Hadoop大数据生态：SpringBoot整合hive，使用spring的jdbcTemplate操作Hive

秒速五厘米 2022-01-30 06:31 4261阅读 0赞

### 简介 ###

1.  使用开发工具连接hive客户端，整合SpringBoot与Hive的连接，同时进行数据的增删查改；
2.  创建hive连接，放入druid数据库连接池，自动注入jdbctemplate；
3.  开发前提：已经安装hive客户端，可参考博客：[阿里云ECS7安装搭建：hive-2.1.1客户端][ECS7_hive-2.1.1]；
4.  版本配置：**hadoop2.7.6，hive1.3.3**；

### 实践 ###

*  1. 首先，pom文件加入hive依赖；

<dependency>
    	<groupId>org.apache.hadoop</groupId>
    	<artifactId>hadoop-common</artifactId>
    	<version>2.6.0</version>
    </dependency>
    
    <dependency>
    	<groupId>org.apache.hadoop</groupId>
    	<artifactId>hadoop-mapreduce-client-core</artifactId>
    	<version>2.6.0</version>
    </dependency>
    
    <dependency>
    	<groupId>org.apache.hadoop</groupId>
    	<artifactId>hadoop-mapreduce-client-common</artifactId>
    	<version>2.6.0</version>
    </dependency>
    
    <dependency>
    	<groupId>org.apache.hadoop</groupId>
    	<artifactId>hadoop-hdfs</artifactId>
    	<version>2.6.0</version>
    </dependency>
    
    <dependency>
    	<groupId>jdk.tools</groupId>
    	<artifactId>jdk.tools</artifactId>
    	<version>1.8</version>
    	<scope>system</scope>
    	<systemPath>${JAVA_HOME}/lib/tools.jar</systemPath>
    </dependency>
    
    <dependency>
    	<groupId>org.springframework.boot</groupId>
    	<artifactId>spring-boot-configuration-processor</artifactId>
    	<optional>true</optional>
    </dependency>
    
    
    <dependency>
    	<groupId>org.apache.hive</groupId>
    	<artifactId>hive-jdbc</artifactId>
    	<version>2.1.1</version>
    	<exclusions>
    		<exclusion>
    			<groupId>org.eclipse.jetty.aggregate</groupId>
    			<artifactId>*</artifactId>
    		</exclusion>
    	</exclusions>
    </dependency>

*  2. 配置文件yml，添加数据源属性配置；

#配置多个数据源(这里测试hive数据源)
    spring:
        datasource:
          hive: #hive数据源
              url: jdbc:hive2://47.100.200.200:9019/default
              type: com.alibaba.druid.pool.DruidDataSource
              username: sixmonth
              password: sixmonth
              driver-class-name: org.apache.hive.jdbc.HiveDriver
          commonConfig: #连接池统一配置，应用到所有的数据源 
             initialSize: 1
             minIdle: 1
             maxIdle: 5
             maxActive: 50
             maxWait: 10000
             timeBetweenEvictionRunsMillis: 10000
             minEvictableIdleTimeMillis: 300000
             validationQuery: select 'x'
             testWhileIdle: true
             testOnBorrow: false
             testOnReturn: false
             poolPreparedStatements: true
             maxOpenPreparedStatements: 20
             filters: stat

*  3. 新建属性统一配置类：DataSourceProperties；

package com.springboot.sixmonth.common.config.druid;
    
    import java.util.Map;
    
    import org.springframework.boot.context.properties.ConfigurationProperties;
    
    /**
     * -统一属性控制类，获取配置文件属性
     * @author sixmonth
     * @Date 2019年5月18日
     *
     */
    @ConfigurationProperties(prefix = DataSourceProperties.DS, ignoreUnknownFields = false)
    public class DataSourceProperties {
    	final static String DS = "spring.datasource";
    	
    	private Map<String,String> hive;
    	
    	private Map<String,String> commonConfig;
    	
    
    	/*为节省空间，这里省略set和get方法，需自行添加上去*/
    
    }

*  4. 新建数据库连接池通配属性类，可应用到所有的数据源：DataSourceCommonProperties；

package com.springboot.sixmonth.common.config.druid;
    
    import org.springframework.boot.context.properties.ConfigurationProperties;
    
    /**
     * -扩展连接池，通用配置属性，可应用到所有数据源
     * @author sixmonth
     * @Date 2019年5月18日
     *
     */
    @ConfigurationProperties(prefix = DataSourceCommonProperties.DS, ignoreUnknownFields = false)
    public class DataSourceCommonProperties {
    	final static String DS = "spring.datasource.commonConfig";
    
    	private int initialSize = 10;
    	private int minIdle;
    	private int maxIdle;
    	private int maxActive;
    	private int maxWait;
    	private int timeBetweenEvictionRunsMillis;
    	private int minEvictableIdleTimeMillis;
    	private String validationQuery;
    	private boolean testWhileIdle;
    	private boolean testOnBorrow;
    	private boolean testOnReturn;
    	private boolean poolPreparedStatements;
    	private int maxOpenPreparedStatements;
    	private String filters;
    
    	private String mapperLocations;
    	private String typeAliasPackage;
    
    	/*为节省空间，这里省略set和get方法，需自行添加上去*/
    	
    }

*  5. 新建hive数据源配置类，加载配置属性文件，注入bean容器：HiveDruidConfig；

package com.springboot.sixmonth.common.config.druid.hive;
    
    import java.sql.SQLException;
    
    import javax.sql.DataSource;
    
    import org.slf4j.Logger;
    import org.slf4j.LoggerFactory;
    import org.springframework.beans.factory.annotation.Autowired;
    import org.springframework.beans.factory.annotation.Qualifier;
    import org.springframework.boot.context.properties.EnableConfigurationProperties;
    import org.springframework.context.annotation.Bean;
    import org.springframework.context.annotation.Configuration;
    
    import com.alibaba.druid.pool.DruidDataSource;
    import com.springboot.sixmonth.common.config.druid.DataSourceCommonProperties;
    import com.springboot.sixmonth.common.config.druid.DataSourceProperties;
    
    /**
     * -配置hive数据源
     * @author sixmonth
     * @Date 2019年5月18日
     *
     */
    @Configuration
    @EnableConfigurationProperties({DataSourceProperties.class,DataSourceCommonProperties.class})//将配置类注入到bean容器，使ConfigurationProperties注解类生效
    public class HiveDruidConfig {
    
    	    private static Logger logger = LoggerFactory.getLogger(HiveDruidConfig.class);
    
    	    @Autowired
    	    private DataSourceProperties dataSourceProperties;
    	    
    	    @Autowired
    	    private DataSourceCommonProperties dataSourceCommonProperties;
    
    	    @Bean("hiveDruidDataSource") //新建bean实例
    	    @Qualifier("hiveDruidDataSource")//标识
    	    public DataSource dataSource(){
    	        DruidDataSource datasource = new DruidDataSource();
    
    	        //配置数据源属性
    	        datasource.setUrl(dataSourceProperties.getHive().get("url"));
    	        datasource.setUsername(dataSourceProperties.getHive().get("username"));
    	        datasource.setPassword(dataSourceProperties.getHive().get("password"));
    	        datasource.setDriverClassName(dataSourceProperties.getHive().get("driver-class-name"));
    	        
    	        //配置统一属性
    	        datasource.setInitialSize(dataSourceCommonProperties.getInitialSize());
    	        datasource.setMinIdle(dataSourceCommonProperties.getMinIdle());
    	        datasource.setMaxActive(dataSourceCommonProperties.getMaxActive());
    	        datasource.setMaxWait(dataSourceCommonProperties.getMaxWait());
    	        datasource.setTimeBetweenEvictionRunsMillis(dataSourceCommonProperties.getTimeBetweenEvictionRunsMillis());
    	        datasource.setMinEvictableIdleTimeMillis(dataSourceCommonProperties.getMinEvictableIdleTimeMillis());
    	        datasource.setValidationQuery(dataSourceCommonProperties.getValidationQuery());
    	        datasource.setTestWhileIdle(dataSourceCommonProperties.isTestWhileIdle());
    	        datasource.setTestOnBorrow(dataSourceCommonProperties.isTestOnBorrow());
    	        datasource.setTestOnReturn(dataSourceCommonProperties.isTestOnReturn());
    	        datasource.setPoolPreparedStatements(dataSourceCommonProperties.isPoolPreparedStatements());
    	        try {
    	            datasource.setFilters(dataSourceCommonProperties.getFilters());
    	        } catch (SQLException e) {
    	            logger.error("Druid configuration initialization filter error.", e);
    	        }
    	        return datasource;
    	    }
    	    
    }

*  6. 新建JdbcTemplate类，自动注入hive数据源：HiveJdbcBaseDaoImpl；

package com.springboot.sixmonth.common.config.jdbcConfig;
    
    import javax.sql.DataSource;
    
    import org.springframework.beans.factory.annotation.Autowired;
    import org.springframework.beans.factory.annotation.Qualifier;
    import org.springframework.jdbc.core.JdbcTemplate;
    import org.springframework.stereotype.Repository;
    
    /**
     * -注入hive数据源
     * @author sixmonth
     * @Date 2019年5月18日
     *
     */
    @Repository
    public class HiveJdbcBaseDaoImpl {
    	
    	private JdbcTemplate jdbcTemplate;
    
    	public JdbcTemplate getJdbcTemplate() {
    		return jdbcTemplate;
    	}
    
    	@Autowired
    	public void setJdbcTemplate(@Qualifier("hiveDruidDataSource") DataSource dataSource) {
    		this.jdbcTemplate = new JdbcTemplate(dataSource);
    	}
    	
    }

*  7. 新建测试dao类，集成hiveJdbctemplate，调用hql，查询hive数据库数据：TestHiveDao；

package com.springboot.sixmonth.dao.jdbcDao.test;
    
    import org.springframework.stereotype.Repository;
    import com.springboot.sixmonth.common.config.jdbcConfig.HiveJdbcBaseDaoImpl;
    
    /**
     * -测试hive连接
     * @author sixmonth
     * @Date 2019年5月18日
     *
     */
    @Repository
    public class TestHiveDao extends HiveJdbcBaseDaoImpl{
    
    	/**
    	 * 测试获取hive数据库数据信息
    	 * @return
    	 */
    	public String test() {
    		String sql = "SELECT name from sixmonth limit 1";
    		String param = this.getJdbcTemplate().queryForObject(sql,String.class);
    		return param;
    	}
    	
    }

*  8. 使用方法：自动注入TestHiveDao，即可直接调用方法使用；

@Autowired
    private TestHiveDao testHiveDao;

### 注意事项 ###

1.  hive连接的用户名和密码可在hive安装的时候自定义，具体可参考博主上一篇博客：[Hadoop大数据生态：Hive自定义设置连接用户名和密码][Hadoop_Hive]；
2.  jdbctemplate操作hive数据源的时候，基本支持所用到的hql语言，可自行测试；
3.  Hive与关系型数据库不一样，关系型数据库都是为了实时查询的业务进行设计的，而**Hive则是为海量数据做数据挖掘设计的，实时性很差**，一般不做实时性查询，如果需要结合实时性查询，可了解整合spark；

### 总结 ###

1.  SpringBoot可整合多种数据源，本篇博客仅针对hive数据源的配置，多数据源整合的话，可参考博主的另一篇博客：[SpringBoot多数据源配置(hive/mysql)][SpringBoot_hive_mysql]；
2.  实践是检验认识真理性的唯一标准，自己动手，丰衣足食~~

[ECS7_hive-2.1.1]: https://blog.csdn.net/alan_liuyue/article/details/90289899
[Hadoop_Hive]: https://blog.csdn.net/alan_liuyue/article/details/90299035
[SpringBoot_hive_mysql]: https://blog.csdn.net/alan_liuyue/article/details/90370844