从nutch索引里面,读取文章内容

nutch作为一个完整的搜索引擎,是广大搜索爱好者学习的一个好项目。然而,因为nutch自己封装了一套严谨的IO操作类,要查看原始网页的内容,也不容易。在毕业设计的过程中,经过一般的摸索,可以由索引得到每个网页的原始内容。因此,后继的文本分类的研究也得以展开。代码如下:

import org.apache.lucene.search.IndexSearcher;
import java.io.*;
import org.apache.lucene.document.Document;
import org.apache.nutch.searcher.FetchedSegments;
import org.apache.nutch.fs.LocalFileSystem;
import java.util.ArrayList;
import java.util.Enumeration;
import org.apache.lucene.document.Field;
import org.apache.nutch.searcher.HitDetails;
public class ExtractText {
 public ExtractText() {
 }
 public static void main(String[] args) throws IOException {

   String path = args[0];//nutch crawl完的主目录 如crawl-20060622203455
   File nutchFileRoot = new File(path);
   File index = new File(nutchFileRoot,”index”);
   File segmentsDir = new  File(nutchFileRoot,”segments”);
   File outPut = new File(nutchFileRoot,”plaitext”);//存放内容的目录
   if(!outPut.isDirectory()){
     outPut.mkdir();
   }

   FetchedSegments segments = new FetchedSegments(new LocalFileSystem(), segmentsDir.toString());

   try {
     IndexSearcher searcher = new IndexSearcher(index.getAbsolutePath());
     
     for(int i = 0;i<searcher.maxDoc();i++){
       Document doc = searcher.doc(i);//依次遍历索引库中的每条记录
        
        String docNo = doc.get(“docNo”);
        String seg  = doc.get(“segment”);

        ArrayList fields = new ArrayList();
        ArrayList values = new ArrayList();

        Enumeration e = doc.fields();
        while (e.hasMoreElements()) {
          Field field = (Field)e.nextElement();
          fields.add(field.name());
          values.add(field.stringValue());
        }
       HitDetails detail = new HitDetails((String[])fields.toArray(new String[fields.size()]),
                              (String[])values.toArray(new String[values.size()]));
       
       String aa = segments.getParseText(detail).toString();//得到内容
       
       File temp =  new File(outPut,seg+”_”+docNo);//输出文件的名字,可以随便起,保证不重复就可以
       DataOutputStream outs = new DataOutputStream(new FileOutputStream(temp));
       outs.write(aa.getBytes());
       outs.close();

     }
   } catch (IOException ex) {
     ex.printStackTrace();
   }

 }
}

转自:http://hi.baidu.com/sunky/blog/item/40702ff5c42a9b25bd310949.html

发表评论

电子邮件地址不会被公开。 必填项已用*标注