Lucene 3.6.2入门:针对索引文件的CRUD

作者: u010395804
发布时间:2015-07-08 16:57:29


    * @see =============================================================================================================
    * @see Lucene官网:http://lucene.Apache.org
    * @see Lucene下载http://archive.apache.org/dist/lucene/java/
    * @see Lucene文档:http://wiki.apache.org/lucene-java/
    * @see =============================================================================================================
    * @see 使用Luke查看分词信息(http://code.google.com/p/luke/
    * @see 1)引言:每一个Lucene版本都会有一个相应的Luke文件
    * @see 2)打开:双击或java -jar lukeall-3.5.0.jar
    * @see 3)选择索引的存放目录后点击OK即可
    * @see 7)如果我们的索引有改变,可以点击右侧的Re-open按钮重新载入索引
    * @see 4)Luke界面右下角的Top ranking terms窗口中显示的就是分词信息。其中Rank列表示出现频率
    * @see 5)Luke菜单下的Documents选项卡中显示的就是文档信息,我们可以根据文档序号来浏览(点击向左和向右的方向箭头)
    * @see 6)Luke菜单下的Search选项卡中可以根据我们输入的表达式来查文档内容
    * @see 比如在Enter search expression here:输入content:my,再在右侧点击一个黑色粗体字的Search大按钮即可
    * @see =============================================================================================================
    * @create Jun 30, 2012 4:34:09 PM
    *
    */

public class HelloIndex {     /*     * 定义一组数据,用来演示搜索(这里有一封邮件为例)     * 假设每一个变量代表一个Document,这里就定义了6个Document     */     //邮件编号     private String[] ids = {"1", "2", "3", "4", "5", "6"};     //邮件主题     private String[] names = {"Michael", "Scofield", "Tbag", "Jack", "Jade", "Jadyer"};     //邮件地址     private String[] emails = {"aa@jadyer.us", "bb@jadyer.cn", "cc@jadyer.cc", "dd@jadyer.tw", "ee@jadyer.hk", "ff@jadyer.me"};     //邮件内容     private String[] contents = {"my blog", "my website", "my name", "I am JavaDeveloper", "I am from Haerbin", "I like Lucene"};     //邮件附件(为数字和日期加索引,与,字符串加索引的方式不同)     private int[] attachs = {9,3,5,4,1,2};     //邮件日期     private Date[] dates = new Date[ids.length];     //它的创建是比较耗时耗资源的,所以这里只让它创建一次,此时reader处于整个生命周期中,实际应用中也可能直接放到ApplicationContext里面     private static IndexReader reader = null;     private Directory directory = null;     public HelloIndex(){     SimpleDateFormat sdf = new SimpleDateFormat("yyyyMMdd");     try {     dates[0] = sdf.parse("20120601");     dates[1] = sdf.parse("20120603");     dates[2] = sdf.parse("20120605");     dates[3] = sdf.parse("20120607");     dates[4] = sdf.parse("20120609");     dates[5] = sdf.parse("20120611");     directory = FSDirectory.open(new File("myExample/02_index/"));     } catch (Exception e) {     e.printStackTrace();     }     }     /**     * 获取IndexReader实例     */     private IndexReader getIndexReader(){     try {     if(reader == null){     reader = IndexReader.open(directory);     }else{     //if the index was changed since the provided reader was opened, open and return a new reader; else,return null     //如果当前reader在打开期间index发生改变,则打开并返回一个新的IndexReader,否则返回null     IndexReader ir = IndexReader.openIfChanged(reader);     if(ir != null){     reader.close(); //关闭原reader     reader = ir; //赋予新reader     }     }     return reader;     }catch(Exception e) {     e.printStackTrace();     }     return null; //发生异常则返回null     }     /**     * 通过IndexReader获取文档数量     */     public void getDocsCount(){     System.out.println("maxDocs:" + this.getIndexReader()。maxDoc());     System.out.println("numDocs:" + this.getIndexReader()。numDocs());     System.out.println("deletedDocs:" + this.getIndexReader()。numDeletedDocs());     }     /**     * 创建索引     */     public void createIndex(){     IndexWriter writer = null;     Document doc = null;     try{     writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_36, new StandardAnalyzer(Version.LUCENE_36)));     writer.deleteAll(); //创建索引之前,先把文档清空掉     for(int i=0; i     doc = new Document();     doc.add(new Field("id", ids[i], Field.Store.YES, Field.Index.NOT_ANALYZED_NO_NORMS));     doc.add(new Field("name", names[i], Field.Store.YES, Field.Index.NOT_ANALYZED_NO_NORMS));     doc.add(new Field("email", emails[i], Field.Store.YES, Field.Index.NOT_ANALYZED));     doc.add(new Field("content", contents[i], Field.Store.NO, Field.Index.ANALYZED));     doc.add(new NumericField("attach", Field.Store.YES, true)。setIntValue(attachs[i])); //为数字加索引(第三个参数指定是否索引)     doc.add(new NumericField("date", Field.Store.YES, true)。setLongValue(dates[i].getTime())); //为日期加索引     /*     * 建立索引时加权     * 定义排名规则,即加权,这里是为指定邮件名结尾的emails加权     */     if(emails[i].endsWith("jadyer.cn")){     doc.setBoost(2.0f);     }else if(emails[i].endsWith("jadyer.me")){     doc.setBoost(1.5f); //为文档加权…默认为1.0,权值越高则排名越高,显示得就越靠前     }else{     doc.setBoost(0.5f); //注意它的参数类型是Float     }     writer.addDocument(doc);     }     }catch(Exception e) {     e.printStackTrace();     }finally{     if(null != writer){     try {     writer.close();     } catch (IOException ce) {     ce.printStackTrace();     }     }     }     }     /**     * 搜索文件     */     public void searchFile(){     IndexSearcher searcher = new IndexSearcher(this.getIndexReader());     Query query = new TermQuery(new Term("content", "my")); //精确搜索:搜索"content"中包含"my"的文档     try{     TopDocs tds = searcher.search(query, 10);     for(ScoreDoc sd : tds.scoreDocs){     Document doc = searcher.doc(sd.doc); //sd.doc得到的是文档的序号     //doc.getBoost()得到的权值与创建索引时设置的权值之间是不相搭的,创建索引时的权值的查看需要使用Luke工具     // 之所以这样,是因为这里的Document对象(是获取到的)与创建索引时的Document对象,不是同一个对象     //sd.score得到的是该文档的评分,该评分规则的公式是比较复杂的,它主要与文档的权值和出现次数成正比     System.out.print("(" + sd.doc + "|" + doc.getBoost() + "|" + sd.score + ")" + doc.get("name") + "[" + doc.get("email") + "]-->");     System.out.println(doc.get("id") + "," + doc.get("attach") + "," + new SimpleDateFormat("yyyyMMdd")。format(new Date(Long.parseLong(doc.get("date")))));     }     }catch(Exception e){     e.printStackTrace();     }finally{     if(null != searcher){     try {     searcher.close();     } catch (IOException e) {     e.printStackTrace();     }     }     }     } /**     * 更新索引     * @see Lucene其实并未提供更新索引的方法,这里的更新操作内部是先删除再添加的方式     * @see 因为Lucene认为更新索引的代价,与删除后重建索引的代价,二者是差不多的     */     public void updateIndex(){     IndexWriter writer = null;     Document doc = new Document();     try{     writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_36, new StandardAnalyzer(Version.LUCENE_36)));     doc.add(new Field("id", "1111", Field.Store.YES, Field.Index.NOT_ANALYZED_NO_NORMS));     doc.add(new Field("name", names[0], Field.Store.YES, Field.Index.NOT_ANALYZED_NO_NORMS));     doc.add(new Field("email", emails[0], Field.Store.YES, Field.Index.NOT_ANALYZED));     doc.add(new Field("content", contents[0], Field.Store.NO, Field.Index.ANALYZED));     doc.add(new NumericField("attach", Field.Store.YES, true)。setIntValue(attachs[0]));     doc.add(new NumericField("date", Field.Store.YES, true)。setLongValue(dates[0].getTime()));     //其实它会先删除索引文档中id为1的文档,然后再将这里的doc对象重新索引,所以即便这里的1!=1111,但它并不会报错     //所以在执行完该方法后:maxDocs=7,numDocs=6,deletedDocs=1,就是因为Lucene会先删除再添加     writer.updateDocument(new Term("id","1"), doc);     }catch(Exception e) {     e.printStackTrace();     }finally{     if(null != writer){     try {     writer.close();     } catch (IOException ce) {     ce.printStackTrace();     }     }     }     }     /**     * 删除索引     * @see -----------------------------------------------------------------------------------------------------     * @see 在执行完该方法后,再执行本类的searchFile()方法,得知numDocs=5,maxDocs=6,deletedDocs=1     * @see 这说明此时删除的文档并没有被完全删除,而是存储在一个回收站中,它是可以恢复的     * @see -----------------------------------------------------------------------------------------------------     * @see 从回收站中清空索引IndexWriter     * @see 对于清空索引,Lucene3.5之前叫做优化,调用的是IndexWriter.optimize()方法,但该方法已被禁用     * @see 因为optimize时它会全部更新索引,这一过程所涉及到的负载是很大的,于是弃用了该方法,使用forceMerge代替     * @see 使用IndexWriter.forceMergeDeletes()方法可以强制清空回收站中的内容     * @see 另外IndexWriter.forceMerge(3)方法会将索引合并为3段,这3段中的被删除的数据也会被清空     * @see 但其在Lucene3.5之后不建议使用,因为其会消耗大量的开销,而Lucene会根据情况自动处理的     * @see -----------------------------------------------------------------------------------------------------     */     public void deleteIndex(){     IndexWriter writer = null;     try{     writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_36, new StandardAnalyzer(Version.LUCENE_36)));     //其参数可以传Query或Term…Query指的是可以查询出一系列的结果并将其全部删掉,而Term属于精确查找     writer.deleteDocuments(new Term("id", "1")); //删除索引文档中id为1的文档     }catch(Exception e) {     e.printStackTrace();     }finally{     if(null != writer){     try {     writer.close();     } catch (IOException ce) {     ce.printStackTrace();     }     }     }     }     /**     * 恢复索引     * @see 建议弃用     */     @Deprecated     public void unDeleteIndex(){     IndexReader reader = null;     try {     //IndexReader.open(directory)此时该IndexReader默认的readOnly=true,而在恢复索引时应该指定其为非只读的     reader = IndexReader.open(directory, false);     //Deprecated. Write support will be removed in Lucene 4.0. There will be no replacement for this method.     reader.undeleteAll();     } catch (Exception e) {     e.printStackTrace();     }finally{     if(null != reader){     try {     reader.close();     } catch (IOException e) {     e.printStackTrace();     }     }     }     }     }
 下面是用JUnit4.x写的一个小测试
    双击代码全选
    package com.jadyer.test;     import org.junit.After;     import org.junit.Before;     import org.junit.Test;     import com.jadyer.lucene.HelloIndex;     public class HelloIndexTest {     private HelloIndex hello;     @Before     public void init(){     hello = new HelloIndex();     }     @After     public void destroy(){     hello.getDocsCount();     }     @Test     public void createIndex(){     hello.createIndex();     }     @Test     public void searchFile(){     hello.searchFile();     }     @Test     public void updateIndex(){     hello.updateIndex();     }     @Test     public void deleteIndex(){     hello.deleteIndex();     }     @Test     @SuppressWarnings("deprecation")     public void unDeleteIndex(){     hello.unDeleteIndex();     }     }

版权声明:本文为博主原创文章,未经博主允许不得转载。

标签: Lucene
来源:http://blog.csdn.net/u010395804/article/details/36410403

推荐: