lucene grouping 实现分组，并查询出分组数据

最新推荐文章于 2019-01-24 00:58:37 发布

liang_zeng

最新推荐文章于 2019-01-24 00:58:37 发布

阅读量466

点赞数

分类专栏： lucene grouping 文章标签： lucene

本文链接：https://blog.csdn.net/liang_zeng/article/details/84244084

版权

lucene grouping 专栏收录该内容

1 篇文章 0 订阅

订阅专栏

查询出分组以后的数据，请注意，这里没做分页。


/**
	 * 获取分组数据
	 * @return
	 * @throws Exception
	 */
	public Map<string list object>&gt;&gt; getGroupSearchData() throws Exception {

		int groupOffset = 0;

		int maxDocsPerGroup = 100;

		int withinGroupOffset = 0;

		int topNGroups = 10;

		TermFirstPassGroupingCollector c1 = new TermFirstPassGroupingCollector(Constant.CATEGORY, Sort.RELEVANCE, topNGroups);

		boolean cacheScores = true;

		double maxCacheRAMMB = 4.0;

		CachingCollector cachedCollector = CachingCollector.create(c1,cacheScores, maxCacheRAMMB);

		AAnalyzer analyzer = new AAnalyzer(true,false);

		BooleanQuery query = this.searchParamter.getBooleanQuery();

		SysContext context = SysContext.getInstance();

		IndexSearcher search = context.getSearcher();

		search.search(query, cachedCollector);

		Collection<searchgroup>&gt; topGroups = c1.getTopGroups(groupOffset, true);

		if (topGroups == null) {

			return new HashMap<string list object>&gt;&gt;(0);
		}

		TermSecondPassGroupingCollector c2 = new TermSecondPassGroupingCollector(Constant.CATEGORY, topGroups, Sort.RELEVANCE, Sort.RELEVANCE,maxDocsPerGroup, true, true, true);

		if (cachedCollector.isCached()) {

			cachedCollector.replay(c2);

		} else {

			context.getSearcher().search(query, c2);
		}

		TopGroups<string> tg = c2.getTopGroups(withinGroupOffset);

		// 同义词
		SynonymDic synDic = this.searchParamter.getAnalyzer().getSynDic();

		//标红显示符合查询的字
		SimpleHTMLFormatter sh = new SimpleHTMLFormatter("<font color="red">", "</font>" , synDic.getSynonym());

		Highlighter high = new Highlighter(sh, new QueryScorer(query));

		high.setTextFragmenter(new SimpleFragmenter(100));

		GroupDocs<string>[] gds = tg.groups;

		Document doc = null;

		Map<string list object>&gt;&gt; ret = new HashMap<string list>&gt;&gt;(gds.length);

		List<map object>&gt; list;

		Map<object object> map;

		IndexUtil iu = new IndexUtil();

		for (GroupDocs<string> gd : gds) {

			/**
			 * 这个就是分组以后的数据了，循环它，放到list里面去
			 */
			ScoreDoc [] docs = gd.scoreDocs;

			list = new ArrayList<map>&gt;(gd.totalHits);

			for (ScoreDoc scoreDoc : docs) {

				map = new HashMap<object object>();

				doc = context.getSearcher().doc(scoreDoc.doc);

				String content = iu.getNoHtml(true,doc.get(Constant.CONTENT));

				TokenStream tokenStream = analyzer.tokenStream(Constant.CONTENT,new StringReader(content));

				String tmpStr = high.getBestFragment(tokenStream, content);

				if (tmpStr != null) {

					content = tmpStr;

				} else{

					content =  iu.getContent(content, 100);// content.substring(0 , content.length() &gt;= 100 ? 100 : content.length());
				}

				String title = iu.getNoHtml(true,doc.get(Constant.TITLE));

				tokenStream = analyzer.tokenStream(Constant.TITLE, new StringReader(title));

				String tmpName = high.getBestFragment(tokenStream, title);

				if (tmpName != null) {

					title = tmpName;

				} else{

					title = iu.getContent(title,100);//title.substring(0 , title.length() &gt;= 100 ? 100 : title.length());
				}

				tokenStream.close();

				map.put(Constant.ID, doc.get(Constant.ID));
				map.put(Constant.TITLE, title);
				map.put(Constant.CONTENT, content);
				map.put(Constant.PATH, doc.get(Constant.PATH));
				map.put(Constant.OUTSIDEURL, doc.get(Constant.OUTSIDEURL));
				map.put(Constant.INDEX_TYPE, doc.get(Constant.INDEX_TYPE));
				map.put(Constant.ATTACH_TYPE, doc.get(Constant.ATTACH_TYPE));
				map.put(Constant.CATEGORY, doc.get(Constant.CATEGORY));
				list.add(map);

			}
			ret.put(gd.groupValue, list);

		}
		// 用完了就释放该查询
		context.releaseSearcher(search);

		return ret;

	}