使用Solrj管理Solr索引

本文主要是介绍使用Solrj管理Solr索引，希望对大家解决编程问题提供一定的参考价值，需要的开发者们随着小编来一起学习吧！

Solrj是Solr搜索服务器的一个比较基础的客户端工具，可以非常方便地与Solr搜索服务器进行交互，最基本的功能就是管理Solr索引，包括添加、更新、删除和查询等。对于一些比较基础的应用，用Solj基本够用，而且你可以非常容易地通过使用Solrj的API实现与Solr搜索服务器进行交互，实现对Solr的基本管理功能。如果你的应用比较复杂，可以扩展Solrj来满足需要。

下面是一个使用Solrj的API实现与Solr服务器交互的工具类SolrPostServer，能够实现索引的添加、更新、删除和查询功能。SolrPostServer类中两个内部类是与访问MongoDB的配置和工具。

在实际应用中，对于是否进行commit，可以有两种方式：

• 一种是直接在客户端进行计算，亦即，进行索引时计算添加的文档数，满足设置的值则进行手动commit，这种方式比较灵活，你可以根据搜索服务器的运行状况选择合理的commit文档数量；
• 另一种是，直接在Solr搜索服务器上进行配置，一般来说，对索引进行大批量更新，一般不会选择在搜索服务器业务繁忙的时候进行，所以能够自动进行commit也便利了对索引的管理，更新文档可以完全可以实现自动化处理。

	<requestHandler name="/update" class="solr.XmlUpdateRequestHandler"><maxPendingDeletes>10000</maxPendingDeletes><autoCommit><maxDocs>20</maxDocs><maxTime>86000</maxTime></autoCommit></requestHandler>

实现代码如下所示：

package org.shirdrn.solr.solrj;import java.io.IOException;
import java.io.Serializable;
import java.net.MalformedURLException;
import java.util.ArrayList;
import java.util.Collection;
import java.util.HashMap;
import java.util.Iterator;
import java.util.List;
import java.util.Map;import org.apache.commons.httpclient.HttpClient;
import org.apache.log4j.Logger;
import org.apache.lucene.document.Document;
import org.apache.solr.client.solrj.ResponseParser;
import org.apache.solr.client.solrj.SolrServerException;
import org.apache.solr.client.solrj.impl.CommonsHttpSolrServer;
import org.apache.solr.client.solrj.impl.XMLResponseParser;
import org.apache.solr.common.SolrDocument;
import org.apache.solr.common.SolrDocumentList;
import org.apache.solr.common.SolrInputDocument;
import org.apache.solr.common.params.SolrParams;import com.mongodb.BasicDBObject;
import com.mongodb.DBCollection;
import com.mongodb.DBCursor;
import com.mongodb.DBObject;
import com.mongodb.Mongo;
import com.mongodb.MongoException;/*** Solr server for indexes operations.* * @author shirdrn* @date   2011-12-20*/
public class SolrPostServer {private static final Logger LOG = Logger.getLogger(SolrPostServer.class);private CommonsHttpSolrServer server; private ResponseParser responseParser;private MongoConfig mongoConfig;private String[] collectionNames;private  int maxCommitCount = 100;private boolean manualOptimize = true;private boolean manualCommit = false;private Collection<SolrInputDocument> docContainer = new ArrayList<SolrInputDocument>();private static int totalCount = 0;public SolrPostServer(String url, HttpClient httpClient, MongoConfig mongoConfig) {try {if(httpClient==null) {server = new CommonsHttpSolrServer(url);server.setSoTimeout(500000);  // socket read timeoutserver.setConnectionTimeout(5000);  server.setDefaultMaxConnectionsPerHost(10);  server.setMaxTotalConnections(100);server.setAllowCompression(true);  server.setMaxRetries(1); // defaults to 0.  > 1 not recommended. } else {server = new CommonsHttpSolrServer(url, httpClient);}} catch (MalformedURLException e) {e.printStackTrace();}this.mongoConfig = mongoConfig;initialize();}/*** Initialize the {@link CommonsHttpSolrServer}'s basic parameters.*/private void initialize() {if(responseParser!=null) {server.setParser(responseParser);} else {server.setParser(new XMLResponseParser());}}@SuppressWarnings("unchecked")public void postUpdate() {DBCursor cursor = null;try {for (String c : collectionNames) {LOG.info("MongoDB collection name: " + c);DBCollection collection = MongoHelper.newHelper(mongoConfig).getCollection(c);DBObject q = new BasicDBObject();cursor = collection.find(q);while(cursor.hasNext()) {try {Map<Object, Object> m = cursor.next().toMap();if(manualCommit) {add(m, true);} else {add(m, false);}++totalCount;LOG.info("Add fragment: _id = " + m.get("_id").toString());} catch (IOException e) {e.printStackTrace();}}cursor.close();}LOG.info("Add totalCount: " + totalCount);finallyCommit();optimize(manualOptimize);} catch (MongoException e) {e.printStackTrace();} catch (SolrServerException e) {e.printStackTrace();} catch (IOException e) {e.printStackTrace();}}/*** Detele lucene {@link Document} by IDs.* @param strings*/public void deleteById(List<String> strings) {try {server.deleteById(strings);} catch (SolrServerException e) {e.printStackTrace();} catch (IOException e) {e.printStackTrace();}}/*** Detele lucene {@link Document} by query.* @param query*/public void deleteByQuery(String query) {try {server.deleteByQuery(query);} catch (SolrServerException e) {e.printStackTrace();} catch (IOException e) {e.printStackTrace();}}/*** Query.* @param params* @param fields* @return*/public List<Map<String, Object>> query(SolrParams params, String[] fields) {List<Map<String, Object>> results = new ArrayList<Map<String, Object>>();try {SolrDocumentList documents = server.query(params).getResults();Iterator<SolrDocument> iter = documents.iterator();while(iter.hasNext()) {SolrDocument doc = iter.next();Map<String, Object> map = new HashMap<String, Object>();for(String field : fields) {map.put(field, doc.getFieldValue(field));}results.add(map);}} catch (SolrServerException e) {e.printStackTrace();}return results;}/*** When controlling the committing action at client side, finally execute committing.* @throws SolrServerException* @throws IOException*/private void finallyCommit() throws SolrServerException, IOException {if(!docContainer.isEmpty()) {server.add(docContainer);commit(false, false);}}/*** Commit.* @param waitFlush* @param waitSearcher* @throws SolrServerException* @throws IOException*/public void commit(boolean waitFlush, boolean waitSearcher) {try {server.commit(waitFlush, waitSearcher);} catch (SolrServerException e) {e.printStackTrace();} catch (IOException e) {e.printStackTrace();}}/*** When controlling the optimizing action at client side, finally execute optimizing.* @param waitFlush* @param waitSearcher* @throws SolrServerException* @throws IOException*/public void optimize(boolean waitFlush, boolean waitSearcher) {try {server.optimize(waitFlush, waitSearcher);commit(waitFlush, waitSearcher);} catch (Exception e) {LOG.error("Encounter error when optimizing.",  e);try {server.rollback();} catch (SolrServerException e1) {e1.printStackTrace();} catch (IOException e1) {e1.printStackTrace();}}}/*** Optimize.* @param optimize* @throws SolrServerException* @throws IOException*/private void optimize(boolean optimize) {if(optimize) {optimize(true, true);}}/*** Add a {@link SolrInputDocument} or collect object and add to the a collection for batch updating* from a mongodb's recored, a Map object.* @param m* @param oneByOne* @throws SolrServerException* @throws IOException*/private void add(Map<Object, Object> m, boolean oneByOne) throws SolrServerException, IOException {SolrInputDocument doc = createDocument(m);if(oneByOne) {server.add(doc);} else {docContainer.add(doc);if(docContainer.size()>maxCommitCount) {server.add(docContainer);server.commit(false, false);docContainer = new ArrayList<SolrInputDocument>();}}}/*** Create a {@link SolrInputDocument} object.* @param record* @return*/private SolrInputDocument createDocument(Map<Object, Object> record) {String id = record.get("_id").toString();String articleId = (String) record.get("articleId");String title = (String) record.get("title");String url = (String) record.get("url");String spiderName = (String) record.get("spiderName");String fragment = makeFragment((BasicDBObject) record.get("fragment"));String word = (String) record.get("word");int pictureCount = (Integer) record.get("pictureCount");int selectedCount = (Integer) record.get("selectedCount");int fragmentSize = (Integer) record.get("fragmentSize");SolrInputDocument doc = new SolrInputDocument();doc.addField("_id", id, 1.0f);doc.addField("articleId", articleId, 1.0f);doc.addField("title", title, 1.0f);doc.addField("url", url, 1.0f);doc.addField("spiderName", spiderName, 1.0f);doc.addField("fragment", fragment, 1.0f);doc.addField("word", word, 1.0f);// Additional processing for lucene payload metadata.doc.addField("pictureCount", word + "|" + pictureCount);doc.addField("coverage", word + "|" + (float)selectedCount/fragmentSize);return doc;}@SuppressWarnings("unchecked")private String makeFragment(BasicDBObject fragment) {StringBuilder builder = new StringBuilder();Iterator<Map.Entry<Integer, String>> iter = fragment.toMap().entrySet().iterator();while(iter.hasNext()) {Map.Entry<Integer, String> entry = iter.next();builder.append(entry.getValue().trim()).append("<br>");}return builder.toString();}/*** Set {@link ResponseParser}, default value is {@link XMLResponseParser}.* @param responseParser*/public void setResponseParser(ResponseParser responseParser) {this.responseParser = responseParser;}/*** Pulling document resource from multiple collections of MongoDB.* @param collectionNames*/public void setCollectionNames(String[] collectionNames) {this.collectionNames = collectionNames;}public void setMaxCommitCount(int maxCommitCount) {this.maxCommitCount = maxCommitCount;}public void setManualCommit(boolean manualCommit) {this.manualCommit = manualCommit;}public void setManualOptimize(boolean manualOptimize) {this.manualOptimize = manualOptimize;}/*** Mongo database configuration.* * @author shirdrn* @date   2011-12-20*/public static class MongoConfig implements Serializable {private static final long serialVersionUID = -3028092758346115702L;private String host;private int port;private String dbname;private String collectionName;public MongoConfig(String host, int port, String dbname, String collectionName) {super();this.host = host;this.port = port;this.dbname = dbname;this.collectionName = collectionName;}@Overridepublic boolean equals(Object obj) {MongoConfig other = (MongoConfig) obj;return host.equals(other.host) && port==other.port&& dbname.equals(other.dbname) && collectionName.equals(other.collectionName);}}/*** Mongo database utility.* * @author shirdrn* @date   2011-12-20*/static class MongoHelper {private static Mongo mongo;private static MongoHelper helper;private MongoConfig mongoConfig;private MongoHelper(MongoConfig mongoConfig) {super();this.mongoConfig = mongoConfig;}public synchronized static MongoHelper newHelper(MongoConfig mongoConfig) {try {if(helper==null) {helper = new MongoHelper(mongoConfig);mongo = new Mongo(mongoConfig.host, mongoConfig.port);Runtime.getRuntime().addShutdownHook(new Thread() {@Overridepublic void run() {if(mongo!=null) {mongo.close();}}});}} catch (Exception e) {e.printStackTrace();}return helper;}			public DBCollection getCollection(String collectionName) {DBCollection c = null;try {c = mongo.getDB(mongoConfig.dbname).getCollection(collectionName);} catch (Exception e) {e.printStackTrace();}return c;}	}
}

下面，我们可以通过写一个测试用例测试一下。

首先，我的Solr搜索服务器已经部署好并启动成功，对应的url为http://192.168.0.197:8080/server/fragment/。测试用例如下所示：

package org.shirdrn.solr.solrj;import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;import junit.framework.TestCase;import org.apache.solr.common.params.CommonParams;
import org.apache.solr.common.params.SolrParams;
import org.apache.solr.request.MapSolrParams;
import org.shirdrn.solr.solrj.SolrPostServer.MongoConfig;@SuppressWarnings("deprecation")
public class TestSolrPostServer extends TestCase {SolrPostServer myServer;MongoConfig config;String url;String[] collectionNames;@Overrideprotected void setUp() throws Exception {url = "http://192.168.0.197:8080/server/fragment/";config = new MongoConfig("192.168.0.184", 27017, "fragment", "");myServer = new SolrPostServer(url, null, config);myServer.setMaxCommitCount(100);}@Overrideprotected void tearDown() throws Exception {super.tearDown();}public void testPostUpdate() {collectionNames = new String[] {"sina","lvping","daodao","go2eu","mafengwo","lotour","17u","sohu","baseSe","bytravel"};myServer.setCollectionNames(collectionNames);myServer.setManualCommit(true);myServer.setManualOptimize(false);myServer.postUpdate();}public void testPostDelete() {List<String> strings = new ArrayList<String>();strings.add("4ef051342c4117a38f63ee97");strings.add("4ef051322c4117a38f63ee36");strings.add("4ef051a42c4117a38f63fb51");strings.add("4ef050d92c4117a38f63dda4");strings.add("4ef051fe2c4117a38f640bc8");strings.add("4ef048ef2c4117a38f6207ce");strings.add("4ef049062c4117a38f620e13");strings.add("4ef046f12c4117a38f6185c0");myServer.deleteById(strings);myServer.commit(false, false);myServer.optimize(true, false);}@SuppressWarnings({ "rawtypes", "unchecked" })public void testQuery() {Map map = new HashMap();map.put(CommonParams.Q, "法国");map.put(CommonParams.START, "0");map.put(CommonParams.ROWS, "10");map.put(CommonParams.FQ, "word:卢浮宫");SolrParams params = new MapSolrParams(map);List<Map<String, Object>> results = myServer.query(params, new String[] {"_id", "title", "url"});assertEquals(10, results.size());}
}

在实际开发的过程中，使用Solrj客户端可以非常容易为测试做一些基本操作，如创建索引，测试Solr基本参数及其开发定制Solr相关接口（Retrieval、Highlighting、Faceted Search、Clustering等等）。

这篇关于使用Solrj管理Solr索引的文章就介绍到这儿，希望我们推荐的文章对编程师们有所帮助！