标签:lucene
lucene通过NRTManager这个类来实现近实时搜索,所谓近实时搜索即在索引发生改变时,通
过线程跟踪,在相对很短的时间反映给给用户程序的调用
NRTManager通过管理IndexWriter对象,并将IndexWriter的一些方法(增删改)例如
addDocument,deleteDocument等方法暴露给客户调用,它的操作全部在内存里面,所以如果
你不调用IndexWriter的commit方法,通过以上的操作,用户硬盘里面的索引库是不会变化的,所
以你每次更新完索引库请记得commit掉,这样才能将变化的索引一起写到硬盘中,实现索引更新后的同步
用户每次获取最新索引(IndexSearcher),可以通过两种方式,第一种是通过调用
NRTManagerReopenThread对象,该线程负责实时跟踪索引内存的变化,每次变化就调用
maybeReopen方法,保持最新代索引,打开一个新的IndexSearcher对象,而用户所要的
IndexSearcher对象是NRTManager通过调用getSearcherManager方法获得SearcherManager对
象,然后通过SearcherManager对象获取IndexSearcher对象返回个客户使用,用户使用完之
后调用SearcherManager的release释放IndexSearcher对象,最后记得关闭NRTManagerReopenThread;
第二种方式是不通过NRTManagerReopenThread对象,而是直接调用NRTManager的
maybeReopen方法来获取最新的IndexSearcher对象来获取最新索引
一:单独使用SearchManager
1.// 实现近实时搜索//所以就不通过Indexread来创建search
private SearcherManager manager;//是线程安全的
2.在构造方法中实现SearchManager
// 创建SearcherManage manager = new SearcherManager(directory, new SearcherWarmer() { @Override public void warm(IndexSearcher arg0) throws IOException { System.out.println("索引改变了..."); } }, Executors.newCachedThreadPool());
// 获得一个search IndexSearcher search = manager.acquire();
manager.maybeReopen();
finally { try { // 通过manager关闭search manager.release(search); } catch (IOException e) { e.printStackTrace(); }
writer.deleteDocuments(new Term("id", "1")); // 使用SearchManage需要对writer进行commit writer.commit();
package org.itat.index; import java.io.File; import java.io.IOException; import java.util.concurrent.Executors; import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.analysis.standard.StandardAnalyzer; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.document.NumericField; import org.apache.lucene.index.CorruptIndexException; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.IndexWriter; import org.apache.lucene.index.IndexWriterConfig; import org.apache.lucene.index.Term; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.SearcherManager; import org.apache.lucene.search.SearcherWarmer; import org.apache.lucene.search.TermQuery; import org.apache.lucene.search.TopDocs; import org.apache.lucene.store.Directory; import org.apache.lucene.store.FSDirectory; import org.apache.lucene.store.LockObtainFailedException; import org.apache.lucene.util.Version; /** * 只使用SearcherManager,不使用NRTManager * * SearcherManager的maybeReopen会自动检查是否需要重新打开,比如重复执行search02几次, * 中间的一次删除一条数据这个删除的数据需要对writer进行commit才行,这样硬盘上的索引才会生效 * 那么使用maybeReopen就可以检测到硬盘中的索引是否改变,并在下次查询的时候就进行生效 * * 但是: * 光使用SearcherManager的话做不到实时搜索,为什么呢? * 因为使用SearcherManager需要进行writer.commit才会检测到,但是我们知道writer的commit是非常 * 消耗性能的,我们不能经常性的commit,那需要怎么做呢? * 我们只能把添加修改删除的操作在内存中生效,然后使用内存中的索引信息并且在搜索时能起到效果, * 过一段时间累计到一定程序才进行writer.commit */ public class IndexUtil1 { private String[] ids = {"1","2","3","4","5","6"}; private String[] emails = {"aa@itat.org","bb@itat.org","cc@cc.org","dd@sina.org","ee@zttc.edu","ff@itat.org"}; private String[] contents = { "welcome to visited the space,I like book", "hello boy, I like pingpeng ball", "my name is cc I like game", "I like football", "I like football and I like basketball too", "I like movie and swim" }; private int[] attachs = {2,3,1,4,5,5}; private String[] names = {"zhangsan","lisi","john","jetty","mike","jake"}; private Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_35); private SearcherManager mgr = null;//是线程安全的 private Directory directory = null; public IndexUtil1() { try { directory = FSDirectory.open(new File("D:\\Workspaces\\realtime\\index")); mgr = new SearcherManager( directory, new SearcherWarmer() { /** * 索引一更新就要重新获取searcher,那获取searcher的时候就会调用这个方法 * 执行maybeReopen的时候会执行warm方法,在这里可以对资源等进行控制 */ @Override public void warm(IndexSearcher search) throws IOException { System.out.println("has change"); } }, Executors.newCachedThreadPool() ); } catch (IOException e) { e.printStackTrace(); } } /** * 删除索引数据,默认不会完全删除,被放入索引回收站 */ public void delete(String id) { IndexWriter writer = null; try { writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_35,analyzer)); //参数是一个选项,可以是一个Query,也可以是一个term,term是一个精确查找的值 //此时删除的文档并不会被完全删除,而是存储在一个回收站中的,可以恢复 //执行完这个操作,索引文件夹下就会多出一个名叫_0_1.del的文件,也就是删除的文件在这个文件中记录了 writer.deleteDocuments(new Term("id",id)); writer.commit(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (LockObtainFailedException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { try { if(writer!=null) writer.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } } public void query() { try { IndexReader reader = IndexReader.open(directory); //通过reader可以有效的获取到文档的数量 System.out.println("numDocs:"+reader.numDocs());//存储的文档数//不包括被删除的 System.out.println("maxDocs:"+reader.maxDoc());//总存储量,包括在回收站中的索引 System.out.println("deleteDocs:"+reader.numDeletedDocs()); reader.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } /** * 索引文件后缀为.fmn为保存的是域的名称等 * .fdt和.fdx保存的是Store.YES的信息,保存域里面存储的数据 * .frq表示这里的域哪些出现多少次,哪些单词出现多少次, * .nrm存储一些评分信息 * .prx存储一些偏移量等 * .tii和.tis专门存储索引里面的所有内容信息 */ public void index() { IndexWriter writer = null; try { //在2.9版本之后,lucene的就不是全部的索引格式都兼容的了,所以在使用的时候必须写明版本号 writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_35, analyzer)); writer.deleteAll();//清空索引 Document doc = null; for(int i=0;i<ids.length;i++) { doc = new Document(); doc.add(new Field("id",ids[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("email","test"+i+"@test.com",Field.Store.YES,Field.Index.NOT_ANALYZED)); doc.add(new Field("content",contents[i],Field.Store.NO,Field.Index.ANALYZED)); doc.add(new Field("name",names[i],Field.Store.YES,Field.Index.NOT_ANALYZED_NO_NORMS)); //存储数字 //NumberTools.stringToLong("");已经被标记为过时了 doc.add(new NumericField("attach",Field.Store.YES,true).setIntValue(attachs[i])); String et = emails[i].substring(emails[i].lastIndexOf("@")+1); writer.addDocument(doc); } } catch (CorruptIndexException e) { e.printStackTrace(); } catch (LockObtainFailedException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { try { if(writer!=null)writer.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } } public void search02() { IndexSearcher searcher = mgr.acquire();//获得一个searcher try { /** * maybeReopen会自动检查是否需要重新打开 * 比如重复执行search02几次,中间一次删除一条数据 * 这个删除的数据需要对writer进行commit才行 * 那么使用maybeReopen就可以检测到硬盘中的索引是否改变 * 并在下次查询的时候把删除的这条给去掉 */ mgr.maybeReopen(); TermQuery query = new TermQuery(new Term("content","like")); TopDocs tds = searcher.search(query, 10); for(ScoreDoc sd:tds.scoreDocs) { Document doc = searcher.doc(sd.doc); System.out.println(doc.get("id")+"---->"+ doc.get("name")+"["+doc.get("email")+"]-->"+doc.get("id")+","+ doc.get("attach")+","+doc.get("date")+","+doc.getValues("email")[1]); } searcher.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally{ try { mgr.release(searcher);//释放一个searcher } catch (IOException e) { e.printStackTrace(); } } } }
package org.itat.index; import java.io.File; import java.io.IOException; import org.apache.lucene.analysis.Analyzer; import org.apache.lucene.analysis.standard.StandardAnalyzer; import org.apache.lucene.document.Document; import org.apache.lucene.document.Field; import org.apache.lucene.document.NumericField; import org.apache.lucene.index.CorruptIndexException; import org.apache.lucene.index.IndexReader; import org.apache.lucene.index.IndexWriter; import org.apache.lucene.index.IndexWriterConfig; import org.apache.lucene.index.Term; import org.apache.lucene.search.IndexSearcher; import org.apache.lucene.search.NRTManager; import org.apache.lucene.search.NRTManagerReopenThread; import org.apache.lucene.search.ScoreDoc; import org.apache.lucene.search.SearcherManager; import org.apache.lucene.search.SearcherWarmer; import org.apache.lucene.search.TermQuery; import org.apache.lucene.search.TopDocs; import org.apache.lucene.store.Directory; import org.apache.lucene.store.FSDirectory; import org.apache.lucene.store.LockObtainFailedException; import org.apache.lucene.util.Version; /** * SearcherManager和NRTManager联合使用 * * SearcherManager的maybeReopen会自动检查是否需要重新打开,比如重复执行search02几次, * 中间的一次删除一条数据这个删除的数据需要对writer进行commit才行,这样硬盘上的索引才会生效 * 那么使用maybeReopen就可以检测到硬盘中的索引是否改变,并在下次查询的时候就进行生效 * * 但是: * 光使用SearcherManager的话做不到实时搜索,为什么呢? * 因为使用SearcherManager需要进行writer.commit才会检测到,但是我们知道writer的commit是非常 * 消耗性能的,我们不能经常性的commit,那需要怎么做呢? * 我们只能把添加修改删除的操作在内存中生效,然后使用内存中的索引信息并且在搜索时能起到效果, * 过一段时间累计到一定程序才进行writer.commit * NRTManage就是这样的功能,把更新的数据存储在内容中,但是lucene搜索的时候也可以搜索到,需要 * writer进行commit才会把索引更新到硬盘中 */ public class IndexUtil2 { private String[] ids = {"1","2","3","4","5","6"}; private String[] emails = {"aa@itat.org","bb@itat.org","cc@cc.org","dd@sina.org","ee@zttc.edu","ff@itat.org"}; private String[] contents = { "welcome to visited the space,I like book", "hello boy, I like pingpeng ball", "my name is cc I like game", "I like football", "I like football and I like basketball too", "I like movie and swim" }; private int[] attachs = {2,3,1,4,5,5}; private String[] names = {"zhangsan","lisi","john","jetty","mike","jake"}; private Analyzer analyzer = new StandardAnalyzer(Version.LUCENE_35); private SearcherManager mgr = null;//是线程安全的 private NRTManager nrtMgr = null;//near real time 近实时搜索 private Directory directory = null; private IndexWriter writer = null; public IndexUtil2() { try { directory = FSDirectory.open(new File("D:\\Workspaces\\realtime\\index")); writer = new IndexWriter(directory,new IndexWriterConfig(Version.LUCENE_35, analyzer)); nrtMgr = new NRTManager(writer, new SearcherWarmer() { /** * 索引一更新就要重新获取searcher,那获取searcher的时候就会调用这个方法 * 执行maybeReopen的时候会执行warm方法,在这里可以对资源等进行控制 */ @Override public void warm(IndexSearcher search) throws IOException { System.out.println("has open"); } } ); //启动NRTManager的Reopen线程 //NRTManagerReopenThread会每隔25秒去检测一下索引是否更新并判断是否需要重新打开writer NRTManagerReopenThread reopen = new NRTManagerReopenThread(nrtMgr, 5.0, 0.025);//0.025为25秒 reopen.setDaemon(true);//设为后台线程 reopen.setName("NrtManager Reopen Thread"); reopen.start(); mgr = nrtMgr.getSearcherManager(true);//true为允许所有的更新 } catch (IOException e) { e.printStackTrace(); } } /** * 删除索引数据,默认不会完全删除,被放入索引回收站 */ public void delete(String id) { try { //参数是一个选项,可以是一个Query,也可以是一个term,term是一个精确查找的值 //此时删除的文档并不会被完全删除,而是存储在一个回收站中的,可以恢复 //执行完这个操作,索引文件夹下就会多出一个名叫_0_1.del的文件,也就是删除的文件在这个文件中记录了 nrtMgr.deleteDocuments(new Term("id",id));//使用使用nrtMgr来删除 } catch (CorruptIndexException e) { e.printStackTrace(); } catch (LockObtainFailedException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } public void query() { try { IndexReader reader = IndexReader.open(directory); //通过reader可以有效的获取到文档的数量 System.out.println("numDocs:"+reader.numDocs());//存储的文档数//不包括被删除的 System.out.println("maxDocs:"+reader.maxDoc());//总存储量,包括在回收站中的索引 System.out.println("deleteDocs:"+reader.numDeletedDocs()); reader.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } /** * 索引文件后缀为.fmn为保存的是域的名称等 * .fdt和.fdx保存的是Store.YES的信息,保存域里面存储的数据 * .frq表示这里的域哪些出现多少次,哪些单词出现多少次, * .nrm存储一些评分信息 * .prx存储一些偏移量等 * .tii和.tis专门存储索引里面的所有内容信息 */ public void index() { IndexWriter writer = null; try { //在2.9版本之后,lucene的就不是全部的索引格式都兼容的了,所以在使用的时候必须写明版本号 writer = new IndexWriter(directory, new IndexWriterConfig(Version.LUCENE_35, analyzer)); writer.deleteAll();//清空索引 Document doc = null; for(int i=0;i<ids.length;i++) { doc = new Document(); doc.add(new Field("id",ids[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("email","test"+i+"@test.com",Field.Store.YES,Field.Index.NOT_ANALYZED)); doc.add(new Field("content",contents[i],Field.Store.NO,Field.Index.ANALYZED)); doc.add(new Field("name",names[i],Field.Store.YES,Field.Index.NOT_ANALYZED_NO_NORMS)); //存储数字 //NumberTools.stringToLong("");已经被标记为过时了 doc.add(new NumericField("attach",Field.Store.YES,true).setIntValue(attachs[i])); String et = emails[i].substring(emails[i].lastIndexOf("@")+1); writer.addDocument(doc); } } catch (CorruptIndexException e) { e.printStackTrace(); } catch (LockObtainFailedException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally { try { if(writer!=null)writer.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } } } public void search02() { IndexSearcher searcher = mgr.acquire();//获得一个searcher try { /** * maybeReopen会自动检查是否需要重新打开 * 比如重复执行search02几次,中间一次删除一条数据 * 这个删除的数据需要对writer进行commit才行 * 那么使用maybeReopen就可以检测到硬盘中的索引是否改变 * 并在下次查询的时候把删除的这条给去掉 */ TermQuery query = new TermQuery(new Term("content","like")); TopDocs tds = searcher.search(query, 10); for(ScoreDoc sd:tds.scoreDocs) { Document doc = searcher.doc(sd.doc); System.out.println(doc.get("id")+"---->"+ doc.get("name")+"["+doc.get("email")+"]-->"+doc.get("id")+","+ doc.get("attach")+","+doc.get("date")+","+doc.getValues("email")[1]); } searcher.close(); } catch (CorruptIndexException e) { e.printStackTrace(); } catch (IOException e) { e.printStackTrace(); } finally{ try { mgr.release(searcher);//释放一个searcher } catch (IOException e) { e.printStackTrace(); } } } }
标签:lucene
原文地址:http://blog.csdn.net/u010366796/article/details/45022397