用lucene为数据库搜索建立 增量索引

用 lucene 建立索引不可能每次都重新开始建立,而是按照新增加的记录,一次次的递增
建立索引的IndexWriter类,有三个参数

IndexWriter writer = new IndexWriter(path, new StandardAnalyzer(),isEmpty);

其中第三个参数是bool型的,指定它可以确定是增量索引,还是重建索引.
对于从数据库中读取的记录,譬如要为文章建立索引,我们可以记录文章的id号,然后下次再次建立索引的时候读取存下的id号,从此id后往下继续增加索引,逻辑如下.

建立增量索引,主要代码如下

public void createIndex(String path)
{
      Statement myStatement = null ;
      String articleId = " 0 " ;
      // 读取文件,获得文章id号码,这里只存最后一篇索引的文章id
     try {
         FileReader fr = new FileReader( " **.txt " );
         BufferedReader br = new BufferedReader(fr);                 
         articleId = br.readLine();
         if (articleId == null || articleId == "" )
         articleId = " 0 " ;
         br.close();
         fr.close();
       } catch (IOException e) {
         System.out.println( " error343! " );
         e.printStackTrace();
       }
     try {
         // sql语句,根据id读取下面的内容
         String sqlText = " ***** " + articleId;
         myStatement = conn.createStatement();
         ResultSet rs = myStatement.executeQuery(sqlText);
        // 写索引
         while (rs.next()) {
          Document doc = new Document();
          doc.add(Field.Keyword( " ** " , DateAdded));
          doc.add(Field.Keyword( " ** " , articleid));
          doc.add(Field.Text( " ** " , URL));    
          doc.add(Field.Text( " ** " , Content));
          doc.add(Field.Text( " ** " , Title));    
          try {
             writer.addDocument(doc);
           }
           catch (IOException e){
             e.printStackTrace();
          }
            // 将我索引的最后一篇文章的id写入文件
           try {
            FileWriter fw = new FileWriter( " **.txt " );
            PrintWriter out = new PrintWriter(fw);    
            out.close();
            fw.close();
            } catch (IOException e) {
              e.printStackTrace();
            }
          }
             ind.Close();
             System.out.println( " ok.end " );
          }
          catch (SQLException e){
             e.printStackTrace();
         }
         finally {
             // 数据库关闭操作
         }        
     }

然后控制是都建立增量索引的时候根据能否都到id值来设置IndexWriter的第三个参数为true 或者是false

boolean isEmpty = true ;
try {
     FileReader fr = new FileReader( " **.txt " );
     BufferedReader br = new BufferedReader(fr);                 
     if (br.readLine() != null ) {
         isEmpty = false ;
      }
      br.close();
      fr.close();
     } catch (IOException e) {
        e.printStackTrace();
   }

转载于:https://blog.51cto.com/ufosoft/64109