工程做完的效果图如下,com.zly.indexManager中两个类,分别创建索引和搜索索引,
com.zly.test.entity中是使用的实体类,分别是NewsType(新闻类型),NewsItem(新闻具体条目),PageControl(分页实体bean) , SearchResultBean(保存搜索结果的bean).
浏览和搜索的前提是有据可查,没有数据什么都实现不了 , 我使用了Htmlparser通过抓取页面信息的形式将新闻添加进数据库 , 添加数据库数据使用了hibernate3
使用了Annotation的方式完成数据库的映射。
//NewsType(新闻类型)
package com.zly.test.entity; import java.io.Serializable; import javax.persistence.Column; import javax.persistence.Entity; import javax.persistence.GeneratedValue; import javax.persistence.GenerationType; import javax.persistence.Id; import javax.persistence.Table; @Entity @Table(name = "t_newsType") public class NewsType implements Serializable { private static final long serialVersionUID = -5092622762025999122L; private Integer id; private String newsTypeName; @Column public String getNewsTypeName() { return newsTypeName; } public void setNewsTypeName(String newsTypeName) { this.newsTypeName = newsTypeName; } @Id @GeneratedValue(strategy = GenerationType.AUTO) public Integer getId() { return id; } public void setId(Integer id) { this.id = id; } }
//NewsItem(新闻具体条目)
package com.zly.test.entity; import java.io.Serializable; import java.util.Date; import javax.persistence.Column; import javax.persistence.Entity; import javax.persistence.GeneratedValue; import javax.persistence.GenerationType; import javax.persistence.Id; import javax.persistence.JoinColumn; import javax.persistence.ManyToOne; import javax.persistence.Table; import javax.persistence.Temporal; import javax.persistence.TemporalType; @Entity @Table(name = "t_newsItem") public class NewsItem implements Serializable { private static final long serialVersionUID = -7532888546907495633L; private Integer id ; private String newsTitle ; private String newsContent; private Date publishTime; private String resource; private NewsType type; private String editor; @Column public String getEditor() { return editor; } public void setEditor(String editor) { this.editor = editor; } @ManyToOne @JoinColumn(name = "t_newsType_id") public NewsType getType() { return type; } public void setType(NewsType type) { this.type = type; } @Id @GeneratedValue(strategy = GenerationType.AUTO) public Integer getId() { return id; } public void setId(Integer id) { this.id = id; } @Column public String getNewsTitle() { return newsTitle; } public void setNewsTitle(String newsTitle) { this.newsTitle = newsTitle; } @Temporal(value = TemporalType.TIMESTAMP) public Date getPublishTime() { return publishTime; } public void setPublishTime(Date publishTime) { this.publishTime = publishTime; } public String getResource() { return resource; } public void setResource(String resource) { this.resource = resource; } public String getNewsContent() { return newsContent; } public void setNewsContent(String newsContent) { this.newsContent = newsContent; } }
添加所有新闻类型的类放在了com.zly.test.newsFetch包下 , 具体代码:
package com.zly.test.newsfetch; import java.text.DateFormat; import java.text.SimpleDateFormat; import java.util.Date; import java.util.LinkedHashSet; import java.util.Set; import java.util.regex.Matcher; import java.util.regex.Pattern; import org.hibernate.Session; import org.hibernate.SessionFactory; import org.hibernate.cfg.AnnotationConfiguration; import org.hibernate.cfg.Configuration; import org.htmlparser.Parser; import org.htmlparser.Tag; import org.htmlparser.filters.NodeClassFilter; import org.htmlparser.tags.Div; import org.htmlparser.tags.LinkTag; import org.htmlparser.util.NodeList; import org.htmlparser.visitors.NodeVisitor; import com.zly.test.entity.NewsItem; import com.zly.test.entity.NewsType; public class GetNews { public static void main(String[] args) throws Exception { //插入数据新闻类型 //insertAllTypes(); //插入所有新闻数据 //国内新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/gn/2009/05" ,"/news.shtml" ,1); //国际新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/gj/2009/05" ,"/news.shtml" ,2); //社会新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/sh/2009/05" ,"/news.shtml" ,3); //港澳新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/ga/2009/05" ,"/news.shtml" ,4); //台湾新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/tw/2009/05" ,"/news.shtml" ,5); //华人新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/hr/2009/05" ,"/news.shtml" ,6); //经济新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/cj/2009/05" ,"/news.shtml" ,7); //文化新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/wh/2009/05" ,"/news.shtml" ,8); //娱乐新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/yl/2009/05" ,"/news.shtml" ,9); //体育新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/ty/2009/05" ,"/news.shtml" ,10); //教育新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/edu/2009/05" ,"/news.shtml" ,11); //健康新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/jk/2009/05" ,"/news.shtml" ,12); //生活新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/life/2009/05" ,"/news.shtml" ,13); //IT新闻 //insertNewsItems("http://www.chinanews.com.cn/scroll-news/it/2009/05" ,"/news.shtml" ,14); } public static void insertAllTypes() { Configuration cfg = new AnnotationConfiguration().configure(); SessionFactory factory = cfg.buildSessionFactory(); Session session = factory.openSession(); String str = new String("国内 国际 社会 港澳 台湾 华人 经济 文化 娱乐 体育 教育 健康 生活 IT"); String[] typesStr = str.split(" "); NewsType[] types = new NewsType[typesStr.length]; session.beginTransaction(); for (int i = 0; i < typesStr.length; i++) { types[i] = new NewsType(); types[i].setNewsTypeName(typesStr[i]); session.save(types[i]); } session.getTransaction().commit(); session.close(); } //处理5.1 - 5.5 所有的具体类型的新闻 public static void insertNewsItems(String before , String after , int type) throws Exception { Configuration cfg = new AnnotationConfiguration().configure(); SessionFactory factory = cfg.buildSessionFactory(); Session session = factory.openSession(); //得到当前新闻所属类别 NewsType newsType = (NewsType) session.get(NewsType.class, type); final Set<NewsItem> set = new LinkedHashSet<NewsItem>(); //获取5月1日-5月5日的新闻 for (int i = 1; i <= 5; i++) { String src = before; if(i < 10) { src = src + "0" + i; }else { src = src + i ; } src = src + after; //使用htmlParser获取每一条新闻的超链接 Parser parser = new Parser(src); parser.setEncoding("gb2312"); NodeList divList = parser.parse(new NodeClassFilter(Div.class)); for (int j = 0; j < divList.size(); j++) { Div div = (Div) divList.elementAt(j); String divClass = div.getAttribute("id"); //取得id为news_list的div if(divClass != null && divClass.equals("news_list")) { div.accept(new NodeVisitor() { //遍历news_list里面的所有超链接 public void visitTag(Tag tag) { if(tag instanceof LinkTag) { String href = ((LinkTag)tag).getLink(); if(!href.startsWith("http")) { href = "http://www.chinanews.com.cn" + href; } System.out.println(href); //找到超链接,将这个超链接所在的页面进行处理,抓取新闻数据,并将其保存在Set中。 try{ insertOneNews(href , set); }catch(Exception e) { e.printStackTrace(); } } } }); } } } //获取新闻完成后,将所有NewsItem添加到数据库 session.beginTransaction(); for (NewsItem newsItem : set) { newsItem.setType(newsType); session.save(newsItem); } session.getTransaction().commit(); session.close(); } //处理每一个具体的新闻超链接,得到NewsItem类 public static void insertOneNews(String src , Set<NewsItem> set) throws Exception { Parser parser = new Parser(src); parser.setEncoding("gb2312"); NodeList divList = parser.extractAllNodesThatMatch(new NodeClassFilter(Div.class)); NewsItem newsItem = new NewsItem(); String title = ""; Date parse = null; String content = ""; String editor = ""; //遍历网页的div。将制定div里面的信息设置到NewsItem实体类中 for (int i = 0; i < divList.size(); i++) { Div div = (Div) divList.elementAt(i); String divString = div.getAttribute("class"); if(divString != null) { //设置标题 if(divString.equals("left_bt")) { title = div.toPlainTextString(); } //设置发布时间 if(divString.equals("left_time")) { String publishStr = ""; Pattern pattern = Pattern.compile("[//d]{4}年[//d]{2}月[//d]{2}日 [//d]{2}:[//d]{2}"); Matcher matcher = pattern.matcher(div.toPlainTextString()) ; if(matcher.find()) { publishStr = matcher.group(); } DateFormat format = new SimpleDateFormat("yyyy年MM月dd日 HH:mm"); parse = format.parse(publishStr); } //设置正文内容 if(divString.equals("left_zw")) { content = div.toHtml(); } //设置记者名称 if(divString.equals("left_name")) { editor = div.toPlainTextString().trim(); editor = editor.substring(editor.indexOf(":") + 1, editor.lastIndexOf("】")); } } } newsItem.setEditor(editor); newsItem.setNewsContent(content); newsItem.setNewsTitle(title); newsItem.setPublishTime(parse); //设置新闻来源 newsItem.setResource("最快新闻网"); //将新闻添加到set中 set.add(newsItem); } }
通过上面的代码完成了所有的数据添加工作。
下面根据ssh的流程分别定制dao , manager , action
com.zly.test.dao包中是所有操作dao的抽象类和接口
我们直接看这些接口的实现
//NewsItemDaoHibernate 新闻实体类dao
@SuppressWarnings("unchecked") //根据分页得到具体页的内容 public List<com.zly.test.entity.NewsItem> getNewsItemByFirstResultAndMaxResult( final int firstResult, final int maxResult , final int type) { return (List<NewsItem>) this.getHibernateTemplate().execute(new HibernateCallback() { public Object doInHibernate(Session session) throws HibernateException, SQLException { return session.createQuery(" from NewsItem where type.id = '" + type + "'").setFirstResult(firstResult).setMaxResults(maxResult).list(); } }); } //得到数据总条数 , 以便计算总页数 public Long getItemCount(final int id) { return (Long) this.getHibernateTemplate().execute(new HibernateCallback(){ public Object doInHibernate(Session session) throws HibernateException, SQLException { return session.createQuery("select count(*) from NewsItem where type.id = '" + id + "'").uniqueResult(); } }); } //显示新闻数据页面用到, 查询具体新闻属于哪个新闻类别 public int getNewsType(final int id) { return (Integer) this.getHibernateTemplate().execute(new HibernateCallback() { public Object doInHibernate(Session session) throws HibernateException, SQLException { NewsItem newsItem = get(id); NewsType type = newsItem.getType(); return type.getId(); } }); }
只用到了一个 NewsManagerImpl
package com.zly.test.service.impl; import java.util.List; import com.zly.test.dao.NewsItemDao; import com.zly.test.entity.NewsItem; import com.zly.test.service.NewsManager; public class NewsManagerImpl implements NewsManager { //新闻条目dao private NewsItemDao newsItemDao; public NewsItemDao getNewsItemDao() { return newsItemDao; } public void setNewsItemDao(NewsItemDao newsItemDao) { this.newsItemDao = newsItemDao; } //根据分页得到内容 public List<NewsItem> getNewsItemByFirstResultAndMaxResult(int firstResult, int maxResult , int type) { return newsItemDao.getNewsItemByFirstResultAndMaxResult(firstResult, maxResult , type); } //得到总记录数,以便计算总页数 public Long getItemCounts(int id) { return newsItemDao.getItemCount(id); } //通过id得到具体新闻 public NewsItem getNewsById(int id) { return newsItemDao.get(id); } //通过id得到所属新闻类别的id public int getNewsType(int id) { return newsItemDao.getNewsType(id); } }
在applicationContext-action.xml中配置action类别
<bean class="com.zly.test.action.NewsAction" id="newsAction" scope="request"> <property name="newsManager" ref="newsManager"></property> <property name="map" ref="map"></property> <property name="map1" ref="map1"></property> </bean>
其中定一个两个map , 因为主页的查看分类新闻的url是采用的这种形式<a href="newsAction.action?category=china" target="_blank">国内</a> 名字为map的Map中保存信息如下
<bean id="map" class="java.util.HashMap"> <constructor-arg> <map> <entry key="china" value="1###国内新闻" /> <entry key="world" value="2###国际新闻" /> <entry key="society" value="3###社会新闻" /> <entry key="compatriot" value="4###港台新闻" /> <entry key="taiwan" value="5###台湾新闻" /> <entry key="huaren" value="6###华人新闻" /> <entry key="economic" value="7###经济新闻" /> <entry key="wenhua" value="8###文化新闻" /> <entry key="entertainment" value="9###娱乐新闻" /> <entry key="sports" value="10###体育新闻" /> <entry key="jiaoyu" value="11###教育新闻" /> <entry key="jiankang" value="12###健康新闻" /> <entry key="life" value="13###生活新闻" /> <entry key="keji" value="14###科技新闻" /> </map> </constructor-arg> </bean>
key是?category后面的值 , value是两部分 , 被###分割开 , 前面的数值是所属新闻类别的id值, 后面的文字是其类别的文字。将其保存在map中,避免不停地查询数据库。
分页类PageControl的代码如下:
package com.zly.test.entity; import java.util.List; public class PageControl { private int curPage ; //当前是第几页 private int maxPage ; //一共有多少页 private Long maxRowCount ; //一共有多少行 private int rowsPerPage = 8 ; //每页有多少行 private List<?> data;//每页的User public int getCurPage() { return curPage; } public void setCurPage(int curPage) { this.curPage = curPage; } public int getMaxPage() { return maxPage; } public void setMaxPage(int maxPage) { this.maxPage = maxPage; } public List<?> getUserList() { return data; } public void setUserList(List<?> data) { this.data = data; } public Long getMaxRowCount() { return maxRowCount; } public void setMaxRowCount(Long amaxRowCountxRowCount) { this.maxRowCount = amaxRowCountxRowCount; } public int getRowsPerPage() { return rowsPerPage; } public void setRowsPerPage(int rowsPerPage) { this.rowsPerPage = rowsPerPage; } public void countMaxPage() { //根据总行数计算总页数 if (this.maxRowCount % this.rowsPerPage ==0){ this.maxPage = (int) (this.maxRowCount/this.rowsPerPage); }else{ this.maxPage = (int) (this.maxRowCount/this.rowsPerPage + 1); } } }
被许多页所包含的page.jsp代码如下:
<%@ page language="java" contentType="text/html;charset=utf-8" pageEncoding="utf-8"%> <%@ taglib prefix="c" uri="http://java.sun.com/jsp/jstl/core" %> <html> <head> <script language="javascript"> function jumping() { document.pageForm.submit(); return true; } function gotoPage(pageNum) { document.pageForm.jumpPage.value = pageNum ; document.pageForm.submit(); return true; } </script> </head> <body> <c:if test="${pageControl.maxPage != 1}"> <form method="post" action="pageAction.action" name="pageForm"> <table> <tr> <td> 每页${pageControl.rowsPerPage}行 </td> <td> 共${pageControl.maxRowCount }行 </td> <td> 第${pageControl.curPage }页 </td> <td> 共${pageControl.maxPage }页 </td> <td> <c:choose> <c:when test="${pageControl.curPage == 1}"> 首页 上一页 </c:when> <c:otherwise> <A HREF="javascript:gotoPage(1)">首页</A> <A HREF="javascript:gotoPage(${pageControl.curPage - 1} )">上一页</A> </c:otherwise> </c:choose> <c:choose> <c:when test="${pageControl.curPage == pageControl.maxPage}"> 下一页 尾页 </c:when> <c:otherwise> <A HREF="javascript:gotoPage(${pageControl.curPage + 1})">下一页</A> <A HREF="javascript:gotoPage(${pageControl.maxPage })">尾页</A> </c:otherwise> </c:choose> 转到第 <SELECT name="jumpPage" onchange="jumping()"> <c:forEach var="i" begin="1" end="${pageControl.maxPage}" step="1"> <c:choose> <c:when test="${i == pageControl.curPage}"> <OPTION selected value="${i}">${i }</OPTION> </c:when> <c:otherwise> <OPTION value="${i}">${i}</OPTION> </c:otherwise> </c:choose> </c:forEach> </SELECT> 页 </td> </tr> </table> </form> </c:if> </body> </html>
下面是struts.xml中关于页面展示新闻的配置
<action name="newsAction" class="newsAction" > <result type="redirect">pageAction.action</result> </action> <action name="pageAction" class="newsAction" method="pageAction"> <result>/result.jsp</result> </action> <action name="detailAction" class="newsAction" method="showDetail"> <result>/detail.jsp</result> </action>
NewsAction代码如下:
package com.zly.test.action; import java.util.List; import java.util.Map; import com.zly.test.entity.NewsItem; import com.zly.test.entity.PageControl; import com.zly.test.service.NewsManager; public class NewsAction extends BaseAction { private static final long serialVersionUID = 7780804627621048756L; //对应分页jsp中的跳转到第几页 private String jumpPage; //对应url?category的值 private String category; //对新闻数据进行处理 private NewsManager newsManager; //保存<entry key="china" value="1###国内新闻" />这样的信息 private Map<String , String> map; //保存<entry key="1" value="china###国内新闻" />这样的信息 private Map<String , String> map1; public Map<String, String> getMap1() { return map1; } public void setMap1(Map<String, String> map1) { this.map1 = map1; } public Map<String, String> getMap() { return map; } public void setMap(Map<String, String> map) { this.map = map; } public String getJumpPage() { return jumpPage; } public void setJumpPage(String jumpPage) { this.jumpPage = jumpPage; } public NewsManager getNewsManager() { return newsManager; } public void setNewsManager(NewsManager newsManager) { this.newsManager = newsManager; } //处理newsAction.action?category=xxx public String execute() { //得到相应的 1###国内新闻 String typeIdInfo = map.get(category); //这里得到的是新闻类别id Integer typeId = Integer.parseInt(typeIdInfo.split("###")[0]); //这里得到的是新闻类别字符串 String typeString = typeIdInfo.split("###")[1]; //将上面三个变量保存在session中,在jsp页面中显示 this.getSession().setAttribute("category", category); this.getSession().setAttribute("typeString" , typeString); this.getSession().setAttribute("id", typeId); //跳转到pageAction中,处理分页 return SUCCESS; } public String pageAction() { //取出新闻类别id Integer id = (Integer) this.getSession().getAttribute("id"); //查看是不是第一次来分页,如果是,当前页设置成第一个 if(jumpPage == null) { jumpPage = "1"; } //从request范围内取出PageControl , 如为空,则第一次分页,创建pageControl对象 PageControl pageControl = (PageControl) this.getRequest().getAttribute("pageControl"); if(pageControl == null) { pageControl = new PageControl(); //第一次的时候得到相应新闻类别所对应的记录的总数 pageControl.setMaxRowCount(newsManager.getItemCounts(id)); //计算一共多少页 pageControl.countMaxPage(); } //设置jumpPage的值为当前页 pageControl.setCurPage(Integer.parseInt(jumpPage)); //计算出hibernate中firstResult的值 int firstResult = (pageControl.getCurPage() - 1) * pageControl.getRowsPerPage() + 1; //设置hibernate中maxResult的值 int maxResult = pageControl.getRowsPerPage(); //利用hibernate得到当前页的数据,并保存在pageControl分页实体类中 List<NewsItem> userList = newsManager.getNewsItemByFirstResultAndMaxResult(firstResult, maxResult , id); pageControl.setData(userList); //将分页实体类保存在request范围内。 this.getRequest().setAttribute("pageControl", pageControl); //将页面视图跳转到result.jsp return SUCCESS; } public String showDetail() { //得到url中的?id=xxx String newsId = this.getRequest().getParameter("id"); int id = Integer.parseInt(newsId); //使用hibernate得到具体id的新闻记录 NewsItem item = newsManager.getNewsById(id); //得到id记录所对应的新闻类型的值 map1这种形式<entry key="1" value="china###国内新闻" /> int typeId = newsManager.getNewsType(id); //得到china,添加到jsp页面的<a href="newsAction?category=">里面 String category = map1.get("" + typeId).split("###")[0]; //得到国内新闻,显示在jsp页面的多个位置 String typeString = map1.get("" + typeId).split("###")[1]; //将以上多个数据添加到request范围内,以便显示。 this.getRequest().setAttribute("news", item); this.getRequest().setAttribute("category", category); this.getRequest().setAttribute("typeString", typeString); return SUCCESS; } public String getCategory() { return category; } public void setCategory(String category) { this.category = category; } }
首页页面index.jsp,里面有几个分类超链接和搜索对话框
<%@ page language="java" pageEncoding="utf-8"%> <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN"> <html> <head> <title>最快新闻网</title> <link href="css/main.css" rel="stylesheet" type="text/css" /> </head> <body> <div id="mainlink"> <span><a href="index.jsp">最快新闻网</a></span><br /><br /> <a href="newsAction.action?category=china" target="_blank">国内</a> | <a href="newsAction.action?category=world" target="_blank">国际</a> | <a href="newsAction.action?category=society" target="_blank">社会</a> | <a href="newsAction.action?category=compatriot" target="_blank">港澳</a> | <a href="newsAction.action?category=taiwan" target="_blank">台湾</a> | <a href="newsAction.action?category=huaren" target="_blank">华人</a> | <a href="newsAction.action?category=economic" target="_blank">经济</a> | <a href="newsAction.action?category=wenhua" target="_blank">文化</a> | <a href="newsAction.action?category=entertainment" target="_blank">娱乐</a> | <a href="newsAction.action?category=sports" target="_blank">体育</a> | <a href="newsAction.action?category=jiaoyu" target="_blank">教育</a> | <a href="newsAction.action?category=jiankang" target="_blank">健康</a> | <a href="newsAction.action?category=life" target="_blank">生活</a> | <a href="newsAction.action?category=keji" target="_blank">IT</a><br /> <form action="searchAction.action" method="post" target="_blank"> 本站搜索: <input type="text" name="searchParam" style="height:20px"/> <select name="searchWhich"> <option value="title"/>标题 <option value="content"/>内容 </select> <input type="submit" value="搜索" style="height:23px"/> </form> </div> </body> </html>
其表现形式如下:
新闻分页展示页面result.jsp代码如下:
<%@ page language="java" contentType="text/html;charset=utf-8" pageEncoding="utf-8"%> <%@taglib prefix="c" uri="http://java.sun.com/jsp/jstl/core" %> <%@taglib prefix="fmt" uri="http://java.sun.com/jsp/jstl/fmt" %> <!DOCTYPE html PUBLIC "-//W3C//DTD HTML 4.01 Transitional//EN" "http://www.w3.org/TR/html4/loose.dtd"> <html> <head> <meta http-equiv="Content-Type" content="text/html; charset=utf-8"> <title>${typeString} -- 最快新闻网</title> </head> <body> <jsp:include page="index.jsp"></jsp:include> <div id="content"> <font color="red" size="5">${typeString}</font> <br /><br /><br /><br /> <div id="newsListTitle" style="float:left;"> <c:forEach items="${pageControl.data}" var="news"> <div style="margin-top: 20px;"> <span> <a href="detailAction.action?id=${news.id }">${news.newsTitle }</a> </span> <br /> </div> </c:forEach> </div> <div id="newsListTime"> <c:forEach items="${pageControl.data}" var="news"> <div style="margin-top: 20px;"> <span> <fmt:formatDate value="${news.publishTime}" pattern="MM-dd HH:mm"/> </span> <br /> </div> </c:forEach> </div> <br /><br /><br /><br /> <%@ include file="page.jsp" %> </div> </body> </html>
显示效果如下:
其中点击具体超链接的效果图如下:
任务1 到此完成,新闻显示工作结束。下面是搜索引擎部分。
搜索的工具类放置在com.zly.indexManager包下面
说明,本程序使用了庖丁解牛中文分词,用户使用时需要中文字典,我的字典放在了c:/dic下面,使用庖丁还需要配置环境变量PAODING_DIC_HOME , 其值为c:/dic , (就是你的字典文件所在的目录)
代码如下:
创建索引类IndexCreateUtil
package com.zly.indexManager; import java.io.File; import java.text.DateFormat; import java.text.SimpleDateFormat; import java.util.List; import net.paoding.analysis.analyzer.PaodingAnalyzer; import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.index.IndexWriter; import org.hibernate.SessionFactory; import org.hibernate.cfg.AnnotationConfiguration; import org.hibernate.cfg.Configuration; import org.hibernate.Session; import org.htmlparser.Parser; import com.zly.test.entity.NewsItem; public class IndexCreateUtil { @SuppressWarnings("unchecked") public void createIndexForNews() throws Exception { //存放索引的文件夹 File indexFile = new File("c:/index/news"); //使用了庖丁解牛分词器 Analyzer analyzer = new PaodingAnalyzer(); //使用索引文件夹,庖丁解牛分词器创建IndexWriter IndexWriter indexWriter = new IndexWriter(indexFile , analyzer , true); //从数据库中读取出所有的新闻记录以便进行索引的创建 Configuration cfg = new AnnotationConfiguration().configure(); SessionFactory factory = cfg.buildSessionFactory(); Session session = factory.openSession(); List<NewsItem> list = session.createQuery(" from NewsItem").list(); DateFormat format = new SimpleDateFormat("yyyy年MM月dd日 HH时mm分ss秒"); //对所有的新闻实体进行索引创建 for (NewsItem newsItem : list) { //建立一个lucene文档 Document doc = new Document(); //得到新闻标题 String newsTitle = newsItem.getNewsTitle(); //得到新闻内容 String newsContent = newsItem.getNewsContent(); //得到新闻事件 String publishDate = format.format(newsItem.getPublishTime()); //得到新闻主键id String id = newsItem.getId() + ""; //将新闻标题加入文档,因为要搜索和高亮,所以index是tokennized,TermVector是WITH_POSITIONS_OFFSETS doc.add(new Field("title" , newsTitle , Field.Store.YES , Field.Index.TOKENIZED , Field.TermVector.WITH_POSITIONS_OFFSETS)); //利用htmlparser得到新闻内容html的纯文本 Parser parser = new Parser(); parser.setInputHTML(newsContent); String strings = parser.parse(null).elementAt(0).toPlainTextString().trim(); //添加新闻内容至文档,与标题相似 doc.add(new Field("content" , strings , Field.Store.COMPRESS , Field.Index.TOKENIZED , Field.TermVector.WITH_POSITIONS_OFFSETS)); //添加时间至文档,因为要按照此字段降序排列排序,所以tokenzied,不用高亮所以TermVector是no就行了 doc.add(new Field("date" , publishDate , Field.Store.YES , Field.Index.TOKENIZED , Field.TermVector.NO)); //添加主键至文档,不分词,不高亮。 doc.add(new Field("id" , id , Field.Store.YES , Field.Index.NO , Field.TermVector.NO)); indexWriter.addDocument(doc); } //创建索引 indexWriter.optimize(); indexWriter.close(); //关闭session session.close(); } public static void main(String[] args) throws Exception { IndexCreateUtil util = new IndexCreateUtil(); util.createIndexForNews(); } }
对索引进行搜索的代码如下:
package com.zly.indexManager; import java.io.File; import java.util.ArrayList; import java.util.List; import net.paoding.analysis.analyzer.PaodingAnalyzer; import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.document.Document; import org.apache.lucene.index.IndexReader; import org.apache.lucene.queryParser.QueryParser; import org.apache.lucene.search.Hits; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.Query; import org.apache.lucene.search.Sort; import org.apache.lucene.search.highlight.Highlighter; import org.apache.lucene.search.highlight.QueryScorer; import org.apache.lucene.search.highlight.SimpleFragmenter; import org.apache.lucene.search.highlight.SimpleHTMLFormatter; import com.zly.test.entity.SearchResultBean; public class IndexSearchUtil { public List<SearchResultBean> getSearchResult(String searchWhich , String searchParam , int firstResult , int maxResult) throws Exception{ //索引所在文件夹 File indexFile = new File("c:/index/news"); //读取索引的indexReader IndexReader reader = IndexReader.open(indexFile); //庖丁解牛分词器 Analyzer analyzer = new PaodingAnalyzer(); //指定对content还是title进行查询 QueryParser parser = new QueryParser(searchWhich , analyzer); //创建indexSearcher IndexSearcher searcher = new IndexSearcher(reader); //对用户的输入进行查询 Query query = parser.parse(searchParam); //根据date字段进行排序,得到查询结果 Hits hits = searcher.search(query , new Sort("date" , true)); //创建list,将结果保存其中,以便在jsp页面中进行显示 List<SearchResultBean> list = new ArrayList<SearchResultBean>(); //模拟hibernate的serFirstResult和setMaxResult以便返回指定条目的结果 for (int i = firstResult - 1; i < firstResult + maxResult - 1; i++) { Document doc = hits.doc(i); //取得该条索引文档 SearchResultBean srb = new SearchResultBean(); //从中取出标题 String title = doc.get("title"); //从中取出内容 String content = doc.get("content"); //从中取出主键id String id = doc.get("id"); //从中取出发布时间 String date = doc.get("date"); //高亮htmlFormatter对象 SimpleHTMLFormatter sHtmlF = new SimpleHTMLFormatter("<b><font color='red'>", "</font></b>"); //高亮对象 Highlighter highlighter = new Highlighter(sHtmlF,new QueryScorer(query)); //设置高亮附近的字数 highlighter.setTextFragmenter(new SimpleFragmenter(100)); //如果查询的是标题,进行处理 if(searchWhich.equals("title")) { String bestFragment = highlighter.getBestFragment(analyzer,searchWhich,title); //获得高亮后的标题内容 srb.setTitle(bestFragment); //如果内容不足150个字,全部设置 if(content.length() < 150) { srb.setContent(content); }else { //如果内容多于150个字,只取出前面150个字 srb.setContent(content.substring(0 , 150)); } } else { //如果查询的是内容字段 String bestFragment = highlighter.getBestFragment(analyzer,searchWhich,content); //取得高亮内容并设置 srb.setContent(bestFragment); //设置标题,全部设置 srb.setTitle(title); } //设置日期 srb.setDate(date); //设置主键 srb.setId(id); //添加到list中,以便在jsp页面上显示 list.add(srb); } return list; } //取得符合搜索条件的所有记录总数,以便分页 , 与上面方法类似 public int getResultCount(String searchWhich , String searchParam) throws Exception { File indexFile = new File("c:/index/news"); IndexReader reader = IndexReader.open(indexFile); Analyzer analyzer = new PaodingAnalyzer(); QueryParser parser = new QueryParser(searchWhich , analyzer); IndexSearcher searcher = new IndexSearcher(reader); Query query = parser.parse(searchParam); Hits hits = searcher.search(query); return hits.length(); } }
分页action代码如下:
package com.zly.test.action; import java.util.List; import com.zly.indexManager.IndexSearchUtil; import com.zly.test.entity.PageControl; import com.zly.test.entity.SearchResultBean; public class SearchAction extends BaseAction { private static final long serialVersionUID = -2387037924517370511L; //查询索引实体类 private IndexSearchUtil indexSearcher; //对应搜索字段是标题还是内容 private String searchWhich; //对应用户输入的搜索内容 private String searchParam; //对应分页跳转到的页面 private String jumpPage; public String getJumpPage() { return jumpPage; } public void setJumpPage(String jumpPage) { this.jumpPage = jumpPage; } public String getSearchWhich() { return searchWhich; } public void setSearchWhich(String searchWhich) { this.searchWhich = searchWhich; } public String getSearchParam() { return searchParam; } public void setSearchParam(String searchParam) { this.searchParam = searchParam; } public String search() throws Exception { //如果为空,说明第一次进入分页 if(jumpPage == null) { jumpPage = "1"; } //从request范围内取得pageControl对象 PageControl pageControl = (PageControl) this.getRequest().getAttribute("pageControl"); //如果为空,则是第一次分页,创建分页对象,并且设置总的记录条数,以便设置最大页数 if(pageControl == null) { pageControl = new PageControl(); pageControl.setMaxRowCount((long)indexSearcher.getResultCount(searchWhich, searchParam)); pageControl.countMaxPage(); } //设置当前页 pageControl.setCurPage(Integer.parseInt(jumpPage)); //计算firstResult int firstResult = (pageControl.getCurPage() - 1) * pageControl.getRowsPerPage() + 1; //计算从当前条数算还有多少条记录 long left = pageControl.getMaxRowCount() - firstResult; int maxResult = -1; //如果剩余的记录数不如每页显示数,就设置maxResult为剩余条数 if(left < pageControl.getRowsPerPage()) { maxResult = Integer.valueOf(left + ""); //如果剩余记录数大于每页显示页数,就设置maxResult为每页条数 }else { maxResult = pageControl.getRowsPerPage(); } //取得查询结果集 List<SearchResultBean> userList = indexSearcher.getSearchResult(searchWhich, searchParam, firstResult, maxResult); //设置为pageControl pageControl.setData(userList); //将pageControl设置到request范围,以便在jsp现实结果 this.getRequest().setAttribute("pageControl", pageControl); //将searchWhich和searchParam设置到request范围,以便添加到分页jsp的form里面的hidden表单域,以便下次分页时,能够将值提交过来 this.getRequest().setAttribute("searchWhich", searchWhich); this.getRequest().setAttribute("searchParam", searchParam); //跳转到分页视图 return SUCCESS; } public IndexSearchUtil getIndexSearcher() { return indexSearcher; } public void setIndexSearcher(IndexSearchUtil indexSearcher) { this.indexSearcher = indexSearcher; } }
搜索的action在struts.xml中设置如下:
<action name="searchAction" class="searchAction" method="search"> <result>/searchResult.jsp</result> </action>
//searchResult.jsp代码如下:
原文地址:
http://blog.csdn.net/mmm123lmj/article/details/4625876