枢轴双和

Question

我正在使用此选择来获取列价格的数据透视。没关系，但我也想对列 M1、M2 和 M3、M4 求和。

SELECT *,25 M1Cijena, 25 M2Cijena, 16 M3Cijena, 16 M4Cijena
FROM (
    select u.pin Pin,u.firstname Name,u.lastname,sum(tmt.Price) Prices, tmt.type MealType
    from TA_Meals tm
    left outer join TA_MealsType tmt
    on tm.MealType = tmt.id     
    full outer join users u
    on u.pin = tm.pin
    where u.department = 1000001001
    group by u.pin,u.firstname,u.lastName,tmt.Type  
) as s
PIVOT
(
    SUM(Prices)
    FOR MealType IN (M1,M2,M3,M4)  **<-- here also want sum for M1,M2 and M3,M4.**

)AS pvt

Answer 1

为什么不只使用条件聚合？

select u.pin, u.firstname, u.lastname,
      sum(case when tmt.type = 'M1' then tmt.Price end) as m1,
      sum(case when tmt.type = 'M2' then tmt.Price end) as m2,
      sum(case when tmt.type = 'M3' then tmt.Price end) as m3,
      sum(case when tmt.type = 'M4' then tmt.Price end) as m4,
      sum(case when tmt.type in ('M1', 'M2') then tmt.Price end) as m1_2,
      sum(case when tmt.type in ('M3', 'M4') then tmt.Price end) as m3_4
from users u left join
     TA_Meals tm
     on u.pin = tm.pin left join
     TA_MealsType tmt
     on tm.MealType = tmt.id join
where u.department = 1000001001
group by u.pin, u.firstname, u.lastName;

请注意，我还修复了您的联接，因此更有意义。将where 与外连接一起使用通常会影响正在实现的连接类型。