Hibernate search入门

原理是运用hibernate结合lucene来实现局部索引

1、引入jar包配置（Hibernate和Lucene包）

<dependency>  <groupId>org.hibernate</groupId>  <artifactId>hibernate-search-orm</artifactId>  <version>4.2.0.Final</version>  
</dependency> 
<dependency>  <groupId>org.apache.lucene</groupId>  <artifactId>lucene-smartcn</artifactId>  <version>3.6.2</version>  
</dependency>

3、给实体类增加注解（注意指定表的索引以及索引设定的分词器）

@Entity  
@Table(name = "PAGEINFO")  
@Indexed(index="PageInfo")/*标记该表可索引，参数index指定存放索引信息的文件名，路径在主配置文件中指定*/  
@Analyzer(impl=SmartChineseAnalyzer.class)//分词器  
public class Pageinfo implements java.io.Serializable {  private static final long serialVersionUID = 5454155825314635342L;  @Id  @GeneratedValue(generator = "custom-id")  @GenericGenerator(name = "custom-id", strategy = "uuid")  @Column(name = "ID", unique = true, nullable = false, insertable = true, updatable = true, length = 32)  @DocumentId  /*以字段id作为文档id*/  public java.lang.String getId() {  return this.id;  }  @Column(name = "TITLE", unique = false, nullable = true, insertable = true, updatable = true, length = 255)  @Field(store=Store.NO)  /*可索引，但不存储*/  public java.lang.String getTitle() {  return this.title;  }  @Column(name = "CONTENT", unique = false, nullable = true, insertable = true, updatable = true)  @Field(store=Store.NO)   /*可索引，但不存储*/  public java.lang.String getContent() {  return this.content;  }  @Column(name = "SOURCE", unique = false, nullable = true, insertable = true, updatable = true)  @Field(store=Store.NO)   /*可索引，但不存储*/  public java.lang.String getSource() {  return this.source;  }  @Column(name = "SUMMARY", unique = false, nullable = true, insertable = true, updatable = true)  @Field(store=Store.NO)   /*可索引，但不存储*/  public java.lang.String getSummary() {  return this.summary;  }  @ManyToOne(cascade = {}, fetch = FetchType.LAZY)  @JoinColumns({ @JoinColumn(name = "SITE_ID", nullable = false, insertable = false, updatable = false) })  @IndexedEmbedded(prefix="site_",depth=1)  /*关联检索，如field为site_name实则是按关联表的那么属性检索*/  public GrabageSiteconfig getGrabageSiteconfig() {  return grabageSiteconfig;  }  }  

4、使用API做单元测试

public class SearchTest {  private static SessionFactory sf;  @BeforeClass  public static void init() {  sf = HibernateConfigTest.sf;//弄一个SessionFactory，不多说  }  @Before  //执行索引  public void index(){  Session session = sf.openSession();  FullTextSession fullTextSession = Search.getFullTextSession(session);  //查出结果  List<Pageinfo> pageinfos = session.createCriteria(Pageinfo.class).list();  session.beginTransaction();  //依次建立索引  for (Iterator iterator = pageinfos.iterator(); iterator.hasNext();) {  Pageinfo pageinfo = (Pageinfo) iterator.next();  fullTextSession.index(pageinfo);  }  session.getTransaction().commit();  session.close();  System.out.println("index over......");  }  @Test  public void searchTest() {  Session session = sf.openSession();  FullTextSession fullTextSession = Search.getFullTextSession(session);  //在字段content中检索  QueryParser queryParser = new QueryParser(Version.LUCENE_36, "content", new SmartChineseAnalyzer(Version.LUCENE_36));  Query luceneqQuery=null;  try {  //检索含有“大风”的信息  luceneqQuery = queryParser.parse("大风");  } catch (ParseException e) {  e.printStackTrace();  }  //执行检索，得到结果集  FullTextQuery fullTextQuery = fullTextSession.createFullTextQuery(luceneqQuery, Pageinfo.class);  List<Pageinfo> pageinfos = fullTextQuery.list();  //查看结果做验证  for (Iterator iterator = pageinfos.iterator(); iterator.hasNext();) {  Pageinfo pageinfo = (Pageinfo) iterator.next();  System.out.println(pageinfo.getContent());  }  }  
}  

更详细的资料：

1、注解说明：

注解说明：@Indexed：让实体支持索引@Analyzer ：设置分词器，我这里使用的是开源的IK中文分词器@DocumentID：索引文档ID@Field ：索引字段，该注解默认属性值为store=Store.NO：是否将数据存储在索引中，经实验无论store=Store.NO还是store=Store.YES都不会影响最终的搜索。如果store=Store.NO值是通过数据库中获取，如果store=Store.YES值是直接从索引文档中获取。index=Index.YES：是否索引analyze=Analyze.YES：是否分词标注了注解后的实体在保存和更新的时候，会自动生成或修改索引。

2、查询索引

public PageModel<Article> searchArticle(int pageNum, int pageSize, String keyword) {FullTextSession fts = Search.getFullTextSession(sessionFactory.getCurrentSession());QueryBuilder qb = fts.getSearchFactory().buildQueryBuilder().forEntity(Article.class).get();Query luceneQuery = qb.keyword().onFields("title", "content", "description").matching(keyword).createQuery();FullTextQuery query = fts.createFullTextQuery(luceneQuery, Article.class);query.setFirstResult((pageNum - 1) * pageSize);query.setMaxResults(pageSize);List<Article> data = query.list();//封装分页数据PageModel<Article> model = new PageModel<>(pageNum, pageSize, data.size());//将数据高亮model.setData(SearchUtils.hightLight(luceneQuery, data, "title", "content", "description"));return model;
}

3、数据高亮

public static List<Article> hightLight(Query query, List<Article> data, String... fields) {List<Article> result = new ArrayList<Article>();Formatter formatter = new SimpleHTMLFormatter("<b style=\"color:red\">", "</b>");QueryScorer queryScorer = new QueryScorer(query);Highlighter highlighter = new Highlighter(formatter, queryScorer);// 使用IK中文分词Analyzer analyzer = new IKAnalyzer();for (Article a : data) {// 构建新的对象进行返回，避免页面错乱（我的页面有错乱）Article article = new Article();for (String fieldName : fields) {// 获得字段值，并给新的文章对象赋值Object fieldValue = ReflectionUtils.invokeMethod(BeanUtils.getPropertyDescriptor(Article.class, fieldName).getReadMethod(),a);ReflectionUtils.invokeMethod(BeanUtils.getPropertyDescriptor(Article.class, fieldName).getWriteMethod(),article, fieldValue);String hightLightFieldValue = null;try {hightLightFieldValue = highlighter.getBestFragment(analyzer, fieldName, String.valueOf(fieldValue));} catch (Exception e) {throw new RuntimeException("高亮显示关键字失败", e);}// 如果高亮成功则重新赋值if (hightLightFieldValue != null) {ReflectionUtils.invokeMethod(BeanUtils.getPropertyDescriptor(Article.class, fieldName).getWriteMethod(),article,hightLightFieldValue);}}// 赋值IDReflectionUtils.invokeMethod(BeanUtils.getPropertyDescriptor(Article.class, "id").getWriteMethod(),article, a.getId());result.add(article);}return result;
}

4、页面迭代显示

<s:iterator value="#request.pageModel.data">
<div class="article"><div class="article_title_area"><span class="article_title"><a href="${pageContext.request.contextPath }/article/show.action?id=${id }">${title }</a></span><span class="article_date">发表时间：<s:date name="postTime" format="yyyy-MM-dd HH:mm:ss"/></span></div><div class="article_content">${description }</div><div class="article_count_info"><span>阅读（${viewCount }）</span><span>评论（${comments.size() }）</span></div>
</div>
</s:iterator>