solr

 

转自：http://blog.sina.com.cn/s/blog_466678e80100l2an.html

 

 

Solrj已经是很强大的solr客户端了。它本身就包装了httpCliet，以完全对象的方式对solr进行交互。很小很好很强大。
不过在实际使用中，设置SolrQuery 的过程中，为了设置多个搜索条件和排序规则等等参数，我们往往会陷入并接字符串的地步，实在是很丑陋，不符合面向对象的思想。扩展性几乎为0,。基于这点，开发了一个小东西，我们只需要设置搜索对象，将对象扔给后台就可以了。
比如，我们搭建的solr服务支持某10个字段的搜索，我们要搜索其中的一些，那么我们只需要传入要搜索的对象POJO，将要搜索的字段内容，set到POJO对象对应额字段即可。
比如如下一个类:
Java代码
package org.uppower.tnt.biz.core.manager.blog.dataobject;


public class SolrPropertyDO {
private String auction_id;
private String opt_tag;
private String exp_tag;
private String title;
private String desc;
private String brand;
private String category;
private String price;
private String add_prov;
private String add_city;
private String quality;
private String flag;
private String sales;
private String sellerrate;
private String selleruid;
private String ipv15;

public String getAuction_id() {
return auction_id;
}

public void setAuction_id(String auctionId) {
auction_id = auctionId;
}

……

public String getExp_tag() {
return exp_tag;
}

public void setExp_tag(String expTag) {
exp_tag = expTag;
}
}
package org.uppower.tnt.biz.core.manager.blog.dataobject;

public class SolrPropertyDO {
private String auction_id;
private String opt_tag;
private String exp_tag;
private String title;
private String desc;
private String brand;
private String category;
private String price;
private String add_prov;
private String add_city;
private String quality;
private String flag;
private String sales;
private String sellerrate;
private String selleruid;
private String ipv15;
public String getAuction_id() {
return auction_id;
}
public void setAuction_id(String auctionId) {
auction_id = auctionId;
}
……
public String getExp_tag() {
return exp_tag;
}
public void setExp_tag(String expTag) {
exp_tag = expTag;
}
}
那么我们在定义搜索对象时候，就按照如下设置：
Java代码
1. SolrPropertyDO propertyDO = new SolrPropertyDO();
2. propertyDO.setAdd_city("(杭州AND成都)OR北京");
3. propertyDO.setTitle("丝绸OR剪刀");
4. ……
SolrPropertyDO propertyDO = new SolrPropertyDO();
propertyDO.setAdd_city("(杭州AND成都)OR北京");
propertyDO.setTitle("丝绸OR剪刀");
……
设置排序条件，也是类似的做法：
Java代码
1. SolrPropertyDO compositorDO = new SolrPropertyDO();
2. compositorDO.setPrice ("desc");
3. compositorDO.setQuality ("asc");
4. ……
SolrPropertyDO compositorDO = new SolrPropertyDO();
compositorDO.setPrice ("desc");
compositorDO.setQuality ("asc");
……
将定义好的两个对象扔给后面的接口就可以了。
接口函数querySolrResult传入四个参数，其中包含搜索字段对象，排序条件对象。为了提供类似limit的操作，用于分页查询，提供了startIndex和pageSize。
函数querySolrResultCount是单纯为了获得搜索条数，配合分页使用。
以下是定义的接口：
Java代码
package org.uppower.tnt.biz.core.manager.blog;

import java.util.List;

import org.uppower.tnt.biz.core.manager.isearch.dataobject.SolrPropertyDO;


public interface SolrjOperator {


public List<Object> querySolrResult(Object propertyDO,
Object compositorDO, Long startIndex, Long pageSize)
throws Exception;


public Long querySolrResultCount(SolrPropertyDO propertyDO,
Object compositorDO) throws Exception;

}
package org.uppower.tnt.biz.core.manager.blog;
import java.util.List;
import org.uppower.tnt.biz.core.manager.isearch.dataobject.SolrPropertyDO;

public interface SolrjOperator {

public List<Object> querySolrResult(Object propertyDO,
Object compositorDO, Long startIndex, Long pageSize)
throws Exception;

public Long querySolrResultCount(SolrPropertyDO propertyDO,
Object compositorDO) throws Exception;
}

实现逻辑为，首先将传入的两个实体对象，解析为<K,V>结构的Map当中，将解析完成的Map放入solrj实际的搜索对象当中。返回的对象为solrj的API提供的SolrDocument,其中结果数量为直接返回SolrDocumentList对象的getNumFound()
具体实现类：
Java代码
package org.uppower.tnt.biz.core.manager.blog;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.TreeMap;

import org.apache.solr.common.SolrDocumentList;

import org.uppower.tnt.biz.core.manager.isearch.common.SolrjCommonUtil;
import org.uppower.tnt.biz.core.manager.isearch.dataobject.SolrPropertyDO;
import org.uppower.tnt.biz.core.manager.isearch.solrj.SolrjQuery;


public class DefaultSolrOperator implements SolrjOperator {

private Logger logger = LoggerFactory.getLogger(this.getClass());
private SolrjQuery solrjQuery;

public void setSolrjQuery(SolrjQuery solrjQuery) {
this.solrjQuery = solrjQuery;
}

@Override
public List<Object> querySolrResult(Object propertyDO,
Object compositorDO, Long startIndex, Long pageSize)
throws Exception {
Map<String, String> propertyMap = new TreeMap<String, String>();
//排序有顺序,使用TreeMap
Map<String, String> compositorMap = new TreeMap<String, String>();
try {
propertyMap = SolrjCommonUtil.getSearchProperty(propertyDO);
compositorMap = SolrjCommonUtil.getSearchProperty(compositorDO);
} catch (Exception e) {
logger.error("SolrjCommonUtil.getSearchProperty() is error !"+ e);
}
SolrDocumentList solrDocumentList = solrjQuery.query(propertyMap, compositorMap,
startIndex, pageSize);
List<Object> resultList = new ArrayList<Object>();
for (int i = 0; i < solrDocumentList.size(); i++) {
resultList.add(solrDocumentList.get(i));
}
return resultList;
}

@Override
public Long querySolrResultCount(SolrPropertyDO propertyDO,
Object compositorDO) throws Exception {
Map<String, String> propertyMap = new TreeMap<String, String>();
Map<String, String> compositorMap = new TreeMap<String, String>();
try {
propertyMap = SolrjCommonUtil.getSearchProperty(propertyDO);
compositorMap = SolrjCommonUtil.getSearchProperty(compositorDO);
} catch (Exception e) {
logger.error("SolrjCommonUtil.getSearchProperty() is error !" + e);
}
SolrDocumentList solrDocument = solrjQuery.query(propertyMap, compositorMap,
null, null);
return solrDocument.getNumFound();
}

}
package org.uppower.tnt.biz.core.manager.blog;
import java.util.ArrayList;
import java.util.HashMap;
import java.util.List;
import java.util.Map;
import java.util.TreeMap;
import org.apache.solr.common.SolrDocumentList;
import org.uppower.tnt.biz.core.manager.isearch.common.SolrjCommonUtil;
import org.uppower.tnt.biz.core.manager.isearch.dataobject.SolrPropertyDO;
import org.uppower.tnt.biz.core.manager.isearch.solrj.SolrjQuery;

public class DefaultSolrOperator implements SolrjOperator {

private Logger logger = LoggerFactory.getLogger(this.getClass());
private SolrjQuery solrjQuery;
public void setSolrjQuery(SolrjQuery solrjQuery) {
this.solrjQuery = solrjQuery;
}
@Override
public List<Object> querySolrResult(Object propertyDO,
Object compositorDO, Long startIndex, Long pageSize)
throws Exception {
Map<String, String> propertyMap = new TreeMap<String, String>();
//排序有顺序,使用TreeMap
Map<String, String> compositorMap = new TreeMap<String, String>();
try {
propertyMap = SolrjCommonUtil.getSearchProperty(propertyDO);
compositorMap = SolrjCommonUtil.getSearchProperty(compositorDO);
} catch (Exception e) {
logger.error("SolrjCommonUtil.getSearchProperty() is error !"+ e);
}
SolrDocumentList solrDocumentList = solrjQuery.query(propertyMap, compositorMap,
startIndex, pageSize);
List<Object> resultList = new ArrayList<Object>();
for (int i = 0; i < solrDocumentList.size(); i++) {
resultList.add(solrDocumentList.get(i));
}
return resultList;
}
@Override
public Long querySolrResultCount(SolrPropertyDO propertyDO,
Object compositorDO) throws Exception {
Map<String, String> propertyMap = new TreeMap<String, String>();
Map<String, String> compositorMap = new TreeMap<String, String>();
try {
propertyMap = SolrjCommonUtil.getSearchProperty(propertyDO);
compositorMap = SolrjCommonUtil.getSearchProperty(compositorDO);
} catch (Exception e) {
logger.error("SolrjCommonUtil.getSearchProperty() is error !" + e);
}
SolrDocumentList solrDocument = solrjQuery.query(propertyMap, compositorMap,
null, null);
return solrDocument.getNumFound();
}
}

其中，对象的解析式利用反射原理，将实体对象中不为空的值，以映射的方式，转化为一个Map，其中排序对象在转化的过程中，使用TreeMap，保证其顺序性。
解析公共类实现如下：
Java代码
package org.uppower.tnt.biz.core.manager.blog.common;

import java.lang.reflect.Field;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method;
import java.util.HashMap;
import java.util.Map;


public class SolrjCommonUtil {

public static Map<String, String> getSearchProperty(Object model)
throws NoSuchMethodException, IllegalAccessException,
IllegalArgumentException, InvocationTargetException {
Map<String, String> resultMap = new TreeMap<String, String>();
// 获取实体类的所有属性，返回Field数组
Field[] field = model.getClass().getDeclaredFields();
for (int i = 0; i < field.length; i++) { // 遍历所有属性
String name = field[i].getName(); // 获取属性的名字
// 获取属性的类型
String type = field[i].getGenericType().toString();
if (type.equals("class java.lang.String")) { // 如果type是类类型，则前面包含"class "，后面跟类名
Method m = model.getClass().getMethod(
"get" + UpperCaseField(name));
String value = (String) m.invoke(model); // 调用getter方法获取属性值
if (value != null) {
resultMap.put(name, value);
}
}
}
return resultMap;
}

// 转化字段首字母为大写
private static String UpperCaseField(String fieldName) {
fieldName = fieldName.replaceFirst(fieldName.substring(0, 1), fieldName
.substring(0, 1).toUpperCase());
return fieldName;
}

}
package org.uppower.tnt.biz.core.manager.blog.common;
import java.lang.reflect.Field;
import java.lang.reflect.InvocationTargetException;
import java.lang.reflect.Method;
import java.util.HashMap;
import java.util.Map;

public class SolrjCommonUtil {
public static Map<String, String> getSearchProperty(Object model)
throws NoSuchMethodException, IllegalAccessException,
IllegalArgumentException, InvocationTargetException {
Map<String, String> resultMap = new TreeMap<String, String>();
// 获取实体类的所有属性，返回Field数组
Field[] field = model.getClass().getDeclaredFields();
for (int i = 0; i < field.length; i++) { // 遍历所有属性
String name = field[i].getName(); // 获取属性的名字
// 获取属性的类型
String type = field[i].getGenericType().toString();
if (type.equals("class java.lang.String")) { // 如果type是类类型，则前面包含"class "，后面跟类名
Method m = model.getClass().getMethod(
"get" + UpperCaseField(name));
String value = (String) m.invoke(model); // 调用getter方法获取属性值
if (value != null) {
resultMap.put(name, value);
}
}
}
return resultMap;
}
// 转化字段首字母为大写
private static String UpperCaseField(String fieldName) {
fieldName = fieldName.replaceFirst(fieldName.substring(0, 1), fieldName
.substring(0, 1).toUpperCase());
return fieldName;
}
}

搜索直接调用solr客户端solrj，基本逻辑为循环两个解析之后的TreeMap，设置到SolrQuery当中，最后直接调用solrj的API，获得搜索结果。最终将搜索结构以List<Object>的形式返回。
具体实现：
Java代码
package org.uppower.tnt.biz.core.manager.blog.solrj;

import java.net.MalformedURLException;
import java.util.Map;

import org.apache.solr.client.solrj.SolrQuery;
import org.apache.solr.client.solrj.impl.CommonsHttpSolrServer;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.apache.solr.common.SolrDocumentList;


public class SolrjQuery {
private String url;
private Integer soTimeOut;
private Integer connectionTimeOut;
private Integer maxConnectionsPerHost;
private Integer maxTotalConnections;
private Integer maxRetries;
private CommonsHttpSolrServer solrServer = null;
private final static String ASC = "asc";

public void init() throws MalformedURLException {
solrServer = new CommonsHttpSolrServer(url);
solrServer.setSoTimeout(soTimeOut);
solrServer.setConnectionTimeout(connectionTimeOut);
solrServer.setDefaultMaxConnectionsPerHost(maxConnectionsPerHost);
solrServer.setMaxTotalConnections(maxTotalConnections);
solrServer.setFollowRedirects(false);
solrServer.setAllowCompression(true);
solrServer.setMaxRetries(maxRetries);
}

public SolrDocumentList query(Map<String, String> propertyMap,
Map<String, String> compositorMap, Long startIndex, Long pageSize)
throws Exception {
SolrQuery query = new SolrQuery();
// 设置搜索字段
if (null == propertyMap) {
throw new Exception("搜索字段不可为空!");
} else {
for (Object o : propertyMap.keySet()) {
StringBuffer sb = new StringBuffer();
sb.append(o.toString()).append(":");
sb.append(propertyMap.get(o));
String queryString = addBlank2expression_r(sb.toString());
query.setQuery(queryString);
}
}
// 设置排序条件
if (null != compositorMap) {
for (Object co : compositorMap.keySet()) {
if (ASC == compositorMap.get(co)
|| ASC.equals(compositorMap.get(co))) {
query.addSortField(co.toString(), SolrQuery.ORDER.asc);
} else {
query.addSortField(co.toString(), SolrQuery.ORDER.desc);
}
}
}

if (null != startIndex) {
query.setStart(Integer.parseInt(String.valueOf(startIndex)));
}
if (null != pageSize && 0L != pageSize.longValue()) {
query.setRows(Integer.parseInt(String.valueOf(pageSize)));
}
try {
QueryResponse qrsp = solrServer.query(query);
SolrDocumentList docs = qrsp.getResults();
return docs;
} catch (Exception e) {
throw new Exception(e);
}
}

private String addBlank2expression_r(String oldExpression) {
String lastExpression;
lastExpression = oldExpression.replace("AND", " AND ").replace("NOT",
" NOT ").replace("OR", " OR ");
return lastExpression;
}

public Integer getMaxRetries() {
return maxRetries;
}

……

public void setMaxTotalConnections(Integer maxTotalConnections) {
this.maxTotalConnections = maxTotalConnections;
}
}
package org.uppower.tnt.biz.core.manager.blog.solrj;
import java.net.MalformedURLException;
import java.util.Map;
import org.apache.solr.client.solrj.SolrQuery;
import org.apache.solr.client.solrj.impl.CommonsHttpSolrServer;
import org.apache.solr.client.solrj.response.QueryResponse;
import org.apache.solr.common.SolrDocumentList;

public class SolrjQuery {
private String url;
private Integer soTimeOut;
private Integer connectionTimeOut;
private Integer maxConnectionsPerHost;
private Integer maxTotalConnections;
private Integer maxRetries;
private CommonsHttpSolrServer solrServer = null;
private final static String ASC = "asc";
public void init() throws MalformedURLException {
solrServer = new CommonsHttpSolrServer(url);
solrServer.setSoTimeout(soTimeOut);
solrServer.setConnectionTimeout(connectionTimeOut);
solrServer.setDefaultMaxConnectionsPerHost(maxConnectionsPerHost);
solrServer.setMaxTotalConnections(maxTotalConnections);
solrServer.setFollowRedirects(false);
solrServer.setAllowCompression(true);
solrServer.setMaxRetries(maxRetries);
}
public SolrDocumentList query(Map<String, String> propertyMap,
Map<String, String> compositorMap, Long startIndex, Long pageSize)
throws Exception {
SolrQuery query = new SolrQuery();
// 设置搜索字段
if (null == propertyMap) {
throw new Exception("搜索字段不可为空!");
} else {
for (Object o : propertyMap.keySet()) {
StringBuffer sb = new StringBuffer();
sb.append(o.toString()).append(":");
sb.append(propertyMap.get(o));
String queryString = addBlank2expression_r(sb.toString());
query.setQuery(queryString);
}
}
// 设置排序条件
if (null != compositorMap) {
for (Object co : compositorMap.keySet()) {
if (ASC == compositorMap.get(co)
|| ASC.equals(compositorMap.get(co))) {
query.addSortField(co.toString(), SolrQuery.ORDER.asc);
} else {
query.addSortField(co.toString(), SolrQuery.ORDER.desc);
}
}
}

if (null != startIndex) {
query.setStart(Integer.parseInt(String.valueOf(startIndex)));
}
if (null != pageSize && 0L != pageSize.longValue()) {
query.setRows(Integer.parseInt(String.valueOf(pageSize)));
}
try {
QueryResponse qrsp = solrServer.query(query);
SolrDocumentList docs = qrsp.getResults();
return docs;
} catch (Exception e) {
throw new Exception(e);
}
}
private String addBlank2expression_r(String oldExpression) {
String lastExpression;
lastExpression = oldExpression.replace("AND", " AND ").replace("NOT",
" NOT ").replace("OR", " OR ");
return lastExpression;
}
public Integer getMaxRetries() {
return maxRetries;
}
……

public void setMaxTotalConnections(Integer maxTotalConnections) {
this.maxTotalConnections = maxTotalConnections;
}
}

整个实现是在Spring的基础上完成的，其中SolrjQuery的init()方法在Spring容器启动是初始化。Init()方法内的属性，也是直接注入的。上层与下层之间也完全用注入的方式解决。具体配置就不贴不出来了，大家都会。
整个代码很简陋，但是几乎支持了你想要搜索的条件设置，而且不会暴露任何与solr相关的内容给上层调用，使整个搜索几乎以sql语言的思想在设置条件。

 

转自：http://blog.sina.com.cn/s/blog_466678e80100l2an.html