Question

我正在进行查询，我在分区内使用行号，我想使用CONCAT组合多个字段来构建全名。当我只是将字段和+一起添加为All Good。当我尝试使用CONCAT函数时 - 我收到一条错误消息。我不太明白为什么 - 有人可以告诉我是否允许在汇总中使用它？

以下是可以正常使用的代码：

USE [AdventureWorks2012]
SELECT 
    count([BusinessEntityID])as NumPeople
    ,[PersonType]
    ,max(case when rnum = 1 then [BusinessEntityID] end) as '1st ID'
    ,max(case when rnum = 2 then [BusinessEntityID] end) as '2nd ID'
    ,max(case when rnum = 1 then [FirstName]+' '+[LastName] end ) as '1st FullName'
    ,max(case when rnum = 2 then [FirstName]+' '+[LastName] end ) as '2nd FullName'

from 
    (   
        Select *, row_number() over (partition by [PersonType]  order by [BusinessEntityID]) as rnum
        FROM
        [Person].[Person] 
    ) x
group by [PersonType]

及其输出：

+-----------+------------+--------+--------+----------------+------------------+
| NumPeople | PersonType | 1st ID | 2nd ID | 1st FullName   | 2nd FullName     |
+-----------+------------+--------+--------+----------------+------------------+
| 273       | EM         | 1      | 2      | Ken Sánchez    | Terri Duffy      |
| 289       | GC         | 2091   | 2092   | David Ortiz    | Qiang Wang       |
| 18484     | IN         | 1699   | 1700   | David Robinett | Rebecca Robinson |
| 753       | SC         | 291    | 293    | Gustavo Achong | Catherine Abel   |
| 17        | SP         | 274    | 275    | Stephen Jiang  | Michael Blythe   |
| 156       | VC         | 1491   | 1493   | Paula Moberly  | Suchitra Mohan   |
+-----------+------------+--------+--------+----------------+------------------+

这是给出错误的代码：

USE [AdventureWorks2012]
SELECT 
    count([BusinessEntityID])as NumPeople
    ,[PersonType]
    ,max(case when rnum = 1 then [BusinessEntityID] end) as '1st ID'
    ,max(case when rnum = 2 then [BusinessEntityID] end) as '2nd ID'
    ,max(case when rnum = 1 then (concat(p.[FirstName], stuff(p.[MiddleName],1,0,' '), stuff(p.[LastName],1,0,' ') end ) as '1st FullName'
    ,max(case when rnum = 2 then (concat(p.[FirstName], stuff(p.[MiddleName],1,0,' '), stuff(p.[LastName],1,0,' ')end ) as '2nd FullName'
from 
    (   
        Select *, row_number() over (partition by [PersonType]  order by [BusinessEntityID]) as rnum
        FROM
        [Person].[Person] 
    ) x
group by [PersonType]

这是错误：

Msg 156, Level 15, State 1, Line 9
Incorrect syntax near the keyword 'end'.
Msg 102, Level 15, State 1, Line 17
Incorrect syntax near 'x'.

我确信这只是Microsoft SQL Server不允许的内容 - 但我想知道它无法完成的内容 - 这样我就可以确保在我完成时避免使用它需要。或者，如果有某种方法可以做到这一点，那么它也会很棒......

Answer 1

我认为括号是这里唯一的问题。使用concat的两行与...) end) as '...类似，它们应该像...))) end) as '...一样读取。完整查询如下。

USE [AdventureWorks2012]
SELECT 
    count([BusinessEntityID])as NumPeople
    ,[PersonType]
    ,max(case when rnum = 1 then [BusinessEntityID] end) as '1st ID'
    ,max(case when rnum = 2 then [BusinessEntityID] end) as '2nd ID'
    ,max(case when rnum = 1 then (concat(p.[FirstName], stuff(p.[MiddleName],1,0,' '), stuff(p.[LastName],1,0,' '))) end)  as '1st FullName'
    ,max(case when rnum = 2 then (concat(p.[FirstName], stuff(p.[MiddleName],1,0,' '), stuff(p.[LastName],1,0,' '))) end) as '2nd FullName'
from 
    (   
        Select *, row_number() over (partition by [PersonType]  order by [BusinessEntityID]) as rnum
        FROM
        [Person].[Person] 
    ) x
group by [PersonType]

Aggregate中的Concat给出错误消息

1 个答案: