学步园

这个在我的搜索引擎里用的就是这个。

package com.laozizhu.search.demo;<br /> import net.paoding.analysis.analyzer.PaodingAnalyzer;<br /> import org.apache.lucene.analysis.Analyzer;<br /> import org.apache.lucene.document.Document;<br /> import org.apache.lucene.queryParser.MultiFieldQueryParser;<br /> import org.apache.lucene.search.BooleanClause;<br /> import org.apache.lucene.search.IndexSearcher;<br /> import org.apache.lucene.search.Query;<br /> import org.apache.lucene.search.ScoreDoc;<br /> import org.apache.lucene.search.TopDocCollector;<br /> import org.apache.lucene.search.highlight.Highlighter;<br /> import org.apache.lucene.search.highlight.QueryScorer;<br /> import org.apache.lucene.search.highlight.SimpleFragmenter;<br /> import org.apache.lucene.search.highlight.SimpleHTMLFormatter;<br /> /**<br /> * Lucene 2.4 + Highlighter 2.4 的分页+高亮显示代码例子.<br><br /> * 高亮的css自己写吧。<br /> *<br /> * @author 老紫竹研究室(laozizhu.com)<br /> */<br /> public class TestLuceneHighlighter {<br /> private static final String FIELD_TITLE = "title";<br /> private static final String FIELD_BODY = "body";<br /> public synchronized Analyzer getAnalyzer() {<br /> return new PaodingAnalyzer();<br /> }<br /> public void test(String queryString, int begin, int number) {<br /> IndexSearcher isearcher = null;<br /> try {<br /> isearcher = new IndexSearcher("d:/index");<br /> /* 下面这个表示要同时搜索这两个域，而且只要一个域里面有满足我们搜索的内容就行 */<br /> BooleanClause.Occur[] clauses = { BooleanClause.Occur.SHOULD, BooleanClause.Occur.SHOULD };<br /> TopDocCollector collector = new TopDocCollector(10);<br /> Query query = MultiFieldQueryParser.parse(queryString, new String[] { FIELD_TITLE, FIELD_BODY }, clauses, getAnalyzer());<br /> isearcher.search(query, collector);<br /> ScoreDoc[] hits = collector.topDocs().scoreDocs;<br /> // 用这个进行高亮显示，默认是<b>..</b><br /> // 用这个指定<read>..</read><br /> SimpleHTMLFormatter simpleHTMLFormatter = new SimpleHTMLFormatter("<read>", "</read>");<br /> // 构造高亮<br /> // 指定高亮的格式<br /> // 指定查询评分<br /> Highlighter highlighter = new Highlighter(simpleHTMLFormatter, new QueryScorer(query));<br /> // 这个一般等于你要返回的，高亮的数据长度<br /> // 如果太小，则只有数据的开始部分被解析并高亮，且返回的数据也少<br /> // 太大，有时太浪费了。<br /> highlighter.setTextFragmenter(new SimpleFragmenter(Integer.MAX_VALUE));<br /> for (int i = begin; i < hits.length && i < begin + number; i++) {<br /> Document doc = isearcher.doc(hits[i].doc);<br /> // 有三个参数<br /> // 分析器<br /> // 要解析的字段名<br /> // 要解析的数据<br /> System.out.println(highlighter.getBestFragment(getAnalyzer(), FIELD_TITLE, doc.get(FIELD_TITLE)));<br /> }<br /> } catch (Exception e) {<br /> e.printStackTrace();<br /> } finally {<br /> if (isearcher != null) {<br /> try {<br /> isearcher.close();<br /> } catch (Exception e) {<br /> e.printStackTrace();<br /> }<br /> }<br /> }<br /> }<br />