发一个Lucene 2.4.0对搜索结果高亮显示的代码

1    /** *//**
2     * search method
3     */
4    private void search(){
5        try {
6            IndexSearcher searcher = new IndexSearcher(path);
7            //Query query= wildcardQuery();
8            //Query query = phraseQuery();;
9            //Query query = booleanQuery();
10            Query query = queryParser();//获取查询内容
11            TopDocCollector collector = new TopDocCollector(10);
12            searcher.search(query,collector);
13            ScoreDoc[] hits = collector.topDocs().scoreDocs;
14
15            //高亮显示设置
16            Highlighter highlighter = null;
17            SimpleHTMLFormatter simpleHTMLFormatter = new SimpleHTMLFormatter("<read>","</read>");  
18            highlighter = new Highlighter(simpleHTMLFormatter,new QueryScorer(query));      
19            highlighter.setTextFragmenter(new SimpleFragmenter(200));//这个100是指定关键字字符串的context的长度，你可以自己设定，因为不可能返回整篇正文内容
20
21            Document doc;
22            for(int i=0;i<hits.length;i++){
23
24                System.out.println(hits[i].doc);
25                System.out.println(hits[i].score);
26                //打印文档的内容
27                doc = searcher.doc(hits[i].doc);
28                System.out.println(doc.toString());
29                //高亮出显示
30                TokenStream tokenStream =new MMAnalyzer().tokenStream("token", new StringReader(doc.get("content")));
31                System.out.println(highlighter.getBestFragment(tokenStream,doc.get("content")));
32                //没查API，我的理解是:先把content域的文本包装成Reader类型,调用MMAnalyzer的tokenStrea方法进行分词；分词结果是具有offset信息的.然后用highlighter对象的getBestFragment方法，把符合query条件的分词在文本中高亮标注
33            }
34                        
35        } catch (Exception e) {
36            // TODO Auto-generated catch block
37            e.printStackTrace();
38        }        
39    }