HIVE自定义函数

Posted 2020-11-05 李杰然

UDF  操作单个数据行，产生单个数据行；

1.
[[email protected] hhh]$ vi TimeFormat.java

import java.sql.Date;
import java.text.SimpleDateFormat;
import org.apache.hadoop.hive.ql.exec.UDF;
public class TimeFormat extends UDF {
 public String evaluate(String num){
  Date d=new Date(Long.decode(num));
  SimpleDateFormat sdf =new SimpleDateFormat("yyyy-MM-dd HH:MM:SS");
  return sdf.format(d) ;
 }
}


2.打包
[[email protected] hhh]$ /usr/jdk1.7.0_25/bin/javac -classpath /home/hadoop/hadoop-0.20.2-cdh3u5/hadoop-core-0.20.2-cdh3u5.jar:/home/hadoop/hive-0.9.0-bin/lib/hive-exec-0.9.0.jar TimeFormat.java

[[email protected] hhh]$ vi main.mf
Manifest-Version: 1.0

[[email protected] hhh]$ /usr/jdk1.7.0_25/bin/jar cvfm TF.jar main.mf TimeFormat.class

3.
hive> add jar /home/hadoop/hhh/TF.jar;
hive> CREATE TEMPORARY FUNCTION TFF AS ‘TimeFormat‘;

create table ss(id bigint)
     row format delimited
     fields terminated by ‘\t‘
     stored as textfile;


hive> load data local inpath ‘/home/hadoop/c.txt‘ into table ss;

[[email protected] ~]$ date +%s （显示当前时间 转换成秒）
（时间戳为 把时间转换成秒 *1000 变为 毫秒点位）

[[email protected] ~]$ vi c.txt
1417792627000



TFF把时间戳 转换为 ("yyyy-MM-dd HH:MM:SS")
hive> select TFF(time) from ha;


-----------------------------------------------
[[email protected] hhh]$ vi hello.java
import org.apache.hadoop.hive.ql.exec.UDF;
public class hello extends UDF {
    public String evaluate(String str) {

        try {

            return "HelloWorld " + str;

        } catch (Exception e) {

            return null;

        }
    }
}

2.打包
[[email protected] hhh]$ /usr/jdk1.7.0_25/bin/javac -classpath /home/hadoop/hadoop-0.20.2-cdh3u5/hadoop-core-0.20.2-cdh3u5.jar:/home/hadoop/hive-0.9.0-bin/lib/hive-exec-0.9.0.jar hello.java

[[email protected] hhh]$ vi main.mf
Manifest-Version: 1.0

[[email protected]h91 hhh]$ /usr/jdk1.7.0_25/bin/jar cvfm H.jar main.mf hello.class

3.
hive> add jar /home/hadoop/hhh/H.jar;
hive> CREATE TEMPORARY FUNCTION HH AS ‘hello‘;

hive> select HH(name) from ha;