lucene 3.0 使用片段(转)

为了便于调试，这里用Jsp程序作为例子：
要引入的包

<%@ page import = "org.apache.lucene.analysis.*" %>
<%@ page import = "org.apache.lucene.analysis.cn.smart.SmartChineseAnalyzer" %>
<%@ page import = "org.apache.lucene.document.*" %>
<%@ page import = "org.apache.lucene.index.*" %>
<%@ page import = "org.apache.lucene.store.*" %>
<%@ page import = "org.apache.lucene.util.*" %>

设定分词器
SmartChinese 是Lucene较新版本内置的强大分词器，当然你也可以使用第三方的，如 IKAnalyzer 等

Analyzer analyzer = new IKAnalyzer();
//Analyzer analyzer = new SmartChineseAnalyzer(Version.LUCENE_CURRENT);
Document doc = new Document();
Field f = null;

索引类

IndexWriter writer = new IndexWriter(FSDirectory.open( new File(request.getRealPath("网站的目录") ) ), 
                                     analyzer, true, IndexWriter.MaxFieldLength.LIMITED);


for( 循环要索引的东东 )
{

    新文档(这里的文档，其实也可以认为相当于数据库的“行”，当然，如果索引的对象正好是个文件，那文档这种叫法就很合理了)
    doc = new Document();
    
    doc.add( new Field("id", rs.getString("id"), Field.Store.YES, Field.Index.NOT_ANALYZED) );

    doc.add( new Field("title", rs.getString("title"), Field.Store.YES, Field.Index.ANALYZED) );
    
    d = new SimpleDateFormat("yyyy-MM-dd").format( new java.util.Date( rs.getLong("shadow_time") ) ).toString();
    f = new Field("shadowtime", d, Field.Store.YES, Field.Index.NOT_ANALYZED);
    doc.add(f);
    
    writer.addDocument(doc);

}

索引必须严格按照这样的方式去创建，如果按2.4版以下的方式，会出现方法已经过期的警告，在3.0中一些方法甚至是直接不能用的，具体是哪些可以看相关的API。

====================================================================

对已经创建的索引进行查询，即是搜索，这包括两个部份，一是如何直接读取某个索引的所有记录，二是如果对某个索引进行搜索。

一、列出索引里的文档
列出索引需引入的类或包

<%@ page import = "org.apache.lucene.document.*" %>
<%@ page import = "org.apache.lucene.index.*" %>
<%@ page import = "org.apache.lucene.store.*" %>
<%@ page import = "org.apache.lucene.util.Version" %>

索引目录

String indexPath = request.getRealPath("网站目录");
IndexReader reader = null;
long startTime = (new java.util.Date()).getTime();

   
打开索引

try {
    reader = IndexReader.open( FSDirectory.open(new File(indexPath)) , true );
}
catch (Exception e) {
    ;
}

版本：<%=reader.getVersion()%> 文档：<%=pagination.getTotalResult()%> 个<br />

TermDocs tdocs = reader.termDocs();
int i = 0;
while( tdocs.next() )
{
    i++;
            
    Document doc = reader.document( tdocs.doc() );
            
    String docid = doc.get("id");
    String title = doc.get("title");
    String area = doc.get("shadowtime");
    
    ....

}

二、搜索索引

除了上面引入的包外，还需要引入

<%@ page import = "org.apache.lucene.search.*" %>
<%@ page import = "org.apache.lucene.queryParser.*" %>

//索引目录
String indexPath = request.getRealPath("网站目录");
    
Analyzer analyzer = new IKAnalyzer();
    
IndexSearcher searcher = null;
IndexReader reader = null;
Query query = null;
TopDocs hits = null;

//打开索引目录
try
{
    reader = IndexReader.open( FSDirectory.open(new File(indexPath)) , true );
    searcher = new IndexSearcher(reader);
}
catch (Exception e)
{
    return;
}

执行查询

try
{  
    //多字段搜索，这里要搜索多少个字段就要有多个个参数
    BooleanClause.Occur[] clauses = { BooleanClause.Occur.SHOULD, BooleanClause.Occur.SHOULD };
    
    String[] fieldnames = new String[] { "title", "content"};
    IKQueryParser ikquery = new IKQueryParser();
    query = ikquery.parseMultiField(fieldnames, queryString , clauses);
                    
    //获取分词结果(这个只是便于对搜索结果输出加亮之类，并非必要选项)
    String wd = "";
    Reader r = new StringReader(queryString);
    org.apache.lucene.analysis.TokenStream tsi = analyzer.tokenStream("", r);
    org.apache.lucene.analysis.Token token;
    while ( tsi.incrementToken() )
    {
        wd = tsi.toString().replaceAll("\(\(|,(.*)", "");
        spword += (spword=="" ? wd : "," + wd );
    }
    tsi.close();
    
}
catch (Exception e)
{                        
     return ;
}

读取搜索结果

hits = searcher.search(query, 0);
for (int i = startindex; i < hits.totalHits; i++)
{
    if(i >= hits.totalHits) { break; }
            
    Document doc = searcher.doc(hits.scoreDocs[i].doc);
            
    String docid = doc.get("id");
    String title = doc.get("title");
   
   .....................

}