SQL Server 2005的压缩结果集返回到值

Question

一行

SQL Server 2005中

我试图压缩设置让每个工作人员一行数据的结果，我的查询是

SELECT Staff +'- '+ CONVERT(CHAR(2),Service_Date, 110) AS Staff 
,(SELECT COUNT (Procedure_) WHERE Procedure_ Like '90791%') AS '90791' 
,(SELECT COUNT (Procedure_) WHERE Procedure_ Like'90832%') AS '90832' 
,(SELECT COUNT (Procedure_) WHERE Procedure_ Like'90834%') AS '90834' 
,(SELECT COUNT (Procedure_) WHERE Procedure_ Like'90837%') AS '90837' 
,(SELECT COUNT (Procedure_) WHERE Procedure_ Like '90847%') AS '90847' 

FROM dbo.Service_History 
WHERE Program='OPCC' 
AND STATUS='Kept' 
AND TYPE = 'Client' 
AND Procedure_ IS NOT NULL 
AND Service_date BETWEEN DATEADD(mm, DATEDIFF(mm, 0,GETDATE())-3, 0) AND GETDATE() 


GROUP BY Staff +'- '+ CONVERT(CHAR(2),Service_Date, 110),Procedure_ 
ORDER BY Staff +'- '+ CONVERT(CHAR(2),Service_Date, 110),Procedure_ 

GROUP BY Staff +'- '+ CONVERT(CHAR(2),Service_Date, 110),Procedure_ 
ORDER BY Staff +'- '+ CONVERT(CHAR(2),Service_Date, 110),Procedure_ 

结果集如下所示:(每个员工一列，返回一列数据，一列全空行）。

Staff      90791 90832 90834 90837 90847 
Smith, Joe (568)- 04 15 NULL NULL NULL NULL 
Smith, Joe (568)- 04 NULL NULL 2 NULL NULL 
Smith, Joe (568)- 04 NULL NULL NULL 1 NULL 
Smith, Joe (568)- 04 NULL NULL NULL NULL NULL 

我想将上面的结果下降到一行。基本上每个工作人员每月有一行总计执行的程序＃。

喜欢的部份：

Staff      90791 90832 90834 90837 90847 
Smith, Joe (568)- 04 15 NULL 2 1 NULL 

我想要的结果可以通过导入结果到一个临时表，然后查询临时表得到的结果来实现，但是这是远远效率，我希望有是更好的解决方案。

Answer 1

您可以取每个字段的MAX（）。即：

ID a  b  c  d 
1 1  NULL NULL NULL 
1 NULL 1  NULL NULL 
1 NULL NULL 1  NULL 

SELECT ID 
    , MAX(a) AS a 
    , MAX(b) AS b 
    , MAX(c) AS c 
    , MAX(d) as d 
FROM Table 
GROUP BY ID 

结果：

ID a b c d 
1 1 1 1 NULL 

Answer 2

做这样的事情：

select staff + convert(char(2),service_date,110) staff, 
count(case when procedure like '90791%' then staff end) '90791', 
count(case when procedure like '90832%' then staff end) '90832', 
count(case when procedure like '90834%' then staff end) '90834', 
count(case when procedure like '90837%' then staff end) '90837', 
count(case when procedure like '90847%' then staff end) '90847' 

from service_history 
where program='OPCC' 
.... 
group by staff + convert(char(2),service_date,110) 

应该这样做！