HIVE中窗口函数、分组排序

Hive

**row_number() over()**直接排序，10个第1名排名为1到10

select user_id,pay_time,amont,Row_Number() OVER (partition by user_id ORDER BY amont desc ) rank from tmp

**rank() over()**是跳跃排序，有两个第二名时接下来就是第四名

select workdept,salary,rank() over(partition by workdept order by salary) as dense_rank_order from emp order by workdept;

**dense_rank() over()**是连续排序，有两个第二名时仍然跟着第三名

select workdept,salary,dense_rank() over(partition by workdept order by salary) as dense_rank_order from emp order by workdept;