hive处理数据

操作命令

数据预处理：剔除数据中任意字段为空值

INSERT OVERWRITE TABLE result01
    select * from salary
    where userid is not null and dept_id is not null and salarys is not null

剔除identity字段数值在0-100以外的值

INSERT OVERWRITE TABLE result
    SELECT * FROM hittable
    WHERE identity between 0 and 100 ;

查看总数

select count(*) from item;

查询提供啤酒或葡萄酒的第一项产品(Item01)的总行数

select count(*) as num from item where item='Beer' or item01 = 'Wine';

查询第一次购买产品中最畅销的产品排名(item01)

select item01,count(item01) as num from item 
group by item01 
order by num desc;

查询牛奶出现在每一行购买供品的概率

select b.num/a.num as rate from(
select count(*) num from item) a,
(select count(*) num from item
where item01=='Milk' 
or item02=='Milk' 
or item02=='Milk'
or item03=='Milk'
or item04=='Milk') b ;