Pactice Every Day_15

               今天完成了对特定格式文件的搜索还有通过文件大小进行搜索，感觉就是对前几天学过去的内容复习了一下，没多少新的东西。今天晚上看了看html的讲解，感觉选修课上的作业可以交上了。

              今天的代码：

//建立索引并实现几种搜索功能：
package MySearcher;

import java.io.File;
import java.io.FileReader;
import java.io.IOException;
import java.text.SimpleDateFormat;
import java.util.Date;
import org.apache.lucene.analysis.Analyzer;
import org.apache.lucene.analysis.standard.StandardAnalyzer;
import org.apache.lucene.document.Document;
import org.apache.lucene.document.Field;
import org.apache.lucene.document.NumericField;
import org.apache.lucene.index.CorruptIndexException;
import org.apache.lucene.index.IndexReader;
import org.apache.lucene.index.IndexWriter;
import org.apache.lucene.index.IndexWriterConfig;
import org.apache.lucene.index.Term;
import org.apache.lucene.queryParser.ParseException;
import org.apache.lucene.search.BooleanClause;
import org.apache.lucene.search.BooleanQuery;
import org.apache.lucene.search.FuzzyQuery;
import org.apache.lucene.search.IndexSearcher;
import org.apache.lucene.search.NumericRangeQuery;
import org.apache.lucene.search.Query;
import org.apache.lucene.search.ScoreDoc;
import org.apache.lucene.search.TopDocs;
import org.apache.lucene.search.WildcardQuery;
import org.apache.lucene.store.Directory;
import org.apache.lucene.store.FSDirectory;
import org.apache.lucene.store.LockObtainFailedException;
import org.apache.lucene.util.Version;

public class IndexAndSearcher
{     
        private final static Analyzer analyzer=new StandardAnalyzer(Version.LUCENE_35);
        private static Directory directory=null;
      public static  void index_01()
      {
      IndexWriter writer=null;
      try
      {  
        directory=FSDirectory.open(new File("f:Index_01"));
             writer=new IndexWriter(directory,new IndexWriterConfig(Version.LUCENE_35,new StandardAnalyzer(Version.LUCENE_35)));
             File file=new File("e:/Lucene实例/example/");
             Document doc=null;
            for(File f:file.listFiles())
           {
             doc=new Document();
             doc.add(new Field("content",new FileReader(f)));
             doc.add(new Field("filename",f.getName(),Field.Store.YES,Field.Index.NOT_ANALYZED));
             doc.add(new Field("path",f.getAbsolutePath(),Field.Store.YES,Field.Index.NOT_ANALYZED));
             doc.add(new NumericField("date",Field.Store.YES,true).setLongValue(f.lastModified()));
             doc.add(new NumericField("size",Field.Store.YES,true).setIntValue((int)(f.length())));
               writer.addDocument(doc);
            }
         writer.close();
      }
      catch(CorruptIndexException e)
      {
       e.printStackTrace();
      }
      catch(LockObtainFailedException e)
      {
       e.printStackTrace();
      }
      catch(IOException e)
      {
          e.printStackTrace();
      }
    } 
 
      public static void check() throws IOException
      {      //检查索引是否被正确建立(打印索引)
    FSDirectory directory = FSDirectory.open(new File("f:Index_04"));
    IndexReader reader = IndexReader.open(directory);
    for(int i = 0;i<reader.numDocs();i++)
    {
     System.out.println(reader.document(i));
    }
      }
     
      //实现功能：f1域中一定含有s1，f2中一定不含有s2，f3中可以含有s3
    public static void searcher1(String f1,String s1,String f2,String s2,String f3,String s3,int n) throws IOException
    {
     directory=FSDirectory.open(new File("f:Index_01"));
     IndexReader reader=IndexReader.open(directory);
     IndexSearcher searcher=new IndexSearcher(reader);
     
     Term t1=new Term(f1,s1);
     FuzzyQuery q1=new FuzzyQuery(t1,0.1f);
     
     Term t2=new Term(f2,s2);
     FuzzyQuery q2=new FuzzyQuery(t2,0.1f);
 
     Term t3=new Term(f3,s3);
     FuzzyQuery q3=new FuzzyQuery(t3,0.1f);
     BooleanQuery q=new BooleanQuery();
     q.add(q1,BooleanClause.Occur.MUST);
     q.add(q2,BooleanClause.Occur.MUST_NOT);
     q.add(q3,BooleanClause.Occur.SHOULD);
     TopDocs tds=searcher.search(q, n);
     ScoreDoc[] sds=tds.scoreDocs;
     System.out.println("共搜到"+tds.totalHits+"条结果,"+"以下显示"+n+"条");
     for(ScoreDoc sd:sds)
     {
      Document d=searcher.doc(sd.doc);
      System.out.println("路径为："+d.get("path")+"   文件名为："+d.get("filename")+"    文件大小为"
        +d.get("size")+"   文件修改日期为："+new SimpleDateFormat("yyyy-MM-dd hh:mm:ss").format
      (new Date(Long.valueOf(d.get("date")))));
     }     
    }
    //实现功能：搜索某个格式的文件，并且对其大小进行限定后显示搜索结果
    public static void searcher2(String type,int size,int n) throws IOException, ParseException
    {
     directory=FSDirectory.open(new File("f:Index_01"));
        IndexReader reader=IndexReader.open(directory);
        IndexSearcher searcher=new IndexSearcher(reader);
        Term t1=new Term("filename","*."+type);
        WildcardQuery q1=new WildcardQuery(t1);
        Query q2=NumericRangeQuery.newIntRange("size",size-1000,size+1000,true,true);
        BooleanQuery q=new BooleanQuery();
        q.add(q1,BooleanClause.Occur.MUST);
        q.add(q2,BooleanClause.Occur.MUST);
        TopDocs tds=searcher.search(q, n);
     ScoreDoc[] sds=tds.scoreDocs;
     System.out.println("共搜到"+tds.totalHits+"条结果,"+"以下显示"+n+"条");
     for(ScoreDoc sd:sds)
     {
      Document d=searcher.doc(sd.doc);
      System.out.println("路径为："+d.get("path")+"   文件名为："+d.get("filename")+"    文件大小为"
        +d.get("size")+"   文件修改日期为："+new SimpleDateFormat("yyyy-MM-dd hh:mm:ss").format
      (new Date(Long.valueOf(d.get("date")))));
     }     
    }
    //实现功能：搜索特定格式的文件
    public static void searcher3(String type,int n) throws IOException, ParseException
    {
     directory=FSDirectory.open(new File("f:Index_01"));
        IndexReader reader=IndexReader.open(directory);
        IndexSearcher searcher=new IndexSearcher(reader);
        Term t1=new Term("filename","*."+type);
        WildcardQuery q=new WildcardQuery(t1);     
        TopDocs tds=searcher.search(q, n);
     ScoreDoc[] sds=tds.scoreDocs;
     System.out.println("共搜到"+tds.totalHits+"条结果,"+"以下显示"+n+"条");
     for(ScoreDoc sd:sds)
     {
      Document d=searcher.doc(sd.doc);
      System.out.println("路径为："+d.get("path")+"   文件名为："+d.get("filename")+"    文件大小为"
        +d.get("size")+"   文件修改日期为："+new SimpleDateFormat("yyyy-MM-dd hh:mm:ss").format
      (new Date(Long.valueOf(d.get("date")))));
     }     
    }
    //实现功能：搜索在指定大小范围内的文件
    public static void searcher4(int start,int end,int n) throws IOException, ParseException
    {
     directory=FSDirectory.open(new File("f:Index_01"));
        IndexReader reader=IndexReader.open(directory);
        IndexSearcher searcher=new IndexSearcher(reader);
        Query q=NumericRangeQuery.newIntRange("size",start,end,true,true);
        TopDocs tds=searcher.search(q, n);
     ScoreDoc[] sds=tds.scoreDocs;
     System.out.println("共搜到"+tds.totalHits+"条结果,"+"以下显示"+n+"条");
     for(ScoreDoc sd:sds)
     {
      Document d=searcher.doc(sd.doc);
      System.out.println("路径为："+d.get("path")+"   文件名为："+d.get("filename")+"    文件大小为"
        +d.get("size")+"   文件修改日期为："+new SimpleDateFormat("yyyy-MM-dd hh:mm:ss").format
      (new Date(Long.valueOf(d.get("date")))));
     }     
    }
}  

 

 

 

 //测试类：

package MySearcher;

import java.io.IOException;

import org.apache.lucene.index.CorruptIndexException;
import org.apache.lucene.queryParser.ParseException;

public class Test {

  public static void main(String[] args) throws IOException ,CorruptIndexException, ParseException
   {
   String f1="content",s1="java",
     f2="filename",s2="java",
     f3="path",s3="",
     type="java";
  
   int n=1895, start=500,end=3000 ;
   System.out.println("一定包含: 域"+f1+"里的"+s1);
   System.out.println("一定不包含： 域"+f2+"里的"+s2);
   System.out.println("可以包含： 域"+f3+"里的"+s3);
   System.out.println("搜索结果为");
   IndexAndSearcher.index_01();
   IndexAndSearcher.searcher1(f1, s1, f2, s2, f3, s3, 10);
   System.out.println();
   System.out.println("搜索"+type+"文件并且文件大小在"+String.valueOf(n)+"左右    搜索结果为：");
   IndexAndSearcher.searcher2(s2,n,10);
   System.out.println();
   System.out.println("搜索文件类型为"+type+"的    结果为：");
   IndexAndSearcher.searcher3(s2,10);
   System.out.println();
   System.out.println("搜索大小在"+String.valueOf(start)+"~"+String.valueOf(end)+"之间的文件"+"   结果为：");
   IndexAndSearcher.searcher4(start,end,10);
  
   }

}

不知不觉写博客有两个星期了，假期也快结束了，明天起要学习新内容了，加油啊！