run.java

简单分词程序 读入一个pdf 输出一个分好词的txt

/*
 * To change this template, choose Tools | Templates
 * and open the template in the editor.
 */

package WordSegment;

import java.io.BufferedReader;
import java.io.FileInputStream;
import java.io.FileWriter;
import java.io.InputStreamReader;
import java.util.Vector;

/**
 *
 * @author icdi
 */
public class Run {      
    
    public static void main(String[] args) throws Exception{
        WordSegment seger;
        seger = new WordSegment();    
        seger.SetDic("D:\\project\\WordSeg\\bin\\dic.dat");
        seger.setStrategy(new composite());        
       
        /*
        Vector vec = seger.Segment("小学生词典asdf,12341234");
        for(int i=0;i<vec.size();i++){            
            System.out.print(vec.get(i).toString());
            System.out.print(" ");
        }
        */
        
        PDFReader.changeToTxt("news.pdf");        
        FileInputStream srcdata = new FileInputStream("news.txt");
	BufferedReader inputbr = new BufferedReader(new InputStreamReader(srcdata,"UTF-8"));
	StringBuffer temp_sen=new StringBuffer();
	String temp="";
	while((temp=inputbr.readLine())!=null)
        {     
                
   	        temp_sen.append(temp);
    	        temp_sen.append("\n");
	}	   
	String Sentence=temp_sen.toString();        
        Vector vec = seger.Segment(Sentence);
        FileWriter outwriter = new FileWriter("D:\\project\\WordSeg\\result.txt");       
        
        for(int i=0;i<vec.size();i++){            
            outwriter.write(vec.get(i).toString());
            outwriter.write(" ");
        }
        outwriter.close();        
    }
}