日期处理UDF

package com.xxx.udf;

import org.apache.hadoop.hive.ql.exec.UDF;

import java.text.ParseException;
import java.text.SimpleDateFormat;
import java.util.Date;

public class DateProcessUDF extends UDF {
    public String evaluate(String input) {
        String start = input.split(",")[0];
        String end = input.split(",")[1];
        SimpleDateFormat simpleDateFormat = new SimpleDateFormat("yyyy-MM-dd");
        StringBuffer sb = new StringBuffer();
        try {
            long startTime = simpleDateFormat.parse(start).getTime();
            long endTime = simpleDateFormat.parse(end).getTime();
            while (startTime <= endTime) {
                sb.append(simpleDateFormat.format(new Date(startTime)).toString() + ",");
                startTime += 86400000;
            }
            return sb.toString().substring(0, sb.toString().length() - 1);
        } catch (ParseException e) {
            e.printStackTrace();
            return null;
        }
    }
}
//select date_process(concat(to_date(start_time),',',to_date(end_date))) 获取开始时间到结束时间的日期序列，逗号分隔
//在使用行转列 获取每个日期

    select cust_code,regexp_replace(trace_date,'-','') as trace_date
    from
   (
    select cust_code,date_process(concat(to_date(start_time),',',to_date(end_time))) as trace_date_array
    from db.xxx
    where del_flag='1' and regexp_replace(to_date(start_time),'-','')<='20190515'
          and '20190515'<=regexp_replace(to_date(end_time),'-','')
    ) tt1 
    lateral view explode (split(trace_date_array,',')) t as trace_date
    where regexp_replace(trace_date,'-','')<='20190515'

转载于:https://www.cnblogs.com/yin-fei/p/10879552.html