hive-udaf开发实例1

Posted 2022-12-13 forrestxingyunfei

1，pom配置：
<dependency>
    <groupId>org.apache.hive</groupId>
    <artifactId>hive-exec</artifactId>
    <version>1.1.0-cdh5.8.3</version>
</dependency>
<dependency>
    <groupId>org.apache.hadoop</groupId>
    <artifactId>hadoop-client</artifactId>
    <version>2.6.0-cdh5.8.3</version>
</dependency>


2,java代码：
package com.ibeifeng.udaf;


import org.apache.hadoop.hive.ql.exec.UDAF;
import org.apache.hadoop.hive.ql.exec.UDAFEvaluator;


public class Wm_concat extends UDAF


public static class myUDAFEval implements UDAFEvaluator


private PartialResult partial = new PartialResult();
public static class PartialResult
String result = "" ;
String delimiter = null;

@Override
public void init()
// TODO Auto-generated method stub
partial.result = "";

public  boolean  iterate(String value, String deli)
if (value == null || "null".equalsIgnoreCase(value))
return true;

if (partial.delimiter == null)
partial.delimiter = deli;

if (partial.result.length() > 0)
partial.result = partial.result.concat(partial.delimiter);

partial.result = partial.result.concat(value);
return true;

public  PartialResult  terminatePartial()
return partial;

public  boolean  merge(PartialResult other)
if (other == null)
return true;

if (partial.delimiter == null)
partial.delimiter = other.delimiter ;
partial.result = other.result ;
else
if(partial.result.length() > 0)
partial.result = partial.result.concat(partial.delimiter);

partial.result = partial.result.concat(other.result);

return true ;

public String terminate()
if (partial == null || partial.result.length() == 0)
return null;

return partial.result ;







3，导出jar包
4，hive中导入jar包
addjar /opt/beifeng/jars/hive_UDF.jar;
createtemporary function Wm_concat as 'com.ibeifeng.udf.GetCmsID ';


5，hive sql使用udf
select getcmsid(url) from info_tab group by id;