【博学谷学习记录】超强总结，用心分享|Hive开窗函数

场景

如果我们需要想对一组数据进行分组,我们通过会使用聚合函数group by,但是它有个缺点,每组数据只能返回一行数据,且只能返回分组字段和聚合字段,如果我们想在分组的同时还保留数据详细的信息,这时我们就可以使用窗口分析函数

演示数据

user1,2018-04-10,1
user1,2018-04-11,5
user1,2018-04-12,7
user1,2018-04-13,3
user1,2018-04-14,2
user1,2018-04-15,4
user1,2018-04-16,4
user2,2018-04-10,2
user2,2018-04-11,3
user2,2018-04-12,5
user2,2018-04-13,6
user2,2018-04-14,3
user2,2018-04-15,9
user2,2018-04-16,7

创建表并加载数据

create table open_window(
    userid string,
    createtime string,
    pv int
)
    row format delimited
    fields terminated by ',';

load data local inpath '/export/testdatas/openwindow.txt' into table open_window;

查询验证

select * from open_window;

分组排名

row_number

从1开始按顺序排序,不间断,无重复

select *,row_number() over (partition by userid order by pv desc ) as rk from open_window;

rank

从1开始按顺序排序,有间断,有重复,碰到相同数据会显示相同排名,但是排名序号会有断层

select *,rank() over (partition by userid order by pv desc ) as rk from open_window;

dense_rank

从1开始按顺序排序,无间断,有重复,碰到相同数据会显示相同排名,排名序号不会断层

select *,rank() over (partition by userid order by pv desc ) as rk from open_window;

分组求和,平均,最大,最小

sum

需要指定计算字段,排序方式,计算窗口(范围),前两者都会影响计算结果

over中常用关键字

partition by: 指定分区(组)字段
order by: 指定排序方式,不指定默认整个分组为计算窗口
rows between 指定计算窗口(范围), 不指定默认从第一行到当前行

rows between 子句,也叫window 子句

preceding:表示往前几行
current row:表示当前行
following:表示往后几行
unbounded preceding:表示从第一行开始
unbounded following:表示到最后一行终止

只指定分区字段 partition by

默认计算范围,整个分组

select *, sum(pv) over (partition by userid) as sum from open_window;

不指定rows between

默认计算范围,第一行到当前行

select *, sum(pv) over (partition by userid order by createtime) as sum from open_window;

-- 等效写法
select *, sum(pv) over (partition by userid order by createtime rows unbounded preceding) as sum from open_window;
select *, sum(pv) over (partition by userid order by createtime rows between unbounded preceding and current row ) as sum from open_window;

指定rows between

从当前行的前3行开始到后1行结束

指定计算范围从当前行的前3行开始到后1行结束

select *, sum(pv) over (partition by userid order by createtime rows between 3 preceding and 1 following) as sum from open_window;

从当前行到最后一行

select *, sum(pv) over (partition by userid order by createtime rows between current row and unbounded following) as sum from open_window;

avg,min,max的开窗函数与sum用法一致,把函数名改一下就可以了