Question

我有这个SQL查询，可以检测到[Type]重复出现。

select Type, count(1) as 'Occurences' from [dbo].[Policy]
 where PolicyDescription ='ABC' 
group by Type
 HAVING COUNT(1)>1

但同一张表[Policy]中也有一列[Id]，而我上面的查询没有基于[Id]进行区分。

我希望上述查询仅考虑[Id]也不同的结果。

我应该实施哪些更改？

样本数据

DECLARE @Policy AS TABLE
(
    ID int,
    PolicyDescription char(3),
    Type int
)

INSERT INTO @Policy(ID, PolicyDescription, Type) VALUES
(1, 'ABC', 123),
(1, 'ABC', 123),
(2, 'ABC', 456),
(3, 'ABC', 567),
(4, 'ABC', 567),
(3, 'ABC', 567)

当前结果

 Type | Occurences
------|------------
   123|           2
   567|           3

所需结果

 Type | Occurences
------|------------
   567|           2

Answer 1

您可以将COUNT与DISTINCT结合使用。

SELECT Type, COUNT(DISTINCT Id) AS 'Occurences'
FROM Policy
WHERE PolicyDescription = 'ABC'
GROUP BY Type
HAVING COUNT(DISTINCT Id) > 1

Answer 2

您只需要更改：

from policy

收件人：

from (select distinct * from policy)

像这样：

select t.type, count(1) as 'Occurences' 
from (select distinct * from policy) t
where t.policydescription ='ABC' 
group by t.type
having count(1) > 1

请参见demo

在这种情况下，如何考虑内部差异来检测重复项？

2 个答案: