KWIC索引系统接受一些行，每行有若干字，每个字由若干字符组成；每行都可以循环移位。重复地把第一个字删除，然后接到行末； KWIC把所有行的各种移位情况按照字母表顺序输出。

在网上找了一个基于管道过滤器的实现，但发现有好象错误，修改了一下使之正确，以下是代码：

Filter类

package com.jason.kwic;import java.io.IOException;public abstract class Filter implements Runnable {// 定义输入管道protected Pipe input;// 定义输出管道protected Pipe output;private boolean isStart = false;Filter(Pipe input, Pipe output) {this.input = input;this.output = output;}// 防止多次调用，调用之后线程开始执行public void start() {if (!isStart) {isStart = true;Thread thread = new Thread(this);thread.start();}}// 线程的 run 方法public void run() {try {this.transform();} catch (IOException e) {e.getMessage();}}// 将输入数据转换为所需数据并写入输出管道// 由子类实现抽象方法protected abstract void transform() throws IOException;
}

Pipe类

package com.jason.kwic;import java.io.IOException;
import java.io.PipedReader;
import java.io.PipedWriter;
import java.io.PrintWriter;
import java.util.Scanner;public class Pipe {//输入管道private Scanner pipereader;//输出管道private PrintWriter pipewriter;public Pipe(){PipedWriter pw = new PipedWriter();PipedReader pr = new PipedReader();try{pw.connect(pr);} catch (IOException e){e.getMessage();}pipewriter = new PrintWriter(pw);pipereader = new Scanner(pr);}//读入一行数据到管道//@return 读入的数据public String readerLine() throws IOException{if (pipereader.hasNextLine()) {return pipereader.nextLine();}return null;}//从管道输出一行数据public void writerLine(String strline) throws IOException{pipewriter.println(strline);}//将读管道关闭，调用该方法后，不能再从管道中读数据//如不能关闭则抛出异public void closeReader() throws IOException{pipereader.close();}//先刷新数据，在将写管道关闭，调用该方法后，不能向管道中写数据//如不能关闭则抛出异常public void closeWriter() throws IOException{pipewriter.flush();pipewriter.close();}
}

Input类：

package com.jason.kwic;import java.io.File;
import java.io.IOException;
import java.util.Scanner;public class Input extends Filter{//输入文件的文件名private File infile;Input(File file, Pipe output){super(null, output);this.infile = file;}@Override//读取数据protected void transform() throws IOException {Scanner sc = new Scanner(infile);String templine = "";while(sc.hasNextLine()){templine = sc.nextLine();//System.out.println("Input:" + templine);output.writerLine(templine);}output.closeWriter();sc.close();}
}

Shift类：

package com.jason.kwic;import java.io.IOException;
import java.util.ArrayList;
public class Shift extends Filter{//单词的列表private ArrayList<String> wordlist = new ArrayList<String>();//重组后的行的列表private ArrayList<String> linelist = new ArrayList<String>();Shift(Pipe input, Pipe output){super(input, output);}@Overrideprotected void transform() throws IOException {String templine = "";//读数据while((templine = input.readerLine()) != null){//将数据拆分为不同单词this.lineSplitWord(templine);//将单词重组为句子this.recombination();//输出重组结果for(int i = 0; i < linelist.size(); i++){//System.out.println("linelist:" + linelist.get(i));output.writerLine(linelist.get(i));}//清空wordlist、linelist和templinewordlist.clear();linelist.clear();templine = "";}input.closeReader();output.closeWriter();}//从一行中提取单词存入单词表中private void lineSplitWord(String line){String word = "";int i = 0;while(i < line.length()){if(line.charAt(i) != ' '){word += line.charAt(i);}else{wordlist.add(word);word = "";}i++;}if (word.length() > 0) {wordlist.add(word);}}private void recombination(){for(int j = 0; j < wordlist.size(); j++){String templine = "";for (int k = wordlist.size() - 1 - j; k < wordlist.size(); k++){templine += wordlist.get(k) + " ";}for (int m = 0; m < wordlist.size() - 1 - j; m++){if(m != wordlist.size() - j - 2){templine += wordlist.get(m) + " ";}else{templine += wordlist.get(m);}}linelist.add(templine);}}
}

Alphabetizer类：

package com.jason.kwic;import java.io.IOException;
//import java.text.Collator;
import java.util.ArrayList;
import java.util.Collections;
import java.util.Comparator;
//import java.util.Locale;
public class Alphabetizer extends Filter{private ArrayList<String> al = new ArrayList<String>();Alphabetizer(Pipe input, Pipe output){super(input, output);}//对读入的数据进行排序protected void transform() throws IOException {String templine = null;//读入数据while((templine = input.readerLine()) != null){al.add(templine);}//按字母表排序Collections.sort(al, new AlphaabetizerComparator());//对排序后的数据进行输出for(int i = 0; i < al.size(); i++){output.writerLine(al.get(i));}input.closeReader();output.closeWriter();}//使用java提供的Collator类来实现比较
//    private class AlphaabetizerComparator implements Comparator<String> {
//
//      private Collator collator;
//      AlphaabetizerComparator(){
//          this.collator = Collator.getInstance(Locale.ENGLISH);
//      }
//
//      @Override
//      public int compare(String o1, String o2) {
//          return this.collator.compare(o1, o2);
//      }
//
//    }//自己写代码实现比较（使用字母的ascii值来进行比较)private class AlphaabetizerComparator implements Comparator<String> {@Overridepublic int compare(String o1, String o2) {if (o1 == null || o2 == null) {throw new NullPointerException();}int compareValue = 0;char o1FirstCharacter = o1.charAt(0);char o2FirstCharacter = o2.charAt(0);if(this.isLetter(o1FirstCharacter) && this.isLetter(o2FirstCharacter)) {//如果是小写的字母的值，则转成对应的大写的字母的值o1FirstCharacter = this.toUpperCase(o1FirstCharacter);o2FirstCharacter = this.toUpperCase(o2FirstCharacter);compareValue = o1FirstCharacter - o2FirstCharacter;} else {throw new RuntimeException("必须是字母");}return compareValue;}private boolean isLetter(char c) {return (c >= 65 && c <= 90) || (c >= 97 && c <= 122);}private char toUpperCase(char c) {if (Character.isLowerCase(c)) {return Character.toUpperCase(c);}return c;}}
}

Output类：

package com.jason.kwic;import java.io.File;
import java.io.IOException;
import java.io.PrintWriter;public class Output extends Filter{//输出文件的文件名private File file;Output(Pipe input, File file){super(input, null);this.file = file;}//输出数据protected void transform() throws IOException {PrintWriter pw = new PrintWriter(file);String templine = "";while((templine = input.readerLine()) != null){pw.write(templine);pw.write("\n");}pw.flush();pw.close();input.closeReader();}
}

Main主程序

package com.jason.kwic;import java.io.File;
import java.util.Scanner;public class Main {public static void main(String[] args) {File infile = new File("d:\\temp\\mykwic_in.txt");File outfile = new File("d:\\temp\\mykwic_out.txt");Scanner inputfile;Scanner outputfile;try {inputfile = new Scanner(infile);outputfile = new Scanner(outfile);// 定义三个管道Pipe pipe1 = new Pipe();Pipe pipe2 = new Pipe();Pipe pipe3 = new Pipe();// 定义四种过滤器Input input = new Input(infile, pipe1);Shift shift = new Shift(pipe1, pipe2);Alphabetizer alph = new Alphabetizer(pipe2, pipe3);Output output = new Output(pipe3, outfile);// 启动四种过滤器的线程
//          input.start();
//          shift.start();
//          alph.start();
//          output.start();//不启用线程，顺序执行四个过滤器input.transform();shift.transform();alph.transform();output.transform();// 直接输出结果System.out.println("-----  infile   -----");String str = null;while (inputfile.hasNextLine()) {str = inputfile.nextLine();System.out.println(str);}System.out.println("input  end");//启用线程时要让当前线程睡一段时间.//Thread.sleep(3000);System.out.println("-----  outfile  -----");while (outputfile.hasNextLine()) {str = outputfile.nextLine();System.out.println(str);}inputfile.close();outputfile.close();} catch (Exception e) {// e.getMessage();e.printStackTrace();}}
}

注意其中的infile和outfile对应的路径要修改成实际的路径，并且必须存在这两个文件。