive();
        try {
            s.getPos().connect(r.getPis()); // 连接管道
        } catch (IOException e) {
            e.printStackTrace();
        }
        new Thread(s).start(); // 启动线程
        new Thread(r).start(); // 启动线程
    }
}

Receive.java

import java.io.IOException;
import java.io.PipedInputStream;

class Receive implements Runnable { // 实现Runnable接口
    private PipedInputStream pis = null;

    public Receive() {
        this.pis = new PipedInputStream(); // 实例化输入流
    }

    public void run() {
        byte b[] = new byte[1024];
        int len = 0;
        try {
            len = this.pis.read(b); // 接收数据
        } catch (IOException e) {
            e.printStackTrace();
        }
        try {
            this.pis.close();
        } catch (IOException e) {
            e.printStackTrace();
        }
        System.out.println("接收的内容为:" + new String(b, 0, len));
    }

    public PipedInputStream getPis() {
        return pis;
    }
}

Send.java

import java.io.IOException;
import java.io.PipedOutputStream;
class Send implements Runnable {
    // 实现Runnable接口
    private PipedOutputStream pos = null; // 管道输出流

    public Send() {
        this.pos = new PipedOutputStream();// 实例化输出流
    }

    public void run() {
        String str = "Hello World!!!";
        try {
            this.pos.write(str.getBytes()); // 输出信息
        } catch (IOException e) {
            e.printStackTrace();
        }
        try {
            this.pos.close(); // 关闭输出流
        } catch (IOException e) {
            e.printStackTrace();
        }
    }

    public PipedOutputStream getPos() { // 通过线程类得到输出流
        return pos;
    }
}

我们可以看到使用管道流,通过connect方法进行连接,实现了Send线程和Receive线程之间的通信。

注意:

PipedInputStream中实际是用了一个1024字节固定大小的循环缓冲区。写入PipedOutputStream的数据实际上保存到对应的 PipedInputStream的内部缓冲区。从PipedInputStream执行读操作时,读取的数据实际上来自这个内部缓冲区。如果对应的 PipedInputStream输入缓冲区已满,任何企图写入PipedOutputStream的线程都将被阻塞。而且这个写操作线程将一直阻塞,直至出现读取PipedInputStream的操作从缓冲区删除数据。这也就是说往PipedOutputStream写数据的线程Send若是和从PipedInputStream读数据的线程Receive是同一个线程的话,那么一旦Send线程发送数据过多(大于1024字节),它就会被阻塞,这就直接导致接受数据的线程阻塞而无法工作(因为是同一个线程嘛),那么这就是一个典型的死锁现象,这也就是为什么javadoc中关于这两个类的使用时告诉大家要在多线程环境下使用的原因了。

JavaConsoleOutput_2

 

应用:过滤器模式

image

使用这个模式的典型例子是Unix的shell命令。这个模式的好处在于过滤器无需知道它与何种东西进行连接,并且这可以实现并行,而且系统的可扩展性可以根据添加删除或者改变Filter进行增强。

在这举一个不断计算平均值的例子,producer作为前端的数据源,不断产生随机数,通过pipe进入filter进行数据处理,然后通过第二个pipe就行后端处理。

import java.util.*;
import java.io.*;

public class PipeTest
/* 建立3个线程(Producer、Filter、Consumer)类和两组通信管道,通过多线程将管道1的数据传送到管道2中,实现管道的通信。
* Producer => pout1->pin1 =>  Filter(pin1->pout2) => pout2->pin2 =>Consumer
*/
{
    public static void main(String args[]) {
        try {
            PipedOutputStream pout1 = new PipedOutputStream();
            PipedInputStream pin1 = new PipedInputStream(pout1);

            PipedOutputStream pout2 = new PipedOutputStream();
            PipedInputStream pin2 = new PipedInputStream(pout2);

            /* construct threads */

            Producer prod = new Producer(pout1);
            Filter filt = new Filter(pin1, pout2);
            Consumer cons = new Consumer(pin2);

            /* start threads */

            prod.start();
            filt.start();
            cons.start();
        } catch (IOException e) {
        }
    }
}

// 前端:该类的作用是产生随机数,并将其放到管道1的输出流中
class Producer extends Thread {
    private DataOutputStream out;// DataOutputStream是用于写入一些基本类型数据的类,此类的实例用于生成伪随机数流
    private Random rand = new Random();

    public Producer(OutputStream os) {
        out = new DataOutputStream(os);
    }

    public void run() {
        while (true) {
            try {
                double num = rand.nextDouble();
                // 将double值直接写入流
                out.writeDouble(num);
                System.out.println("写入流中的值是 :" + num);
                out.flush();
                sleep(Math.abs(rand.nextInt()%10));//随机休眠一段时间
            } catch (Exception e) {
                System.out.println("Error:   " + e);
            }
        }
    }
}

// 过滤器,起数据处理作用,读取管道1中输入流的内容,并将其放到管道2的输出流中
class Filter extends Thread {
    private DataInputStream in;
    private DataOutputStream out;
    private double total = 0;
    private int count = 0;

    public Filter(InputStream is, OutputStream os) {
        in = new DataInputStream(is);
        out = new DataOutputStream(os);
    }

    public void run() {
        for (;;) {
            try {
                double x = in.readDouble(); // 读取流中的数据
                total += x;
                count++;
                if (count != 0) {
                    double d = total / count;
                    out.writeDouble(d); // 将得到的数据平均值写入流
                }
            } catch (IOException e) {
                System.out.println("Error:   " + e);
            }
        }
    }
}

// 后端:读取管道2输入流的内容
class Consumer extends Thread {
    private double old_avg = 0;
    private DataInputStream in;

    public Consumer(InputStream is) {
        in = new DataInputStream(is);
    }

    public void run() {
        for (;;) {
            try {
                double avg = in.readDouble();
                if (Math.abs(avg - old_avg) > 0.01) {
                    System.out.println("现在的平均值是:   " + avg);
                    System.out.println();
                    old_avg = avg;
                }
            } catch (IOException e) {
                System.out.println("Error:   " + e);
            }
        }
    }

}

 

2)字符流

Java利用 java.io.PipedWriter和java.io.PipedReader在线程之间传输字符信息。与 java.io.PipedOutputStream和java.io.PipedInputStream类似,java.io.PipedWriter 是java.io.Writer的直接子类,java.io.PipedReader是java.io.Reader的直接子类。PipedWriter拥有一个允许指定输入管道字符流的构造方法,而PipedReader拥有一个允许指定输出管道字符流的构造方法。从而使得PipedWriter和PipedReader往往成对出现、配合使用。

 

以典型KWIC系统为例,下边的代码演示了如何使用字符流并且使用了过滤器模式:ReadLineThread --Pipe1 --> ShiftThread -- Pipe2 --> SortLinesThread

import java.io.BufferedReader;
import java.io.BufferedWriter;
import java.io.FileNotFoundException;
import java.io.FileReader;
import java.io.FileWriter;
import java.io.IOException;
import java.io.PipedReader;
import java.io.PipedWriter;
import java.util.ArrayList;
import java.util.Collections;
import java.util.StringTokenizer;

public class KwicPipe {
    public static void main(String[] args) {
        try {
            //get the input and output path
            String src = args[0];
            String dest = args[1];


            //(writeToShiftThread => readFromShiftThread) = Pipe1
            PipedReader readFromShiftThread = new PipedReader();
            PipedWriter writeToShiftThread = new PipedWriter(readFromShiftThread);


            //(writeToSortLinesThread => readFromSortLinesThread) = Pipe2
            PipedReader readFromSortLinesThread = new PipedReader();
            PipedWriter writeToSortLinesThread = new PipedWriter(readFromSortLinesThread);


            //ReadLineThread --Pipe1 --> ShiftThread -- Pipe2 --> SortLinesThread
            ReadLineThread R1 = new ReadLineThread(writeToShiftThread,src);
            ShiftThread R2 = new ShiftThread(readFromShiftThread,writeToSortLinesThread);
            SortLinesThread R3 = new SortLinesThread(readFromSortLinesThread,dest);


            //Start the three processing thread
            R1.start();
            R2.start();
            R3.start();
        }
        catch (IOException e) {
            System.out.println("NO I/O");
        }
    }
}

// read the content of kwici.dat and send the lines to another thread
class ReadLineThread extends Thread {
    PipedWriter PipeIn;
    String InputFilename= null;
    ReadLineThread(PipedWriter PlaceInPipe, String InputFilename) {
        PipeIn = PlaceInPipe;
        this.InputFilename = InputFilename;
    }
    private BufferedReader fileopen(String InputFilename) {
        BufferedReader input_file = null;
        try {
            input_file = new BufferedReader(new FileReader(InputFilename));
        } catch (IOException e) {
            System.err.println(("File not open" + e.toString()));
            System.exit(1);
        }
        return input_file;
    }
    public void run() {
        try {
            String Input;
            BufferedReader TheInput = fileopen(InputFilename);
            while ( (Input = TheInput.readLine()) != null) {
                System.out.println(Input);
                PipeIn.write(Input + "/n"); // Read from the file and then write to the pipe1
            }
        }
        catch (FileNotFoundException e) {
            System.out.println("NO FILE ");
        }
        catch (IOException e) {
            System.out.println("NO I/O");
        }
    }
}

// read the lines from ReadLineThread and shift them. Send all the shifted lines to SortLinesThread
class ShiftThread extends Thread {
    PipedReader PipeOut;
    PipedWriter PipeIn;
    ShiftThread(PipedReader ReadFromPipe, PipedWriter WriteToPipe) {
        PipeOut = ReadFromPipe;
        PipeIn = WriteToPipe;
    }
    public void run() {
        char[] cbuf = new char[80];
        int i, j;
        StringBuffer linebuff = new StringBuffer();
        try {
            // read from ReadLineThread
            i = PipeOut.read(cbuf, 0, 80);
            while (i != -1) {               
                for (j = 0; j < i; j++) {
                    //if new line
                    if (cbuf[j]=='/n'){
                        // When reach the end of line,shift it
                        shiftline(linebuff.toString());
                        // empty the buffer
                        linebuff.delete(0, linebuff.length());
                    }
                    else {
                        linebuff.append(cbuf[j]);
                    }
                }
                i = PipeOut.read(cbuf, 0, 80); //get next buffer's worth
            }
        }
        catch (FileNotFoundException e) {
            System.out.println("NO FILE ");
        }
        catch (IOException e) {
            System.out.println("NO I/O or end of stream (ShiftThread terminated)");
        }
        /* BECAUSE
         * If a thread was providing data characters to the connected piped output,
         * but the thread is no longer alive, then an IOException is thrown. (javadoc)
         */
    }
    private void shiftline( String line )
    {
        String onetoken = new String ();
        StringTokenizer tokens =
            new StringTokenizer( line );
        ArrayList<String> Tokens = new ArrayList<String> ();
        int count = tokens.countTokens();
        for ( int i = 0; i < count; i++)
        {
            onetoken = tokens.nextToken();
            if (!((onetoken.compareTo( "a" ) == 0) && (onetoken.compareTo( "an" ) == 0) && (onetoken.compareTo( "and" ) == 0) && (onetoken.compareTo( "the" ) == 0)))
            {
                Tokens.add(onetoken);
            }
        }
        for ( int tokencount = 0; tokencount < count; tokencount++ )
        {
            StringBuffer linebuffer = new StringBuffer ();
            int index = tokencount;
            for ( int i = 0; i< count; i++ )
            {
                if (index >= count)
                    index = 0;
                linebuffer.append ( Tokens.get(index)  );
                linebuffer.append (" ");
                index++;
            }  //for i
            line = linebuffer.toString();
            // send the line to the SortLinesThread
            try {
                PipeIn.write(line+ "/n");
            } catch (IOException e) {
                e.printStackTrace();
            }
        }  // for token count
        return;
    }
}

class SortLinesThread extends Thread {
    PipedReader PipeOut;
    String OutputFilename;
    ArrayList<String>  KwicList = new ArrayList<String>();
    SortLinesThread(PipedReader ReadFromPipe, String OutputFilename) {
        PipeOut = ReadFromPipe;
        this.OutputFilename = OutputFilename;
    }
    public void run() {
        char[] cbuf = new char[80];
        int i, j;
        StringBuffer linebuff = new StringBuffer();
        try {
            // read from ShiftLineThread
            i = PipeOut.read(cbuf, 0, 80);
            while (i != -1) { // I don't know we're using that (The method Read blocks until at least one character of input is available.)
                for (j = 0; j < i; j++) {
                    //if new line
                    if (cbuf[j]=='/n'){
                        // add it to the ArrayList
                        KwicList.add(linebuff.toString());
                        // adn empty the buffer
                        linebuff.delete(0, linebuff.length());
                    }
                    else {
                        //append the character to the line
                        linebuff.append(cbuf[j]);
                    }
                }           
                i = PipeOut.read(cbuf, 0, 80); //get next buffer's worth
            }
        }
        catch (FileNotFoundException e) {
            System.out.println("NO FILE ");
        }
        catch (IOException e) {
            System.out.println("NO I/O or end of stream (SortLinesThread terminated)");
        }
        /* BECAUSE
         * If a thread was providing data characters to the connected piped output,
         * but the thread is no longer alive, then an IOException is thrown. (javadoc)
         */
        // when the reading is finished, sort the ArrayList and diplay
        Collections.sort(KwicList);//sort when added
        displaylist ( KwicList );//Standard Output
        //Export to file
        try {
            export(KwicList, OutputFilename);
        } catch (Exception e) {
            System.out.println("Error Output File ");
        }
    }
    private void displaylist (ArrayList<String> KwicList )
    {
        System.out.println ("/nList : " );
        for ( int count = 0; count < KwicList.size(); count++ )
            System.out.println (KwicList.get (count) );
    }
    private void export(ArrayList<String> List, String oufFilename) throws Exception{
        BufferedWriter writer = null;
        try {
            writer = new BufferedWriter(new FileWriter(oufFilename));
        } catch (FileNotFoundException e) {
            System.err.println(("File not open" + e.toString()));
            System.exit(1);
        }

        for (int count = 0; count < List.size(); count++) {
              writer.write(List.get(count));
              writer.write("/r/n");
        }
        writer.flush();
        writer.close();
        System.out.println("Processed Finished");
    }
}

作者:gnuhpc
出处:http://www.cnblogs.com/gnuhpc/

原文地址:https://www.cnblogs.com/gnuhpc/p/2844084.html