前言
最早做非结构化数据搜索时用的还是lucene.net,一直说在学习java的同时把lucene这块搞一搞,这拖了2年多了,终于开始搞这块了。
开发环境
idea2016、lucene6.0、jdk1.8
使用lucene准备条件
1、pom.xml
2、测试数据。 我从博客园首页拿了几条数据,直接写了个伪Dao返回一个List<Blog>
public List<Blog> listBlogs(){ List<Blog> list = new ArrayList<Blog>(); Blog b1=new Blog(6538488,"风过无痕-唐","你真的了解volatile吗,关于volatile的那些事","http://www.cnblogs.com/tangyanbo/p/6538488.html"); Blog b2=new Blog(6541312,"布鲁克石","TypeScript设计模式之职责链、状态","http://www.cnblogs.com/brookshi/p/6541312.html"); Blog b7=new Blog(6547215,"嵘么么","轻松理解JavaScript闭包","http://www.cnblogs.com/rongmm/p/6547215.html"); Blog b3=new Blog(6541220,"一线码农","使用Task的一些知识优化了一下同事的多线程协作取消的一串代码","http://www.cnblogs.com/huangxincheng/p/6541220.html"); Blog b4=new Blog(6516387,"xybaby","并发与同步、信号量与管程、生产者消费者问题","http://www.cnblogs.com/xybaby/p/6516387.html"); Blog b5=new Blog(6532713,"王福朋","深入理解 JavaScript 异步系列(4)—— Generator","http://www.cnblogs.com/wangfupeng1988/p/6532713.html"); Blog b6=new Blog(6517788,"欢醉","入坑系列之HAProxy负载均衡","http://www.cnblogs.com/zhangs1986/p/6517788.html"); list.add(b1); list.add(b2); list.add(b3); list.add(b4); list.add(b5); list.add(b6); list.add(b7); return list; }
IndexSearcher搜索
1、搜索体验,基于servlet开发,直接接受路径参数key
2、先写入索引
public class IndexWriterServlet extends HttpServlet { @Override protected void doGet(HttpServletRequest req, HttpServletResponse resp) throws ServletException, IOException { String index_path="d:\indexDir"; Directory dir=null; Analyzer analyzer=null; IndexWriterConfig config=null; IndexWriter indexWriter=null; try { dir =new SimpleFSDirectory(Paths.get(index_path)); analyzer=new StandardAnalyzer(); config =new IndexWriterConfig(analyzer); indexWriter =new IndexWriter(dir,config); BlogDao blogDao=new BlogDao(); List<Blog> list = blogDao.listBlogs(); for (Blog blog:list){ Document document=new Document(); document.add(new Field("id",blog.getId().toString(),TextField.TYPE_STORED)); document.add(new Field("author",blog.getAuthor(),TextField.TYPE_STORED)); document.add(new Field("title",blog.getTitle(),TextField.TYPE_STORED)); document.add(new Field("url",blog.getUrl(),TextField.TYPE_STORED)); indexWriter.addDocument(document); } indexWriter.commit(); System.out.println("====索引创建完成===="); }catch (Exception ex){ ex.printStackTrace(); }finally { if(indexWriter !=null){ indexWriter.close(); } } } @Override protected void doPost(HttpServletRequest req, HttpServletResponse resp) throws ServletException, IOException { doGet(req,resp); } }
2、查询索引数据
public class IndexReaderServlet extends HttpServlet { @Override protected void doGet(HttpServletRequest req, HttpServletResponse resp) throws ServletException, IOException { Directory dir =null; DirectoryReader reader =null; IndexSearcher searcher=null; String index_path="d:\indexDir"; try{ //dir =new RAMDirectory(); dir =new SimpleFSDirectory(Paths.get(index_path)); reader = DirectoryReader.open(dir); searcher=new IndexSearcher(reader); String key = req.getParameter("key"); Query query = new TermQuery(new Term("title",key)); TopDocs rs=searcher.search(query,100); resp.setContentType("text/html"); resp.setCharacterEncoding("UTF-8"); PrintWriter out = resp.getWriter(); StringBuilder strHtml =new StringBuilder(); for (int i = 0; i < rs.scoreDocs.length; i++) { Document firstHit = searcher.doc(rs.scoreDocs[i].doc); strHtml.append("title:"+firstHit.getField("title").stringValue()+"<br>"); strHtml.append("author:"+firstHit.getField("author").stringValue()+"<br>"); strHtml.append("<a href='"+firstHit.getField("url").stringValue()+"'>"+ firstHit.getField("url").stringValue()+"</a><br>"); strHtml.append("====================================================<br>"); } out.write(strHtml.toString()); }catch (Exception ex){ ex.printStackTrace(); }finally { reader.close(); } } @Override protected void doPost(HttpServletRequest req, HttpServletResponse resp) throws ServletException, IOException { doGet(req,resp); } }
总结
读写索引用到的几个关键类:IndexWriter、IndexSearcher、Directory、DirectoryReader、Document、TermQuery、Field