Hive窗口函数

一、概述

• 窗口函数是一组特殊函数

1. 扫描多个输入行来计算每个输出值，为每行数据生成一行结果
2. 可以通过窗口函数来实现复杂的计算和聚合

• 语法

function (arg1,..., arg n) over ([partition by <...>] [order by <....>] [<window_clause>])

1. partition by类似于group by，未指定则按整个结果集
2. 只有指定order by子句之后才能进行窗口定义
3. 可同时使用多个窗口函数
4. 过滤窗口函数计算结果必须在外面一层

• 按功能可划分为：排序，聚合，分析

案例：

#得到男性个数,女性个数
select name,count(id) over(partition by info.sex) from emp_id;

create database test;
use test;
create table test(name string,date string,num int)
row format delimited
fields terminated by ',';

select *,count(*) over() from test;

select *,count(*) over(partition by name) from test;

//组内会按照顺序count
select *,count(*) over(partition by name order by num) from test;

//过滤计算,外面套一层子查询
select * from(select *,count(*) over(partition by name order by date desc) win1 from test) a where a.win1<3;

select *,count(1) over() from test where date like '2018-01-%';

二、排序

• ROW_NUMBER()
  对所有数值输出不同的序号，序号唯一连续

select *,row_number() over(partition by name) from test;

• RANK()
  对相同数值，输出相同的序号，下一个序号跳过（1,1,3）

select *, rank() over() win1 from test;

select *, rank() over(partition by name) win1 from test;

select *, rank() over(partition by name order by date) win1 from test;

• DENSE_RANK()
  对相同数值，输出相同的序号，下一个序号连续（1,1,2）

//不会跳过序号
select *,dense_rank() over(partition by name order by date) win1 from test;

• NTILE(n)
  将有序的数据集合平均分配到n个桶中, 将桶号分配给每一行，根据桶号，选取前或后 n分之几的数据

select *,ntile(4) over(partition by name) from test;

• LAG和LEAD
  lag返回当前数据行的上一行数据
  lead返回当前数据行的下一行数据

//查询顾客上次购买的时间
select *,lag(date) over(partition by name order by date) from test;

• first_value和last_value

select *,lag(date) over(partition by name order by date) lastdate,lag(num) over(partition by name order by date),first_value(num) over(partition by name order by date) from test;

• PERCENT_RANK()
  (目前排名- 1)/(总行数- 1)，值相对于一组值的百分比排名

三、窗口定义

• 窗口定义由[<window_clause>]子句描述
  用于进一步细分结果并应用分析函数

• 支持两类窗口定义
  行类型窗口
  范围类型窗口

• rank、ntile、dense_rank、cume_dist、percent_rank、lead、lag和row_number函数不支持与窗口子句一起使用

• 行窗口：根据当前行之前或之后的行号确定的窗口
  rows between <start_expr> and <end_expr>

• <start_expr>可以为下列值
  unbounded preceding: 窗口起始位置（分组第一行）
  current row：当前行
  n preceding/following：当前行之前/之后n行

• <end_expr>可以为下列值
  unbounded following : 窗口结束位置（分组最后一行）
  current row：当前行
  n preceding/following：当前行之前/之后n行

select *,count(1) over(partition by name order by buydate desc rows between unbounded preceding and current row ) from test;

select *,count(1) over(partition by name order by buydate desc rows between 1 preceding and current row) from test;

select *,count(1) over(partition by name order by buydate desc rows between 1 preceding and 1 following) from test;

• 范围窗口是取分组内的值在指定范围区间内的行
  该范围值/区间必须是数字或日期类型
  目前只支持一个order by列

select *,count(num) over(partition by name order by num range between 20 preceding and 10 following) rst from test;