Question

我有一个带有A列和B列的sql表，以及第三列C = A-B。

例如：

Row 1: A = 10, B = 8, C = 2
Row 2: A = 7, B = 8, C = -1
Row 3: A = 0, B = 8, C = -8

我想在第四列D中总结第三列C的值。列D应该是最后一行的D值加上当前行的C值。例如：

Row 1: A = 10, B = 8, C = 2, D = 2
Row 2: A = 7, B = 8, C = -1, D = 1
Row 3: A = 0, B = 8, C = -8, D = -7

我尝试使用以下SQL：

SELECT myTable2.*, D = LAG(D) + C OVER (ORDER BY Id)
FROM 
(
    SELECT myTable.*, C = myTable.A - myTable.B
    FROM 
    (
        SELECT Id = 1, A = 10, B = 8
        UNION
        SELECT Id = 2, A = 7, B = 8
        UNION
        SELECT Id = 3, A = 0, B = 8
    ) myTable
) myTable2
ORDER BY myTable2.Id ASC

我可以LAG(C)，但不能LAG(D)。我有什么想法可以实现这个目标吗？

感谢。

Answer 1

由于您的数据没有ID，我按A降序排序。

SELECT A, B, A-B C, SUM(A-B) OVER (ORDER BY A DESC) D FROM (VALUES (10,8),(7,8),(0,8)) A(A,B)

A           B           C           D
----------- ----------- ----------- -----------
10          8           2           2
7           8           -1          1
0           8           -8          -7

(3 row(s) affected)

Answer 2

你可以试试这个：

SELECT myTable2.*,  D = ISNULL(SUM(C) OVER (ORDER BY Id), 0)
FROM 
(
    SELECT myTable.*, C = myTable.A - myTable.B
    FROM 
    (
        SELECT Id = 1, A = 10, B = 8
        UNION
        SELECT Id = 2, A = 7, B = 8
        UNION
        SELECT Id = 3, A = 0, B = 8
    ) myTable
) myTable2

Answer 3

create table 
#test
(
a int,
b int,
c int
)

insert into #test
select 1,2,3
union all
select 3,4,5

with cte
as
(
select *,0 as 'd', row_number() over (order by a,b,c) as rn
from #test
)
select *,(select sum(c+d) from cte c1 where c1.rn<=c2.rn) 'D'
from cte c2

如何在单个sql select语句中使用滞后的前一行添加值？

3 个答案: