A simple example about full-text search based Java:Lucene

80酷酷网    80kuku.com

  索引源代码:
package lucene;

/**
* <p>Title: </p>
* <p>Description: </p>
* <p>Copyright: Copyright (c) 2003</p>
* <p>Company: </p>
* author Shirley
* version 1.0
*/

import org.apache.lucene.index.*;
import org.apache.lucene.analysis.*;
import java.io.*;
import org.apache.lucene.document.*;

public class IndexFiles {
  //使用方法:: IndexFiles [索引输出目录] [索引的文件列表] ...
  public static void main(String[] arg) throws Exception {
    String[] args = new String[2];
    //索引后存放索引信息的路径
    args[0] = System.getProperty("java.io.tmpdir", "tmp") + System.getProperty("file.separator") + "index-1";
    //待索引文件
    args[1] = "E:\\AppWork\\lucene\\rfc2047.txt";
    args[2] = "E:\\AppWork\\cyberoffice\\CO\\Sheldon Java Mail.htm";
    args[3] = "E:\\AppWork\\lucene\\englishtest.doc";
    args[4] = "E:\\AppWork\\cyberoffice\\CO\\xls1.xls";
    args[5] = "E:\\AppWork\\cyberoffice\\CO\\ppt1.ppt";

    String indexPath = args[0];
    IndexWriter writer;
    //用指定的语言分析器构造一个新的写索引器(第3个参数表示是否为追加索引)
    writer = new IndexWriter(indexPath, new SimpleAnalyzer(), false);

    for (int i=1; i<args.length; i++) {
      System.out.println("Indexing file " + args[i]);
      InputStream is = new FileInputStream(args[i]);

      //构造包含2个字段Field的Document对象
      //一个是路径path字段,不索引,只存储
      //一个是内容body字段,进行全文索引,并存储
      Document doc = new Document();
      doc.add(Field.UnIndexed("path", args[i]));
      doc.add(Field.Text("body", (Reader) new InputStreamReader(is)));
      //将文档写入索引
      writer.addDocument(doc);
      is.close();
    };
    //关闭写索引器
    writer.close();
  }
}




搜索源代码:
package lucene;

/**
* <p>Title: </p>
* <p>Description: </p>
* <p>Copyright: Copyright (c) 2003</p>
* <p>Company: </p>
* author Shirley
* version 1.0
*/

import org.apache.lucene.search.*;
import org.apache.lucene.queryParser.*;
import org.apache.lucene.analysis.*;

public class Search {
  public static void main(String[] arg) throws Exception {
    String[] args = new String[2];
    //索引后存放索引信息的路径
    args[0] = System.getProperty("java.io.tmpdir", "tmp") + System.getProperty("file.separator") + "index-1";
    //搜索關鍵字
    args[1] = "sending";

    String indexPath = args[0];
    String queryString = args[1];

    //指向索引目录的搜索器
    Searcher searcher = new IndexSearcher(indexPath);
    //查询解析器:使用和索引同样的语言分析器
    Query query = QueryParser.parse(queryString, "body", new SimpleAnalyzer());
    //搜索结果使用Hits存储
    Hits hits = searcher.search(query);
    //通过hits可以访问到相应字段的数据和查询的匹配度
    for (int i=0; i<hits.length(); i++) {
      System.out.println(hits.doc(i).get("path") + "; Score: " +  hits.score(i));
    };
  }
}


注:目前程序只支持英文索引,可以过滤文件类型为.txt  .doc  .htm  .xls  .ppt

中文索引及其它类型文件的索引正在研究中......



分享到
  • 微信分享
  • 新浪微博
  • QQ好友
  • QQ空间
点击: