Hive之分析函数

目录

• 一、sum() over(partition by)
• 二、avg()、min()、max() over(partition)
• 三、row_number() over(partition by)
• 四、用over(partition by) 还是 group by

一、sum() over(partition by)

• 数据准备

cookie1,2015-04-10,1
cookie1,2015-04-11,5
cookie1,2015-04-12,7
cookie1,2015-04-13,3
cookie1,2015-04-14,2
cookie1,2015-04-15,4
cookie1,2015-04-16,4

• 查询语句

select 
   cookieid, 
   createtime, 
   pv, 
   sum(pv) over (partition by cookieid order by createtime rows between unbounded preceding and current row) as pv1, 
   sum(pv) over (partition by cookieid order by createtime) as pv2, 
   sum(pv) over (partition by cookieid) as pv3, 
   sum(pv) over (partition by cookieid order by createtime rows between 3 preceding and current row) as pv4, 
   sum(pv) over (partition by cookieid order by createtime rows between 3 preceding and 1 following) as pv5, 
   sum(pv) over (partition by cookieid order by createtime rows between current row and unbounded following) as pv6 
from cookie1;

• 查询结果

• 查询结果说明

  • pv1: 分组内从起点到当前行的pv累积，如，11号的pv1=2015-04-10号的pv + 2015-04-11号的pv, 2015-04-12号=10号+11号+12号
  • pv2: 同pv1
  • pv3: 分组内(cookie1)所有的pv累加
  • pv4: 分组内当前行+往前3行，如，11号=10号+11号， 12号=10号+11号+12号， 13号=10号+11号+12号+13号， 14号=11号+12号+13号+14号
  • pv5: 分组内当前行+往前3行+往后1行，如，14号=11号+12号+13号+14号+15号=5+7+3+2+4=21
  • pv6: 分组内当前行+往后所有行，如，13号=13号+14号+15号+16号=3+2+4+4=13，14号=14号+15号+16号=2+4+4=10

• partition by 的参数说明

如果不指定ROWS BETWEEN,默认为从起点到当前行;
如果不指定ORDER BY，则将分组内所有值累加;
关键是理解ROWS BETWEEN含义,也叫做WINDOW子句：
PRECEDING：往前
FOLLOWING：往后
CURRENT ROW：当前行
UNBOUNDED：起点，
UNBOUNDED PRECEDING 表示从前面的起点，
UNBOUNDED FOLLOWING：表示到后面的终点
–其他AVG，MIN，MAX，和SUM用法一样。

二、avg()、min()、max() over(partition)

  avg()、min()、max() over(partition) 与 sum() over(partition) 类似，都是对窗口做操作

• min() over(partition) 的查询语句

select 
   cookieid, 
   createtime, 
   pv, 
   min(pv) over (partition by cookieid order by createtime rows between unbounded preceding and current row) as pv1, -- 默认为从起点到当前行
   min(pv) over (partition by cookieid order by createtime) as pv2, --从起点到当前行，结果同pv1
   min(pv) over (partition by cookieid) as pv3, --分组内所有行
   min(pv) over (partition by cookieid order by createtime rows between 3 preceding and current row) as pv4, --当前行+往前3行
   min(pv) over (partition by cookieid order by createtime rows between 3 preceding and 1 following) as pv5, --当前行+往前3行+往后1行
   min(pv) over (partition by cookieid order by createtime rows between current row and unbounded following) as pv6  --当前行+往后所有行
from cookie1;

• 结果展示

三、row_number() over(partition by)

  row_number()从1开始，为每一条分组记录返回一个数字

  row_number() OVER (ORDER BY id DESC) 是先把id列降序，再为降序以后的每条id记录返回一个序号。

  row_number() OVER (PARTITION BY COL1 ORDER BY COL2) 表示根据COL1分组，在分组内部根据 COL2排序，而此函数计算的值就表示每组内部排序后的顺序编号（组内连续的唯一的)

• 数据准备

• 查询：需根据部门分组，显示每个部门的工资等级

SELECT *, Row_Number() OVER (partition by deptid ORDER BY salary desc) rank FROM employee

四、用over(partition by) 还是 group by

• 总结区别：over(partition by) 和 group by的区别

  • group by：单纯分组，要查询非group by字段时需要用 collect_set()[0]处理，或者子查询处理
  • over(partition by)：不仅能分组，还能同时查询非分区字段，不仅可以使用sum()、avg()、min()、max()等功能，还可以使用row_number() 对数据进行排名功能

• group by

  在hive中使用group by时，是不能select 非group by 字段的。

select name,sex from people group by sex;
---------------------------------------------------
会报错：
FAILED: SemanticException [Error 10025]: Line 1:15 Expression not in GROUP BY key 'name'

  解决这个问题的方式有很多：在子查询中做group by然后用left join 连接，在外层selec。还有就是用collect_set()包围这个非group by字段

select collect_set(name)[0],sex from people group by sex;

• over(partition by)

  当然，用over(partition by)也能解决分组问题,在分组的同时会对相同key的进行回填处理

  数据准备

  查询语句

select v1,v2,sum(v2) over(partition by v1) as sum from wmg_test;

  结果展示

  所以要做到取一条分组数据，就在外层去重

select distinct v1,sum_01                          
from (
select v1,sum(v2) over(partition by v1) as sum_01
from wmg_test
) a;

  结果展示