首页 > 编程语言 > 关于注解式的分布式Elasticsearch的封装案例
2021
01-26

关于注解式的分布式Elasticsearch的封装案例

原生的Rest Level Client不好用,构建检索等很多重复操作。

对bboss-elasticsearch进行了部分增强:通过注解配合实体类进行自动构建索引和自动刷入文档,复杂的业务检索需要自己在xml中写Dsl。用法与mybatis-plus如出一辙。

依赖

<dependency>
			<groupId>org.elasticsearch</groupId>
			<artifactId>elasticsearch</artifactId>
		</dependency>
		<dependency>
			<groupId>com.bbossgroups.plugins</groupId>
			<artifactId>bboss-elasticsearch-spring-boot-starter</artifactId>
			<version>5.9.5</version>
			<exclusions>
				<exclusion>
					<artifactId>slf4j-log4j12</artifactId>
					<groupId>org.slf4j</groupId>
				</exclusion>
			</exclusions>
		</dependency>
		<!-- https://mvnrepository.com/artifact/org.projectlombok/lombok -->
		<dependency>
			<groupId>org.projectlombok</groupId>
			<artifactId>lombok</artifactId>
			<version>1.18.6</version>
			<scope>provided</scope>
		</dependency>

配置:

 
import com.rz.config.ElsConfig;
import org.frameworkset.elasticsearch.boot.ElasticSearchBoot;
import org.springframework.beans.factory.annotation.Autowired;
import org.springframework.boot.ApplicationArguments;
import org.springframework.boot.ApplicationRunner;
import org.springframework.core.annotation.Order;
import org.springframework.stereotype.Component; 
import java.util.HashMap;
import java.util.Map;
 
/**
 * 启动时初始化bBoss
 *
 * @author sunziwen
 * @version 1.0
 * @date 2019/12/12 16:54
 **/
@Component
@Order(value = 1)
public class StartElastic implements ApplicationRunner {
 @Autowired
 private ElsConfig config;
 
 @Override
 public void run(ApplicationArguments args) throws Exception {
 Map properties = new HashMap();
 properties.put("elasticsearch.rest.hostNames", config.getElsClusterNodes());
 ElasticSearchBoot.boot(properties);
 }
}

注解和枚举:

package com.rz.szwes.annotations;
 
import java.lang.annotation.*;
 
/**
 * 标识实体对应的索引信息
 *
 * @author sunziwen
 * 2019/12/13 10:14
 * @version 1.0
 **/
@Target({ElementType.TYPE})
@Retention(RetentionPolicy.RUNTIME)
@Documented
public @interface ESDsl {
 /**
 * xml的位置
 */
 String value();
 
 String indexName();
 
 /**
 * elasticsearch7.x版本已经删除该属性
 */
 String indexType() default "";
}
package com.rz.szwes.annotations;
 
 
import java.lang.annotation.*;
 
/**
 * 为字段指定映射类型
 *
 * @author sunziwen
 * 2019/12/14 10:06
 * @version 1.0
 **/
@Target({ElementType.FIELD})
@Retention(RetentionPolicy.RUNTIME)
@Documented
public @interface ESMapping {
 //映射类型
 ESMappingType value();
 
 //加权
 int boost() default 1;
 
 //分词标识analyzed、not_analyzed
 String index() default "analyzed";
 
 //分词器ik_max_word、standard
 String analyzer() default "ik_max_word";
 
 //String作为分组聚合字段的时候需要设置为true
 boolean fildData() default false;
}
package com.rz.szwes.annotations;
 
/**
 * Es映射类型枚举(定义了大部分,有缺失请使用者补全)当前版本基于elasticsearch 6.8
 *
 * @author sunziwen
 * 2019/12/14 10:09
 * @version 1.0
 **/
public enum ESMappingType {
 /**
 * 全文搜索。
 */
 text("text"),
 /**
 * keyword类型适用于索引结构化(排序、过滤、聚合),只能通过精确值搜索到。
 */
 keyword("keyword"),
 
 /
 
 /**
 * -128~127 在满足需求的情况下,尽可能选择范围小的数据类型。
 */
 _byte("byte"),
 /**
 * -32768~32767
 */
 _short("short"),
 /**
 * -2^31~2^31-1
 */
 _integer("integer"),
 /**
 * -2^63~2^63-1
 */
 _long("long"),
 
 /
 /**
 * 64位双精度IEEE 754浮点类型
 */
 _doule("doule"),
 /**
 * 32位单精度IEEE 754浮点类型
 */
 _float("float"),
 /**
 * 16位半精度IEEE 754浮点类型
 */
 half_float("half_float"),
 /**
 * 缩放类型的的浮点数
 */
 scaled_float("scaled_float"),
 
 /
 /**
 * 时间类型
 */
 date("date"),
 
 _boolean("boolean"),
 /**
 * 范围类型
 */
 range("range"),
 /**
 * 嵌套类型
 */
 nested("nested"),
 /**
 * 地理坐标
 */
 geo_point("geo_point"),
 /**
 * 地理地图
 */
 geo_shape("geo_shape"),
 /**
 * 二进制类型
 */
 binary("binary"),
 /**
 * ip 192.168.1.2
 */
 ip("ip");
 
 private String value;
 
 ESMappingType(String value) {
 this.value = value;
 }
 
 public String getValue() {
 return value;
 }
}

工具类:对HashMap进行了增强

package com.rz.szwes.util; 
import java.util.HashMap;
import java.util.function.Supplier;
 
/**
 * 原始HashMap不支持Lambda表达式,特此包装一个
 *
 * @author sunziwen
 * @version 1.0
 * @date 2019/12/13 11:09
 **/
public class LambdaHashMap<K, V> extends HashMap<K, V> { 
 public static <K, V> LambdaHashMap<K, V> builder() {
 return new LambdaHashMap<>();
 }
 
 public LambdaHashMap<K, V> put(K key, Supplier<V> supplier) {
 super.put(key, supplier.get());
 //流式
 return this;
 }
}

核心类两个:

package com.rz.szwes.core; 
import cn.hutool.core.util.ClassUtil;
import com.alibaba.fastjson.JSON;
import com.frameworkset.orm.annotation.ESId;
import com.rz.szwes.annotations.ESDsl;
import com.rz.szwes.annotations.ESMapping;
import com.rz.szwes.util.LambdaHashMap;
import org.springframework.util.StringUtils; 
import java.lang.reflect.Field;
import java.lang.reflect.ParameterizedType;
import java.time.LocalDate;
import java.time.LocalDateTime;
import java.time.LocalTime;
import java.util.ArrayList;
import java.util.Arrays;
import java.util.Collections;
import java.util.Date;
 
/**
 * 抽象类解析泛型
 *
 * @author sunziwen
 * 2019/12/14 16:04
 * @version 1.0
 **/
public abstract class AbstractElasticBase<T> {
 { //初始化解析
 ParameterizedType pt = (ParameterizedType) this.getClass().getGenericSuperclass();
 // 获取第一个类型参数的真实类型
 Class<T> clazz = (Class<T>) pt.getActualTypeArguments()[0];
 parseMapping(clazz);
 }
 
 /**
 * 索引名称
 */
 protected String indexName;
 /**
 * 索引类型
 */
 protected String indexType;
 /**
 * es写dsl的文件路径
 */
 protected String xmlPath;
 /**
 * 索引映射
 */
 protected String mapping;
 
 //将Class解析成映射JSONString
 private void parseMapping(Class<T> clazz) {
 if (clazz.isAnnotationPresent(ESDsl.class)) {
  ESDsl esDsl = clazz.getAnnotation(ESDsl.class);
  this.xmlPath = esDsl.value();
  this.indexName = esDsl.indexName();
  //如果类型为空,则采用索引名作为其类型
  this.indexType = StringUtils.isEmpty(esDsl.indexType()) ? esDsl.indexName() : esDsl.indexType();
 } else {
  throw new RuntimeException(clazz.getName() + "缺失注解[@ESDsl]");
 }
 //构建索引映射
 LambdaHashMap<Object, Object> put = LambdaHashMap.builder()
        .put("mappings", () -> LambdaHashMap.builder()
            .put(indexType, () -> LambdaHashMap.builder()
                .put("properties", () -> {
                 Field[] fields = clazz.getDeclaredFields();
                 LambdaHashMap<Object, Object> builder = LambdaHashMap.builder();
                 for (Field field : fields) {
                 builder.put(field.getName(), () -> toEsjson(field));
                 }
                 return builder;
                })))
 ;
 this.mapping = JSON.toJSONString(put);
 }
 
 private LambdaHashMap<Object, Object> toEsjson(Field field) {
 //基本数据类型
 if (ClassUtil.isSimpleTypeOrArray(field.getType())) {
  //对字符串做大小限制、分词设置
  if (new ArrayList<Class>(Collections.singletonList(String.class)).contains(field.getType())) {
  LambdaHashMap<Object, Object> put = LambdaHashMap.builder()
         .put("type", () -> "text")
         .put("fields", () -> LambdaHashMap.builder()
             .put("keyword", () -> LambdaHashMap.builder()
                 .put("type", () -> "keyword")
                 .put("ignore_above", () -> 256)));
  if (field.isAnnotationPresent(ESMapping.class)) {
   ESMapping esMapping = field.getAnnotation(ESMapping.class);
   //设置聚合分组
   if (esMapping.fildData()) {
   put.put("fildData", () -> true);
   }
   //设置加权
   if (esMapping.boost() != 1) {
   put.put("boost", esMapping::boost);
   }
   //设置是否进行分词
   if (!"analyzed".equals(esMapping.index())) {
   put.put("analyzed", esMapping::analyzer);
   }
   //分词器
   put.put("analyzer", esMapping::analyzer);
  }
  return put;
  }
  //设置默认类型
  return LambdaHashMap.builder().put("type", () -> {
  if (field.isAnnotationPresent(ESMapping.class)) {
   ESMapping esMapping = field.getAnnotation(ESMapping.class);
   return esMapping.value().getValue();
  }
  if (new ArrayList<Class>(Arrays.asList(byte.class, Byte.class, short.class, Short.class, int.class, Integer.class, long.class, Long.class)).contains(field.getType())) {
   return "long";
  } else if (new ArrayList<Class>(Arrays.asList(double.class, Double.class, float.class, Float.class)).contains(field.getType())) {
   return "double";
  } else if (new ArrayList<Class>(Arrays.asList(Date.class, java.sql.Date.class, LocalDate.class, LocalDateTime.class, LocalTime.class)).contains(field.getType())) {
   return "date";
  } else if (new ArrayList<Class>(Arrays.asList(boolean.class, Boolean.class)).contains(field.getType())) {
   return "boolean";
  }
  return "text";
  });
 } else {
  //设置对象类型
  LambdaHashMap<Object, Object> properties = LambdaHashMap.builder()
         .put("properties", () -> {
         Field[] fields = field.getType().getDeclaredFields();
         LambdaHashMap<Object, Object> builder = LambdaHashMap.builder();
         for (Field field01 : fields) {
          builder.put(field01.getName(), toEsjson(field01));
         }
         return builder;
         });
  if (field.isAnnotationPresent(ESMapping.class)) {
  ESMapping esMapping = field.getAnnotation(ESMapping.class);
  properties.put("type", esMapping.value().getValue());
  }
  return properties;
 }
 } 
}
package com.rz.szwes.core; 
import lombok.extern.slf4j.Slf4j;
import org.frameworkset.elasticsearch.boot.BBossESStarter;
import org.frameworkset.elasticsearch.client.ClientInterface;
import org.frameworkset.elasticsearch.client.ClientUtil;
import org.springframework.beans.factory.annotation.Autowired; 
import java.util.*;
 
/**
 * Elastic基础函数
 *
 * @author sunziwen
 * @version 1.0
 * @date 2019/12/13 9:56
 **/
@Slf4j
public class ElasticBaseService<T> extends AbstractElasticBase<T> {
 
 @Autowired
 private BBossESStarter starter;
 
 /**
 * Xml创建索引
 */
 protected String createIndexByXml(String xmlName) {
 ClientInterface restClient = starter.getConfigRestClient(xmlPath);
 boolean existIndice = restClient.existIndice(this.indexName);
 if (existIndice) {
  restClient.dropIndice(indexName);
 }
 return restClient.createIndiceMapping(indexName, xmlName);
 }
 
 /**
 * 自动创建索引
 */
 protected String createIndex() {
 ClientInterface restClient = starter.getRestClient();
 boolean existIndice = restClient.existIndice(this.indexName);
 if (existIndice) {
  restClient.dropIndice(indexName);
 }
 log.debug("创建索引:" + this.mapping);
 return restClient.executeHttp(indexName, this.mapping, ClientUtil.HTTP_PUT);
 }
 
 /**
 * 删除索引
 */
 protected String delIndex() {
 return starter.getRestClient().dropIndice(this.indexName);
 }
 
 /**
 * 添加文档
 *
 * @param t 实体类
 * @param refresh 是否强制刷新
 */
 protected String addDocument(T t, Boolean refresh) {
 return starter.getRestClient().addDocument(indexName, indexType, t, "refresh=" + refresh);
 }
 
 /**
 * 添加文档
 *
 * @param ts 实体类集合
 * @param refresh 是否强制刷新
 */
 protected String addDocuments(List<T> ts, Boolean refresh) {
 return starter.getRestClient().addDocuments(indexName, indexType, ts, "refresh=" + refresh);
 }
 
 /**
 * 分页-添加文档集合
 *
 * @param ts 实体类集合
 * @param refresh 是否强制刷新
 */
 protected void addDocumentsOfPage(List<T> ts, Boolean refresh) {
 this.delIndex();
 this.createIndex();
 int start = 0;
 int rows = 100;
 Integer size;
 do {
  List<T> list = pageDate(start, rows);
  if (list.size() > 0) {
  //批量同步信息
  starter.getRestClient().addDocuments(indexName, indexType, ts, "refresh=" + refresh);
  }
  size = list.size();
  start += size;
 } while (size > 0);
 }
 
 /**
 * 使用分页添加文档必须重写该类
 *
 * @param start 起始
 * @param rows 项数
 * @return
 */
 protected List<T> pageDate(int start, int rows) {
 return null;
 }
 
 /**
 * 删除文档
 *
 * @param id id
 * @param refresh 是否强制刷新
 * @return
 */
 protected String delDocument(String id, Boolean refresh) {
 return starter.getRestClient().deleteDocument(indexName, indexType, id, "refresh=" + refresh);
 }
 
 /**
 * 删除文档
 *
 * @param ids id集合
 * @param refresh 是否强制刷新
 * @return
 */
 protected String delDocuments(String[] ids, Boolean refresh) {
 return starter.getRestClient().deleteDocumentsWithrefreshOption(indexName, indexType, "refresh=" + refresh, ids);
 }
 
 /**
 * id获取文档
 *
 * @param id
 * @return
 */
 protected T getDocument(String id, Class<T> clazz) {
 return starter.getRestClient().getDocument(indexName, indexType, id, clazz);
 }
 
 /**
 * id更新文档
 *
 * @param t 实体
 * @param refresh 是否强制刷新
 * @return
 */
 protected String updateDocument(String id, T t, Boolean refresh) {
 return starter.getRestClient().updateDocument(indexName, indexType, id, t, "refresh=" + refresh);
 }
}

写复杂Dsl的xml:(如何写Dsl请参考bBoss-elasticsearch文档,用法类似mybatis标签)

<properties> 
</properties>

框架集成完毕,以下是使用示例:

定义数据模型:

package com.rz.dto;
 
import com.frameworkset.orm.annotation.ESId;
import com.rz.szwes.annotations.ESDsl;
import com.rz.szwes.annotations.ESMapping;
import com.rz.szwes.annotations.ESMappingType;
import lombok.Data; 
import java.util.List;
 
/**
 * 对应elasticsearch服务器的数据模型
 *
 * @author sunziwen
 * @version 1.0
 * @date 2019/12/16 11:08
 **/
@ESDsl(value = "elasticsearch/zsInfo.xml", indexName = "zsInfo")
@Data
public class ElasticZsInfoDto {
 @ESMapping(ESMappingType._byte)
 private int least_hit;
 private int is_must_zz;
 private int zs_level;
 private int cur_zs_ct;
 private int least_score_yy;
 private int least_score_yw;
 private int area_id;
 private String coll_name;
 private String coll_code;
 private long coll_pro_id;
 private int is_must_wl;
 private int cur_year;
 private int is_two;
 private long logo;
 @ESId
 private int id;
 private String area;
 private int college_id;
 private String is_must_yy;
 private int is_double;
 private int least_score_zz;
 private int least_score_wl;
 private String grade;
 private int is_nine;
 private String pro_name;
 private int least_score_sx;
 private int relevanceSort;
 private int pre_avg;
 private String is_must_dl;
 private String profession_code;
 private int least_score_sw;
 private String is_must_ls;
 private int grade_zk;
 private int least_score_wy;
 private int is_must_hx;
 private int profession_id;
 private String is_grad;
 private String is_must_yw;
 private int is_must_sw;
 private int least_score_ls;
 private int least_score_dl;
 private String zs_memo;
 private String is_must_sx;
 private String introduce;
 private int is_must_wy;
 private int grade_bk;
 private String pre_name;
 private int least_score_hx;
 private String coll_domain;
 private int pre_wch;
 private List<String> courses;
}

定义服务

package com.rz.service; 
import com.rz.dto.ElasticZsInfoDto;
import com.rz.szwes.core.ElasticBaseService; 
 
/**
 * 招生索引操作服务
 *
 * @author sunziwen
 * @version 1.0
 * @date 2019/12/16 11:02
 **/
public class ElasticZsInfoService extends ElasticBaseService<ElasticZsInfoDto> { 
}

完毕。

已经可以进行索引和文档的crud操作了,至于复杂的检索操作就需要在xml中定义了。这里只介绍了我增强的功能,大部分功能都在bBoss中定义好了,读者可以去看bBoss文档(笔者认为的他的唯一缺陷是不能通过实体配合注解实现自动索引,还要每次手动指定xml位置,手动写mapping是很痛苦的事情,特此进行了增强)。

以上为个人经验,希望能给大家一个参考,也希望大家多多支持自学编程网。如有错误或未考虑完全的地方,望不吝赐教。

编程技巧