合并两个具有不同 WHERE 子句的 SELECT 查询

Question

我有一张服务表。我需要合并两个 SELECT 查询。两者都有不同的 where 子句。例如

SELECT 
  U_REGN as 'Region', 
  COUNT(callID) as 'OpenServices',
  SUM(CASE WHEN descrption LIKE '%DFC%' THEN 1 ELSE 0 END) 'DFC'
FROM OSCL
WHERE     
  ([status] = - 3) 
GROUP BY 
  U_REGN
ORDER BY 
  'OpenServices' desc

这给了我结果

Region    | OpenServices | DFC
Karaci    | 14           | 4
Lahore    | 13           | 3
Islamabad | 10           | 4

我还有一个疑问

SELECT 
  U_REGN as 'Region', 
  COUNT(callID) as 'ClosedYesterday'
FROM OSCL
WHERE 
  DATEDIFF(day, closeDate, GETDATE()) = 1
GROUP BY 
  U_REGN
ORDER BY 
  'ClosedYesterday' desc

它给了我结果

Region    | ClosedServices
Karachi   | 8
Lahore    | 7
Islamabad | 4

我需要合并两个结果，并在 DFC 列旁边显示 ClosedServices。

Answer 1

将您当前的两个查询的结果集视为表并加入它们：

select
    FirstSet.Region,
    FirstSet.OpenServices,
    FirstSet.DFC,
    SecondSet.ClosedYesterday
from 
(
    SELECT U_REGN as 'Region', COUNT(callID) as 'OpenServices',
    SUM(CASE WHEN descrption LIKE '%DFC%' THEN 1 ELSE 0 END) 'DFC'
    FROM OSCL 
    WHERE ([status] = - 3) 
    GROUP BY U_REGN 
    --ORDER BY 'OpenServices' desc
) as FirstSet
inner join
(
    SELECT U_REGN as 'Region', 
    COUNT(callID) as 'ClosedYesterday'
    FROM OSCL
    WHERE DATEDIFF(day, closeDate, GETDATE()) = 1
    GROUP BY U_REGN
    --ORDER BY 'ClosedYesterday' desc
) as SecondSet
on FirstSet.Region = SecondSet.Region
order by FirstSet.Region

这不是我写过的最漂亮的 SQL，但希望您能看到它是如何工作的并了解如何维护它。

我怀疑性能更好的查询是SELECT来自 OSCL的单个查询，按 U_REGN 分组，三个计数器中的每一个作为单独的SUM(CASE ...)语句，类似于您目前为 DFC 所做的工作。这最多是单表扫描，具体取决于您的索引和架构。

Answer 2

建立迈克尔的建议：

SELECT
    U_REGN AS 'Region',
    SUM(CASE WHEN [status] = -3 THEN 1 ELSE 0 END) AS 'OpenServices',
    SUM(CASE WHEN [status] = -3 AND [description] LIKE '%DFC%' THEN 1 ELSE 0 END) AS 'DFC',
    SUM(CASE WHEN DATEDIFF(day, closeDate, GETDATE()) = 1 THEN 1 ELSE 0 END) AS 'ClosedYesterday'
FROM
    OSCL
GROUP BY 
    U_REGN
ORDER BY
    'OpenServices' desc