hive创建udf

1. 目的
   根据hive表中的多个字段创建一个json字符串，因此需要实现一个udf
2. 创建udf

package com.ke.search_reco.toJson;
import com.alibaba.fastjson.JSON;
import org.apache.hadoop.hive.ql.exec.UDF;

import java.util.HashMap;
import java.util.Map;

public class toJsonStr extends UDF {
    public String evaluate(String... args) {
        Map<String, String> res = new HashMap<>();
        for(int i=0;i<args.length;i+=2) {
            res.put(args[i], args[i+1]);
        }
        return JSON.toJSONString(res);

    }
}

3.pom.xml中添加依赖

    <dependencies>
        <dependency>
            <groupId>org.apache.hive</groupId>
            <artifactId>hive-exec</artifactId>
            <version>1.2.1</version>
        </dependency>
        <dependency>
            <groupId>com.alibaba</groupId>
            <artifactId>fastjson</artifactId>
            <version>1.2.70</version>
        </dependency>
    </dependencies>

<build>
    <plugins>
     <!--以下配置是为了保证打jar包时相关依赖一同打包-->
        <plugin>
            <groupId>org.apache.maven.plugins</groupId>
            <artifactId>maven-assembly-plugin</artifactId>
            <configuration>
                <descriptorRefs>
                    <descriptorRef>jar-with-dependencies</descriptorRef>
                </descriptorRefs>
            </configuration>
            <executions>
                <execution>
                    <id>make-assembly</id>
                    <phase>package</phase>
                    <goals>
                        <goal>single</goal>
                    </goals>
                </execution>
            </executions>
        </plugin>
     </plugins>
   </build>

4.通过maven打jar包
（1）点击idea右侧maven（2）点击package
（3）运行后在target目录下生成两个jar文件，一个是带依赖的jar文件，另一个是不带依赖的jar文件

5.将带依赖的jar文件put到hdfs任意目录

hadoop fs -put hive-udf-1.0-SNAPSHOT-jar-with-dependencies.jar /user/lip/tools/

6.在hive sql中使用udf

add jar /user/lip/tools/hive-udf-1.0-SNAPSHOT-jar-with-dependencies.jar;
create temporary function toJsonStr as 'com.ke.search_reco.toJson.toJsonStr';
select
    id,
	toJsonStr(
		'key_1', col_1,
		'key_2', col_2)
from test_tb