我有一个问题我创建了去年我所有接收器的总数，我们返回的所有产品以及原因。我遇到的问题是如何正确地将表格与每个客户的总接收者和表格中的问题列表正确地连接起来。
当我进行连接时，我最终得到了来自接收器表的总和重复。我理解为什么会这样，我只是不知道如何解决它。
FirstCTE
Receiver Qty fvendno
---------------------------
21 ACO
2 AEG
16 AER
139 AET
SecondCTE
fdincidate Fkey_ID fcvendno FPROBLEM
-----------------------------------------------------
3/8/16 9:57 2451 AET Wrong Quantity
3/8/16 9:59 2452 AET Wrong Quantity
4/18/16 10:41 2527 AET Wrong Quantity
4/28/16 11:57 2540 AET Defective Material
6/24/16 14:19 2595 AET Wrong Material
11/2/16 10:51 2700 AET Wrong Material
12/1/16 12:29 2804 AET Wrong Material
1/9/17 14:40 2838 AET Wrong Material
1/18/05 16:02 122 AIN
11/4/15 11:52 2316 ALF Wrong Quantity
6/2/14 10:58 1565 AVL Wrong Quantity
6/2/14 12:24 1567 AVL Defective Material
查询：
SELECT
ap.fcompany,
FC.[Receiver Qty],
COUNT(SC.FKey_ID) AS 'Number of issues per vendor',
CASE sc.FPROBLEM
WHEN 'Wrong Quantity'
THEN 'Shortage'
WHEN 'Defective Material'
THEN 'Defective Material'
WHEN 'Wrong Material'
THEN 'Wrong Material'
END AS fproblem
FROM
SecondCTE AS SC
LEFT OUTER JOIN
FirstCTE AS FC ON FC.fvendno = SC.fcvendno
LEFT OUTER JOIN
apvend AS ap ON SC.fcvendno = ap.fvendno
WHERE
(SC.fdincidate BETWEEN DATEADD(YEAR, DATEDIFF(YEAR, 0, GETDATE()) - 1, 0)
AND DATEADD(MONTH, DATEDIFF(MONTH, -1, GETDATE()) -1, -1))
GROUP BY
FC.[Receiver Qty], ap.fcompany, SC.FPROBLEM
ORDER BY
ap.fcompany, SC.FPROBLEM
我想要的结果
fcompany Receiver Qty Number oF issues per vendor fProblem
A 139 1 Defective Material
A 3 Wrong Material
A 3 Shortage
B 597 2 Defective Material
B 2 Wrong Material
C 16 3 Shortage
D 13 3 Shortage
E 400 25 Shortage
F 10 1 Defective Material
G 90 1 Shortage
我得到的结果
fcompany Receiver Qty Number or issues per vendor fProblem
---------------------------------------------------------------
A 139 1 Defective Material
A 139 3 Wrong Material
A 139 3 Shortage
B 597 2 Defective Material
B 597 2 Wrong Material
C 16 3 Shortage
D 13 3 Shortage
E 400 25 Shortage
F 10 1 Defective Material
G 90 1 Shortage
答案 0 :(得分：1)
使用 row_number() 将[Reciver Qty]的重复实例设置为null：
, thirdcte as (
select
ap.fcompany
, fc.[Reciver Qty]
, [Number of issues per vendor] = count(sc.fkey_id)
, fproblem = case sc.fproblem when 'Wrong Quantity' then 'Shortage'
when 'Defective Material' then 'Defective Material'
when 'Wrong Material' then 'Wrong Material'
end
, rn = row_number() over (
partition by ap.fcompany
order by (case sc.fproblem when 'Wrong Quantity' then 'Shortage'
when 'Defective Material' then 'Defective Material'
when 'Wrong Material' then 'Wrong Material'
end)
)
from Secondcte as sc
left outer join Firstcte as fc on fc.fvendno = sc.fcvendno
left outer join apvend as ap on sc.fcvendno = ap.fvendno
where (sc.fdincidate between dateadd(year, datediff(year, 0, getdate()) - 1, 0)
and dateadd(month, datediff(month, - 1, getdate()) - 1, - 1))
group by fc.[Reciver Qty]
, ap.fcompany
, sc.fproblem
)
select
fcompany
, [Reciver Qty] = case when rn = 1 then [Reciver Qty] else null end
, [Number of issues per vendor]
, fproblem
from thirdcte
order by fcompany, fproblem
注意：请注意 between Bad habits to kick : mis-handling date / range queries - Aaron Bertrand - 2009-10-16
答案 1 :(得分：0)
如果我正确阅读了您的问题，您希望在前一条记录具有相同值时清空Reciver。您可以使用LAG() - 函数：
WITH result AS (
SELECT ap.fcompany, FC.[Reciver Qty], COUNT(SC.FKey_ID) AS 'Number of issues per vendor'
, CASE sc.FPROBLEM WHEN 'Wrong Quantity' THEN 'Shortage'
WHEN 'Defective Material' THEN 'Defective Material'
WHEN 'Wrong Material' THEN 'Wrong Material'
END AS fproblem
FROM SecondCTE AS SC left OUTER JOIN FirstCTE AS FC
ON FC.fvendno = SC.fcvendno LEFT OUTER JOIN
apvend AS ap
ON SC.fcvendno = ap.fvendno
WHERE (SC.fdincidate BETWEEN DATEADD(YEAR, DATEDIFF(YEAR, 0, GETDATE()) - 1, 0) AND DATEADD(MONTH, DATEDIFF(MONTH, - 1, GETDATE()) - 1, - 1))
GROUP BY FC.[Reciver Qty]
, ap.fcompany
, SC.FPROBLEM
)
SELECT fcompany
, ISNULL(NULLIF(Reciver, LAG(Reciver,1,0) OVER (ORDER BY fcompany, FPROBLEM)),'') AS Reciver
, [Qty Number of issues per vendor]
, fProblem
from result
ORDER BY fcompany
, FPROBLEM