6.postgresql--grouping()

CREATE TABLE sales (
    brand VARCHAR NOT NULL,
    segment VARCHAR NOT NULL,
    quantity INT NOT NULL,
    PRIMARY KEY (brand, segment)
);

 INSERT INTO sales (brand, segment, quantity)
VALUES
    ('ABC', 'Premium', 100),
    ('ABC', 'Basic', 200),
    ('XYZ', 'Premium', 100),
    ('XYZ', 'Basic', 300);

union all的实现

select brand,segment,sum(quantity) from sales group by brand,segment
union all
select brand,null,sum(quantity) from sales group by brand
union all
select null,segment,sum(quantity) from sales group by segment
union all
select null,null,sum(quantity) from sales ;

grouping sets实现

 select coalesce (brand,'所有品牌') brand,coalesce(segment,'所有类型') segment,sum(quantity) from sales  
   group by grouping sets((brand,segment),(brand),(segment),())

pg既然提供了grouping sets，就提供了对应了对应的grouping()函数，用来判断是否该纬度在该指标统计中是否被用到

用到，返回0
没用到，返回1

   select 
	grouping(brand) as brand_used,
	grouping(segment) as segment_used,
  brand,
  segment,
  sum(quantity) 
from sales
group by grouping sets((brand,segment),(brand),(segment),())
having grouping(brand) = 1