Question

我有一个问题，我整个星期都在研究。我仍然是新人，我很感激迄今为止所获得的帮助。

此查询从3个表中提取数据，将其转储到视图中，然后我有第二个查询来按照老板想要的方式格式化数据。问题是，这必须在SAP B1上运行，它似乎不支持视图。所以我需要以一种在一个查询中完成所有操作的方式来编写它。

这是第一个查询：

SELECT
t2.cardcode as 'BP_Code',
--t0.Recontact as 'Date',
t2.CardName as 'BP_Name',
SubString(T3.Name,1,2) as 'Salesman',
replace(T0.Street,',',' ') as 'Street_Address',
T0.City, 
T0.State, 
SUM(CASE WHEN YEAR(t0.recontact) = 2011 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2011',
SUM(CASE WHEN YEAR(t0.recontact) = 2012 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2012',
SUM(CASE WHEN YEAR(t0.recontact) = 2013 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2013',
convert(decimal(10,2),SUM(t0.U_sold)) as 'Total_Sold'
From 
OCLG t0 -- OCLG is Activities
inner join OCRD t2 -- OCRD is Customer Definitions
on T0.cardcode like t2.cardcode
inner join OCLS t3 -- OCLS is Activity Definitions
on T0.CntctSbjct=T3.Code 
where 
t0.U_sold > 0
and T0.CntctSbjct=T3.Code 
and T0.CardCode=T2.CardCode
and T0.CntctType='3' 
and t2.CardCode = 'a239' --This was added to simplify output
and T0.Recontact >= Convert(date, '2011-01-01' ) and T0.Recontact <= Convert(date,    '2013-12-31' )
group by t2.cardcode, t0.city, t0.state, t0.street, t2.CardName, T3.Name, t0.Recontact
order by t2.CardCode

现在，这会吐出一个看起来像这样的表：

BP_Code    BP_Name     Salesman     Street_Address     City     State     Year2011     Year2012     Year 2013     Total_Sold
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    8993.84 0.00    0.00    8993.84
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    16474.54    0.00    0.00    16474.54
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    0.00    7170.79 0.00    7170.79
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    0.00    9207.73 0.00    9207.73
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    0.00    0.00    6960.20 6960.20
A239    Buddy's 01  123 WASHINGTON WASHINGTON     MO    0.00    0.00    6787.73 6787.73

现在我将该输出粘贴在一个名为CallReport的视图中，然后在其上运行此查询：

;WITH x AS 
(
  SELECT BP_Code, BP_Name, SalesMan, Street_Address, 
    s = SUM(Total_Sold) OVER (PARTITION BY BP_Code),
    a = SUM(Year2011) OVER (PARTITION BY BP_Code),
    b = SUM(Year2012) OVER (PARTITION BY BP_Code),
    c = SUM(Year2013) OVER (PARTITION BY BP_Code),
    r = ROW_NUMBER() OVER (PARTITION BY BP_Code ORDER BY Date DESC)
  FROM CallReport
)
SELECT BP_Code, BP_Name, Salesman, Street_Address, Year2011 = a, Year2012 = b, Year2013 = c, Total_Sold = s

  FROM x
  WHERE r = 1 and Salesman = 01;

然后我得到了我想要的输出：

 BP_Code    BP_Name Salesman    Street_Address  Year2011    Year2012    Year2013    Total_Sold
 A239   Buddy's 01  123 WASHINGTON 25468.38 16378.52    13747.93    106804.83

看看问题是什么？我有数百个BP_Codes，我只是缩小了这个例子的A239。我需要每个BP_code只在一行上，每年的总销售数据和总计。我知道有一种方法可以更容易地做到这一点，但我仍然是游戏新手。有任何想法吗？这是在MSSQL 2008 R2

上

Answer 1

如果我理解这一点你正在尝试修改视图以获得所需的输出，下面是它的代码

SELECT
t2.cardcode as 'BP_Code',
--t0.Recontact as 'Date',
t2.CardName as 'BP_Name',
SubString(T3.Name,1,2) as 'Salesman',
replace(T0.Street,',',' ') as 'Street_Address',
T0.City, 
T0.State, 
SUM(CASE WHEN YEAR(t0.recontact) = 2011 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2011',
SUM(CASE WHEN YEAR(t0.recontact) = 2012 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2012',
SUM(CASE WHEN YEAR(t0.recontact) = 2013 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2013',
convert(decimal(10,2),SUM(t0.U_sold)) as 'Total_Sold'
From 
OCLG t0 -- OCLG is Activities
inner join OCRD t2 -- OCRD is Customer Definitions
on T0.cardcode like t2.cardcode
inner join OCLS t3 -- OCLS is Activity Definitions
on T0.CntctSbjct=T3.Code 
where 
t0.U_sold > 0
and T0.CntctSbjct=T3.Code 
and T0.CardCode=T2.CardCode
and T0.CntctType='3' 
and t2.CardCode = 'a239' --This was added to simplify output
and T0.Recontact >= Convert(date, '2011-01-01' ) and T0.Recontact <= Convert(date,    '2013-12-31' )
group by t2.cardcode, t0.city, t0.state, replace(T0.Street,',',' '), t2.CardName, SubString(T3.Name,1,2)
order by t2.CardCode

我所做的只是纠正你的groupby子句，你得到多个记录，因为t0.Recontact in group by。

如果在t0.city，t0.state中有多个值，则替换（T0.Street，'，'，''），t2.CardName，SubString（T3.Name，1,2）列为BP_Code和你想选择这些列中的一个值然后使用下面的代码

SELECT
t2.cardcode as 'BP_Code',
--t0.Recontact as 'Date',
max(t2.CardName) as 'BP_Name',
max(SubString(T3.Name,1,2)) as 'Salesman',
max(replace(T0.Street,',',' ')) as 'Street_Address',
max(T0.City) City, 
max(T0.State) State, 
SUM(CASE WHEN YEAR(t0.recontact) = 2011 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2011',
SUM(CASE WHEN YEAR(t0.recontact) = 2012 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2012',
SUM(CASE WHEN YEAR(t0.recontact) = 2013 THEN convert(decimal(10,2),t0.U_sold) ELSE 0 END) AS 'Year2013',
convert(decimal(10,2),SUM(t0.U_sold)) as 'Total_Sold'
From 
OCLG t0 -- OCLG is Activities
inner join OCRD t2 -- OCRD is Customer Definitions
on T0.cardcode like t2.cardcode
inner join OCLS t3 -- OCLS is Activity Definitions
on T0.CntctSbjct=T3.Code 
where 
t0.U_sold > 0
and T0.CntctSbjct=T3.Code 
and T0.CardCode=T2.CardCode
and T0.CntctType='3' 
and t2.CardCode = 'a239' --This was added to simplify output
and T0.Recontact >= Convert(date, '2011-01-01' ) and T0.Recontact <= Convert(date,    '2013-12-31' )
group by t2.cardcode
order by t2.CardCode

无法消除SQL视图

1 个答案: