Sql 从同一表中查找不匹配ID的总和

例如，我在下表

ID    DetailedID    ReferenceID    ItemID    Qty
01           101             80      A101    100
01           102             90      A102    200
01           103              0      A101    050
01           104              0      A109    100
02           105             81      1010    100
02           106             82      1010    100
03           107              0      1111    020
03           108             81      1010    100
03           109              0      X200    010
03           110              0      1010    020

现在，我需要ReferenceID=0和ItemID不匹配的数量之和，根据ID-wise，我将显示该数量之和，以添加一列和该ID的detaileId的最小值

我想要输出像

ID    DetailedID    ReferenceID    ItemID    Qty    AddQty
01           101             80      A101    100    055   
01           102             90      A102    200    NULL
01           103              0      A101    050    NULL
01           104              0      A109    055    NULL
02           105             81      1010    100    NULL
02           106             82      1010    100    NULL
03           107              0      1111    020    030
03           108             81      1010    100    NULL
03           109              0      X200    010    NULL
03           110              0      1010    020    NULL

您可以尝试以下方法：

DECLARE @DataSource TABLE
(
    [ID] TINYINT
   ,[DetailedID] TINYINT
   ,[ReferenceID] TINYINT
   ,[ItemID] VARCHAR(12)
   ,[Qty] TINYINT
);

INSERT INTO @DataSource ([ID], [DetailedID], [ReferenceID], [ItemID], [Qty])
VALUES (01, 101, 80, 'A101', 100)
      ,(01, 102, 90, 'A102', 200)
      ,(01, 103, 0, 'A101', 50)
      ,(01, 104, 0, 'A109', 55)
      ,(02, 105, 81, '1010', 100)
      ,(02, 106, 82, '1010', 100)
      ,(03, 107, 0, '1111', 20)
      ,(03, 108, 81, '1010', 100)
      ,(03, 109, 0, 'X200', 10)
      ,(03, 110, 0, '1010', 20);

WITH DataSource AS
(
    SELECT DS.[ID]
          ,SUM(DS.[Qty]) AS [AddQty]
    FROM @DataSource DS
    WHERE DS.[ReferenceID] = 0
        AND NOT EXISTS(SELECT 1 FROM @DataSource I WHERE I.[ItemID] = DS.[ItemID] AND I.[ReferenceID] <> 0) 
    GROUP BY DS.[ID]
), DataSourceWithMinDetailedID AS
(
    SELECT [ID]     
          ,MIN([DetailedID]) AS [MinDetailedID]
    FROM @DataSource 
    GROUP BY [ID]
)
SELECT A.*
      ,C.[AddQty]
FROM @DataSource A
LEFT JOIN  DataSourceWithMinDetailedID B
    ON A.[ID] = B.[ID]
LEFT JOIN DataSource C
    ON B.[ID] = C.[ID]
    AND A.[DetailedID] = B.[MinDetailedID];

请检查以下内容 您将认识到，除了使用CASE语句外，我还使用了带有partitionby子句的和聚合函数 我希望这是清楚的

/*
create table ItemQuantity (
ID varchar(2),
DetailedID int,
ReferenceID int,
ItemID varchar(5),
Qty int)
insert into ItemQuantity values 
('01',           101,             80      ,'A101',    100),
('01',           102,             90      ,'A102',    200),
('01',           103,              0      ,'A101',    050),
('01',           104,              0      ,'A109',    100),
('02',           105,             81      ,'1010',    100),
('02',           106 ,            82      ,'1010',    100),
('03',           107  ,            0      ,'1111',    020),
('03',           108   ,          81      ,'1010',    100),
('03',           109    ,          0      ,'X200',    010),
('03',           110     ,         0      ,'1010',    020)
*/
;with cte as (
select 
    * ,
    QtyTemp = case when ReferenceID = 0 then Qty else 0 end,
    Flag = case when exists(select * from ItemQuantity k where i.ItemID = k.ItemID and k.ReferenceID <> 0) then 1 else 0 end,
    rn = row_number() over (partition by ID order by DetailedID)
from ItemQuantity i
)
select ID, DetailedID, ReferenceID, ItemID, Qty,
AddQty = case when rn = 1 then SUM(case when ReferenceID = 0 and Flag = 0 then QtyTemp else 0 end) OVER (Partition By ID) else null end
from cte

关于您的样本，您在所需输出的一行中有55个作为数量。但该行的样本数据有100个。请根据这一点检查数据和输出

我的脚本执行以下输出

---

我希望这会有所帮助，

您能显示您尝试过的查询吗？第一行的055在哪里。试图理解逻辑