hive自定义udf函数

1.创建一个简单的maven项目

2.pom文件设置需要的依赖

<?xml version="1.0" encoding="UTF-8"?>
<project xmlns="http://maven.apache.org/POM/4.0.0"
         xmlns:xsi="http://www.w3.org/2001/XMLSchema-instance"
         xsi:schemaLocation="http://maven.apache.org/POM/4.0.0 http://maven.apache.org/xsd/maven-4.0.0.xsd">
    <modelVersion>4.0.0</modelVersion>

    <groupId>org.example</groupId>
    <artifactId>hiveTest</artifactId>
    <version>1.0-SNAPSHOT</version>


    <!--根据要连接的hadoop和hive，设置与自己集群版本匹配的版本-->
   <properties>
       <hadoop.version>2.6.0-cdh5.13.3</hadoop.version>
       <hive.version>1.1.0-cdh5.13.3</hive.version>
   </properties>

   <!--因为使用cdh的hadoop和hive，因此要添加cdh官方的repository，才能够下载相对应的依赖包-->
    <!--如果使用apache版本的hadoop和hive，则不需要添加-->
    <repositories>
        <repository>
            <id>cloudera</id>
            <url>http://repository.cloudera.com/artifactory/cloudera-repos</url>
        </repository>
    </repositories>


    <dependencies>

        <dependency>
            <groupId>org.apache.hadoop</groupId>
            <artifactId>hadoop-common</artifactId>
            <version>${hadoop.version}</version>
        </dependency>


        <dependency>
            <groupId>org.apache.hive</groupId>
            <artifactId>hive-exec</artifactId>
            <version>${hive.version}</version>
        </dependency>

        <dependency>
            <groupId>junit</groupId>
            <artifactId>junit</artifactId>
            <version>4.10</version>
            <scope>test</scope>
        </dependency>

    </dependencies>


</project>

3.具体实现代码

import org.apache.hadoop.hive.ql.exec.*;
import org.apache.hadoop.io.Text;


public class HelloUDF01 extends UDF{

    public Text evaluate(Text input){
        return  new Text("hello" +':'+ input);

    }
    public static void main(String[] args) {
        HelloUDF01 udf =new HelloUDF01();
        Text result = udf.evaluate(new Text("zhangsan"));
        System.out.println(result.toString());
    }

}

4.打包jar

view->tool_windows->build->package->run

5.上传jar到hdfs目录

hdfs   dfs   -put    /home/hadoop/read_xwf/hiveTest-1.0-SNAPSHOT.jar     /user/xin/

6.注册udf函数

 create   function   func.HelloUDF    as   'HelloUDF01'   using    jar     'hdfs://nameservice/user/xin/hiveTest-1.0-SNAPSHOT.jar';
as前面是自己定义的函数名，as后面是自己代码的方法名(见下面的截图)，jar后面是jar的具体位置要补全(可以从建表语句中的location看到具体的位置，复制到前面即可)

7.测试

select 'lisi',func.HelloUDF('lisi');