Lucene 4.7 --高亮显示

jar

lucene-analyzers-common-4.7.0.jar
lucene-analyzers-smartcn-4.7.0.jar
lucene-core-4.7.0.jar
lucene-queryparser-4.7.0.jar
lucene-memory-4.7.0.jar
lucene-queries-4.7.0.jar
lucene-highlighter-4.7.0.jar

 

先一段查询代码

Set<String> fields = new HashSet<String>();  
fields.add("carId");  
fields.add("carName");  
  
org.apache.lucene.document.Document document = searcher.doc(scoreDoc.doc, fields);  
CarVO resultVO = new CarVO();  
resultVO.setCarId(document.getValues("carId")[0]);  
resultVO.setCarName(document.getValues("carName")[0]);  
return resultVO; 

在这里,如果说我们通过carName搜索到了某个car,那么我们希望能够高亮显示这个carName中匹配的值,那么就对这个resultVO中setCarName时进行高亮处理,方法如下:

这样调用:

resultVO.setCarName(highlightFormat(document.getValues("carName")[0],"carName",query,scoreDoc,document));

 

实现方法:

private String highlightFormat(String value, String field,   
            Query query, ScoreDoc scoreDoc, Document document) throws Exception {  
          
        QueryScorer scorer  = new QueryScorer(query,field);       
        SimpleHTMLFormatter hlightFormater = new SimpleHTMLFormatter("<span class=\"hlight\">", "</span>");  
        Highlighter hl = new Highlighter(hlightFormater,scorer);  
        hl.setTextFragmenter(new SimpleSpanFragmenter(scorer));   
        TokenStream stream  = TokenSources.getAnyTokenStream(  
                searcher.getIndexReader(), scoreDoc.doc, field, document, analyzer);  
          
        String valueWithHighlight = hl.getBestFragment(stream, value);    
        return valueWithHighlight;  
    } 

searcher就是IndexSearcher,anlyzer是分词器,这个我一般都作为静态变量长时间持有

其余的这一“大堆”东西我并不打算做太多说明,一则是这里和原来的版本变化不大,二是基本按照这些点已经足够满足基本的开发应用需求了,唯一需要大家关注的,也需要大家自己处理得,就是

SimpleHTMLFormatter hlightFormater = new SimpleHTMLFormatter("<span class=\"hlight\">", "</span>");  

当然这里提醒一句,如果是你的carName这种属性是可以让用户自定义输入的,请一定要注意对注入攻击的拦截

 

posted @ 2015-04-20 00:55  243573295  阅读(277)  评论(0编辑  收藏  举报