wenchanter 2014-11-13
使用SolrJ操作Solr会比利用httpClient来操作Solr要简单。SolrJ是封装了httpClient方法,来操作solr的API的。SolrJ底层还是通过使用httpClient中的方法来完成Solr的操作。
1、首先,你需要添加如下jar包
其中apache-solr-solrj-3.4.0.jar、slf4j-api-1.6.1.jar可以在下载的apache-solr-3.4.0的压缩包中的dist中能找到。
2、其次,建立一个简单的测试类,完成Server对象的相关方法的测试工作,代码如下:
package com.hoo.test; import java.io.IOException; import java.net.MalformedURLException; import java.util.ArrayList; import java.util.Collection; import java.util.List; import org.apache.solr.client.solrj.SolrQuery; import org.apache.solr.client.solrj.SolrServer; import org.apache.solr.client.solrj.SolrServerException; import org.apache.solr.client.solrj.impl.CommonsHttpSolrServer; import org.apache.solr.client.solrj.response.QueryResponse; import org.apache.solr.client.solrj.response.UpdateResponse; import org.apache.solr.common.SolrDocumentList; import org.apache.solr.common.SolrInputDocument; import org.apache.solr.common.params.ModifiableSolrParams; import org.apache.solr.common.params.SolrParams; import org.junit.After; import org.junit.Before; import org.junit.Test; import com.hoo.entity.Index; /** * <b>function:</b> Server TestCase * @author hoojo * @createDate 2011-10-19 下午01:49:07 * @file ServerTest.java * @package com.hoo.test * @project SolrExample * @blog http://blog.csdn.net/IBM_hoojo * @email [email protected] * @version 1.0 */ public class ServerTest { private SolrServer server; private CommonsHttpSolrServer httpServer; private static final String DEFAULT_URL = "http://localhost:8983/solr/"; @Before public void init() { try { server = new CommonsHttpSolrServer(DEFAULT_URL); httpServer = new CommonsHttpSolrServer(DEFAULT_URL); } catch (MalformedURLException e) { e.printStackTrace(); } } @After public void destory() { server = null; httpServer = null; System.runFinalization(); System.gc(); } public final void fail(Object o) { System.out.println(o); } /** * <b>function:</b> 测试是否创建server对象成功 * @author hoojo * @createDate 2011-10-21 上午09:48:18 */ @Test public void server() { fail(server); fail(httpServer); } /** * <b>function:</b> 根据query参数查询索引 * @author hoojo * @createDate 2011-10-21 上午10:06:39 * @param query */ public void query(String query) { SolrParams params = new SolrQuery(query); try { QueryResponse response = server.query(params); SolrDocumentList list = response.getResults(); for (int i = 0; i < list.size(); i++) { fail(list.get(i)); } } catch (SolrServerException e) { e.printStackTrace(); } } }
测试运行servercase方法,如果成功创建对象,那你就成功的链接到。
注意:在运行本方法之前,请启动你的solr官方自动的项目。http://localhost:8983/solr/保证能够成功访问这个工程。因为接下来的所有工作都是围绕这个solr工程完成的。如果你现在还不知道,怎么部署、发布官方solr工程,请参考前面的具体章节。
3、Server的有关配置选项参数,server是CommonsHttpSolrServer的实例
server.setSoTimeout(1000); // socket read timeout server.setConnectionTimeout(100); server.setDefaultMaxConnectionsPerHost(100); server.setMaxTotalConnections(100); server.setFollowRedirects(false); // defaults to false // allowCompression defaults to false. // Server side must support gzip or deflate for this to have any effect. server.setAllowCompression(true); server.setMaxRetries(1); // defaults to 0. > 1 not recommended. //sorlr J 目前使用二进制的格式作为默认的格式。对于solr1.2的用户通过显示的设置才能使用XML格式。 server.setParser(new XMLResponseParser()); //二进制流输出格式 //server.setRequestWriter(new BinaryRequestWriter());
4、利用SolrJ完成IndexDocument的添加操作
/**
* <b>function:</b> 添加doc文档
* @author hoojo
* @createDate 2011-10-21 上午09:49:10
*/
@Test
public void addDoc() {
//创建doc文档
SolrInputDocument doc = new SolrInputDocument();
doc.addField("id", 1);
doc.addField("name", "Solr Input Document");
doc.addField("manu", "this is SolrInputDocument content");
try {
//添加一个doc文档
UpdateResponse response = server.add(doc);
fail(server.commit());//commit后才保存到索引库
fail(response);
fail("query time:" + response.getQTime());
fail("Elapsed Time:" + response.getElapsedTime());
fail("status:" + response.getStatus());
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
query("name:solr");
}在apache-solr-3.4.0\example\solr\conf目录下的schema.xml中可以找到有关于field属性的配置,schema.xml中的field就和上面Document文档中的field(id、name、manu)对应。如果出现ERROR:unknownfield'xxxx'就表示你设置的这个field在schema.xml中不存在。如果一定要使用这个field,请你在schema.xml中进行filed元素的配置。具体请参考前面的章节。
注意:在schema.xml中配置了uniqueKey为id,就表示id是唯一的。如果在添加Document的时候,id重复添加。那么后面添加的相同id的doc会覆盖前面的doc,类似于update更新操作,而不会出现重复的数据。
5、利用SolrJ添加多个Document,即添加文档集合
/**
* <b>function:</b> 添加docs文档集合
* @author hoojo
* @createDate 2011-10-21 上午09:55:01
*/
@Test
public void addDocs() {
Collection<SolrInputDocument> docs = new ArrayList<SolrInputDocument>();
SolrInputDocument doc = new SolrInputDocument();
doc.addField("id", 2);
doc.addField("name", "Solr Input Documents 1");
doc.addField("manu", "this is SolrInputDocuments 1 content");
docs.add(doc);
doc = new SolrInputDocument();
doc.addField("id", 3);
doc.addField("name", "Solr Input Documents 2");
doc.addField("manu", "this is SolrInputDocuments 3 content");
docs.add(doc);
try {
//add docs
UpdateResponse response = server.add(docs);
//commit后才保存到索引库
fail(server.commit());
fail(response);
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
query("solr");
}就是添加一个List集合
6、添加JavaEntityBean,这个需要先创建一个JavaBean,然后来完成添加操作;
JavaBean:Index的代码
package com.hoo.entity; import org.apache.solr.client.solrj.beans.Field; /** * <b>function:</b> JavaEntity Bean;Index需要添加相关的Annotation注解,便于告诉solr哪些属性参与到index中 * @author hoojo * @createDate 2011-10-19 下午05:33:27 * @file Index.java * @package com.hoo.entity * @project SolrExample * @blog http://blog.csdn.net/IBM_hoojo * @email [email protected] * @version 1.0 */ public class Index { //@Field setter方法上添加Annotation也是可以的 private String id; @Field private String name; @Field private String manu; @Field private String[] cat; @Field private String[] features; @Field private float price; @Field private int popularity; @Field private boolean inStock; public String getId() { return id; } @Field public void setId(String id) { this.id = id; } //getter、setter方法 public String toString() { return this.id + "#" + this.name + "#" + this.manu + "#" + this.cat; } }
注意上面的属性是和在apache-solr-3.4.0\example\solr\conf目录下的schema.xml中可以找到有关于field属性的配置对应的。如果你IndexJavaBean中出现的属性在schema.xml的field配置无法找到,那么出出现unknownfiled错误。
添加Bean完成doc添加操作
/**
* <b>function:</b> 添加JavaEntity Bean
* @author hoojo
* @createDate 2011-10-21 上午09:55:37
*/
@Test
public void addBean() {
//Index需要添加相关的Annotation注解,便于告诉solr哪些属性参与到index中
Index index = new Index();
index.setId("4");
index.setName("add bean index");
index.setManu("index bean manu");
index.setCat(new String[] { "a1", "b2" });
try {
//添加Index Bean到索引库
UpdateResponse response = server.addBean(index);
fail(server.commit());//commit后才保存到索引库
fail(response);
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
queryAll();
}7、添加Bean集合
/**
* <b>function:</b> 添加Entity Bean集合到索引库
* @author hoojo
* @createDate 2011-10-21 上午10:00:55
*/
@Test
public void addBeans() {
Index index = new Index();
index.setId("6");
index.setName("add beans index 1");
index.setManu("index beans manu 1");
index.setCat(new String[] { "a", "b" });
List<Index> indexs = new ArrayList<Index>();
indexs.add(index);
index = new Index();
index.setId("5");
index.setName("add beans index 2");
index.setManu("index beans manu 2");
index.setCat(new String[] { "aaa", "bbbb" });
indexs.add(index);
try {
//添加索引库
UpdateResponse response = server.addBeans(indexs);
fail(server.commit());//commit后才保存到索引库
fail(response);
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
queryAll();
}8、删除索引Document
/**
* <b>function:</b> 删除索引操作
* @author hoojo
* @createDate 2011-10-21 上午10:04:28
*/
@Test
public void remove() {
try {
//删除id为1的索引
server.deleteById("1");
server.commit();
query("id:1");
//根据id集合,删除多个索引
List<String> ids = new ArrayList<String>();
ids.add("2");
ids.add("3");
server.deleteById(ids);
server.commit(true, true);
query("id:3 id:2");
//删除查询到的索引信息
server.deleteByQuery("id:4 id:6");
server.commit(true, true);
queryAll();
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
}9、查询索引
/**
* <b>function:</b> 查询所有索引信息
* @author hoojo
* @createDate 2011-10-21 上午10:05:38
*/
@Test
public void queryAll() {
ModifiableSolrParams params = new ModifiableSolrParams();
// 查询关键词,*:*代表所有属性、所有值,即所有index
params.set("q", "*:*");
// 分页,start=0就是从0开始,,rows=5当前返回5条记录,第二页就是变化start这个值为5就可以了。
params.set("start", 0);
params.set("rows", Integer.MAX_VALUE);
// 排序,,如果按照id 排序,,那么将score desc 改成 id desc(or asc)
params.set("sort", "score desc");
// 返回信息 * 为全部 这里是全部加上score,如果不加下面就不能使用score
params.set("fl", "*,score");
try {
QueryResponse response = server.query(params);
SolrDocumentList list = response.getResults();
for (int i = 0; i < list.size(); i++) {
fail(list.get(i));
}
} catch (SolrServerException e) {
e.printStackTrace();
}
}10、其他和Server有关方法
/**
* <b>function:</b> 其他server相关方法测试
* @author hoojo
* @createDate 2011-10-21 上午10:02:03
*/
@Test
public void otherMethod() {
fail(server.getBinder());
try {
fail(server.optimize());//合并索引文件,可以优化索引、提供性能,但需要一定的时间
fail(server.ping());//ping服务器是否连接成功
Index index = new Index();
index.setId("299");
index.setName("add bean index199");
index.setManu("index bean manu199");
index.setCat(new String[] { "a199", "b199" });
UpdateResponse response = server.addBean(index);
fail("response: " + response);
queryAll();
//回滚掉之前的操作,rollback addBean operation
fail("rollback: " + server.rollback());
//提交操作,提交后无法回滚之前操作;发现addBean没有成功添加索引
fail("commit: " + server.commit());
queryAll();
} catch (SolrServerException e) {
e.printStackTrace();
} catch (IOException e) {
e.printStackTrace();
}
}11、文档查询
/**
* <b>function:</b> query 基本用法测试
* @author hoojo
* @createDate 2011-10-20 下午04:44:28
*/
@Test
public void queryCase() {
//AND 并且
SolrQuery params = new SolrQuery("name:apple AND manu:inc");
//OR 或者
params.setQuery("name:apple OR manu:apache");
//空格 等同于 OR
params.setQuery("name:server manu:dell");
//params.setQuery("name:solr - manu:inc");
//params.setQuery("name:server + manu:dell");
//查询name包含solr apple
params.setQuery("name:solr,apple");
//manu不包含inc
params.setQuery("name:solr,apple NOT manu:inc");
//50 <= price <= 200
params.setQuery("price:[50 TO 200]");
params.setQuery("popularity:[5 TO 6]");
//params.setQuery("price:[50 TO 200] - popularity:[5 TO 6]");
//params.setQuery("price:[50 TO 200] + popularity:[5 TO 6]");
//50 <= price <= 200 AND 5 <= popularity <= 6
params.setQuery("price:[50 TO 200] AND popularity:[5 TO 6]");
params.setQuery("price:[50 TO 200] OR popularity:[5 TO 6]");
//过滤器查询,可以提高性能 filter 类似多个条件组合,如and
//params.addFilterQuery("id:VA902B");
//params.addFilterQuery("price:[50 TO 200]");
//params.addFilterQuery("popularity:[* TO 5]");
//params.addFilterQuery("weight:*");
//0 < popularity < 6 没有等于
//params.addFilterQuery("popularity:{0 TO 6}");
//排序
params.addSortField("id", ORDER.asc);
//分页:start开始页,rows每页显示记录条数
//params.add("start", "0");
//params.add("rows", "200");
//params.setStart(0);
//params.setRows(200);
//设置高亮
params.setHighlight(true); // 开启高亮组件
params.addHighlightField("name");// 高亮字段
params.setHighlightSimplePre("<font color='red'>");//标记,高亮关键字前缀
params.setHighlightSimplePost("</font>");//后缀
params.setHighlightSnippets(1);//结果分片数,默认为1
params.setHighlightFragsize(1000);//每个分片的最大长度,默认为100
//分片信息
params.setFacet(true)
.setFacetMinCount(1)
.setFacetLimit(5)//段
.addFacetField("name")//分片字段
.addFacetField("inStock");
//params.setQueryType("");
try {
QueryResponse response = server.query(params);
/*List<Index> indexs = response.getBeans(Index.class);
for (int i = 0; i < indexs.size(); i++) {
fail(indexs.get(i));
}*/
//输出查询结果集
SolrDocumentList list = response.getResults();
fail("query result nums: " + list.getNumFound());
for (int i = 0; i < list.size(); i++) {
fail(list.get(i));
}
//输出分片信息
List<FacetField> facets = response.getFacetFields();
for (FacetField facet : facets) {
fail(facet);
List<Count> facetCounts = facet.getValues();
for (FacetField.Count count : facetCounts) {
System.out.println(count.getName() + ": " + count.getCount());
}
}
} catch (SolrServerException e) {
e.printStackTrace();
}
}12、分片查询、统计
/**
* <b>function:</b> 分片查询, 可以统计关键字及出现的次数、或是做自动补全提示
* @author hoojo
* @createDate 2011-10-20 下午04:54:25
*/
@Test
public void facetQueryCase() {
SolrQuery params = new SolrQuery("*:*");
//排序
params.addSortField("id", ORDER.asc);
params.setStart(0);
params.setRows(200);
//Facet为solr中的层次分类查询
//分片信息
params.setFacet(true)
.setQuery("*:*")
.setFacetMinCount(1)
.setFacetLimit(5)//段
//.setFacetPrefix("electronics", "cat")
.setFacetPrefix("cor")//查询manu、name中关键字前缀是cor的
.addFacetField("manu")
.addFacetField("name");//分片字段
try {
QueryResponse response = server.query(params);
//输出查询结果集
SolrDocumentList list = response.getResults();
fail("Query result nums: " + list.getNumFound());
for (int i = 0; i < list.size(); i++) {
fail(list.get(i));
}
fail("All facet filed result: ");
//输出分片信息
List<FacetField> facets = response.getFacetFields();
for (FacetField facet : facets) {
fail(facet);
List<Count> facetCounts = facet.getValues();
for (FacetField.Count count : facetCounts) {
//关键字 - 出现次数
fail(count.getName() + ": " + count.getCount());
}
}
fail("Search facet [name] filed result: ");
//输出分片信息
FacetField facetField = response.getFacetField("name");
List<Count> facetFields = facetField.getValues();
for (Count count : facetFields) {
//关键字 - 出现次数
fail(count.getName() + ": " + count.getCount());
}
} catch (SolrServerException e) {
e.printStackTrace();
}
}分片查询在某些统计关键字的时候还是很有用的,可以统计关键字出现的次数,可以通过统计的关键字来搜索相关文档的信息。
13、Document文档和JavaBean相互转换
这里转换的Bean是一个简单的User对象
package com.hoo.entity; import java.io.Serializable; import org.apache.solr.client.solrj.beans.Field; /** * <b>function:</b> User Entity Bean;所有被添加Annotation @Field 注解的属性将参与index操作 * @author hoojo * @createDate 2011-10-19 下午04:16:00 * @file User.java * @package com.hoo.entity * @project SolrExample * @blog http://blog.csdn.net/IBM_hoojo * @email [email protected] * @version 1.0 */ public class User implements Serializable { /** * @author Hoojo */ private static final long serialVersionUID = 8606788203814942679L; //@Field private int id; @Field private String name; @Field private int age; /** * 可以给某个属性重命名,likes就是solr index的属性;在solrIndex中将显示like为likes */ @Field("likes") private String[] like; @Field private String address; @Field private String sex; @Field private String remark; public int getId() { return id; } //setter 方法上面也可以 @Field public void setId(int id) { this.id = id; } public String getName() { return name; } //getter、setter @Override public String toString() { return this.id + "#" + this.name + "#" + this.age + "#" + this.like + "#" + this.address + "#" + this.sex + "#" + this.remark; } }
测试类代码如下
package com.hoo.test; import org.apache.solr.client.solrj.beans.DocumentObjectBinder; import org.apache.solr.common.SolrDocument; import org.apache.solr.common.SolrDocumentList; import org.apache.solr.common.SolrInputDocument; import org.apache.solr.common.SolrInputField; import org.junit.Test; import com.hoo.entity.User; /** * <b>function:</b>SolrInputDocument implements Map, Iterable * @author hoojo * @createDate 2011-10-19 下午03:54:54 * @file SolrInputDocumentTest.java * @package com.hoo.test * @project SolrExample * @blog http://blog.csdn.net/IBM_hoojo * @email [email protected] * @version 1.0 */ public class SolrInputDocumentTest { public final void fail(Object o) { System.out.println(o); } /** * <b>function:</b> 创建SolrInputDocument * @author hoojo * @createDate 2011-10-21 下午03:38:20 */ @Test public void createDoc() { SolrInputDocument doc = new SolrInputDocument(); doc.addField("id", System.currentTimeMillis()); doc.addField("name", "SolrInputDocument"); doc.addField("age", 22, 2.0f); doc.addField("like", new String[] { "music", "book", "sport" }); doc.put("address", new SolrInputField("guangzhou")); doc.setField("sex", "man"); doc.setField("remark", "china people", 2.0f); fail(doc); } /** * <b>function:</b> 利用DocumentObjectBinder对象将SolrInputDocument 和 User对象相互转换 * @author hoojo * @createDate 2011-10-21 下午03:38:40 */ @Test public void docAndBean4Binder() { SolrDocument doc = new SolrDocument(); doc.addField("id", 456); doc.addField("name", "SolrInputDocument"); doc.addField("likes", new String[] { "music", "book", "sport" }); doc.put("address", "guangzhou"); doc.setField("sex", "man"); doc.setField("remark", "china people"); DocumentObjectBinder binder = new DocumentObjectBinder(); User user = new User(); user.setId(222); user.setName("JavaBean"); user.setLike(new String[] { "music", "book", "sport" }); user.setAddress("guangdong"); fail(doc); // User ->> SolrInputDocument fail(binder.toSolrInputDocument(user)); // SolrDocument ->> User fail(binder.getBean(User.class, doc)); SolrDocumentList list = new SolrDocumentList(); list.add(doc); list.add(doc); //SolrDocumentList ->> List fail(binder.getBeans(User.class, list)); } /** * <b>function:</b> SolrInputDocument的相关方法 * @author hoojo * @createDate 2011-10-21 下午03:44:30 */ @Test public void docMethod() { SolrInputDocument doc = new SolrInputDocument(); doc.addField("id", System.currentTimeMillis()); doc.addField("name", "SolrInputDocument"); doc.addField("age", 23, 1.0f); doc.addField("age", 22, 2.0f); doc.addField("age", 24, 0f); fail(doc.entrySet()); fail(doc.get("age")); //排名有用,类似百度竞价排名 doc.setDocumentBoost(2.0f); fail(doc.getDocumentBoost()); fail(doc.getField("name")); fail(doc.getFieldNames());//keys fail(doc.getFieldValues("age")); fail(doc.getFieldValues("id")); fail(doc.values()); } }