Question

我在下面有一个表，它是许多临时表的结果，它涉及join和order by子句。

Type                Col1       Col2         Col3        Col4       Col5
Within 30 days      493868      426428      401194      268929      399480
Within 60 days      745         13118       35577       31577       30536
Within 90 days      NULL        28222       27312       45085       10432
Within 120 days     NULL        12039       13999       38239       11777
Grater than120 days NULL        14806       16531       110783      13091
Blank               NULL        NULL         NULL       NULL        29297
Total               494613      494613       494613     494613      494613

我想要输出如下：当前行值应该是前一行值+当前值..否则直到它与总数匹配为止。

Type                    Col1       Col2         Col3       Col4     Col5
Within 30 days          493868     426428       401194     268929      399480
Within 60 days           494613    439546       436771      300506     430016
Within 90 days           494613    467768       464083      345591     440448
Within 120 days          494613    479807       478082      383830     452225
Greater than 120 days    494613    494613       494613      494613     465316
Blank                    494613    494613       494613      494613     494613
Total                    494613    494613       494613      494613     494613

由于我使用了orderby子句，因此我无法使用temptable和派生表。请帮助。我使用了查询：

select e.[type],o.[Col1],h.[Col2],s.[Col3],d.[Col4],e.[Col5] from #table1 e left outer join #table2 d on e.[type]=d.[type] left outer join #table3 s on e.[type]=s.[type] left outer join #table4 h on e.[type]=h.[type] left outer join #table5 o on e.[type]=o.[type] order by case when e.[type] ='Within 30 days' then 1 when e.[type] ='Within 60 days' then 2 when e.[type] ='Within 90 days' then 3 when e.[type] ='Within 120 days' then 4 when e.[type] ='Greater than 120 days' then 5 when e.[type] ='Blank' then 6 else 7 end

Answer 1

使用SUM OVER()窗口聚合函数计算运行总计

Select type,
        sum([Count(A)]) Over(order by order_col) as [Count(A)],
        sum([Count(B)]) Over(order by order_col) as [Count(B)],
        sum([Count(C)]) Over(order by order_col) as [Count(C)],
        sum([Count(D)]) Over(order by order_col) as [Count(D)],
        sum([Count(E)]) Over(order by order_col) as [Count(E)]
From your_result

您可以在CASE页面上添加sum over()语句，以避免汇总type = Total

对于年长的诽谤

SELECT type,
        cs.[Count(A)],
        cs.[Count(B)],
        cs.[Count(C)],
        cs.[Count(D)],
        cs.[Count(E)]
FROM   your_result t
        CROSS apply (SELECT Sum([Count(A)]) [Count(A)],
                            Sum([Count(B)]) [Count(B)],
                            Sum([Count(C)]) [Count(C)],
                            Sum([Count(D)]) [Count(D)],
                            Sum([Count(E)]) [Count(E)]
                    FROM   your_result t1
                    WHERE  t1.order_col <= t.order_col) cs

将上一行添加到当前行 - sql server 2008

1 个答案: