Question

我处于这样一种情况，我需要选择在两个特定列中具有相同内容的行，并在第三个列中选择不同的内容。到目前为止，我在两个类似的专栏中得到了这个：

SELECT id, Title,
COUNT(*) AS NumOccurrences
FROM Table
GROUP BY id, Title
HAVING ( COUNT(*) > 1 )

我现在需要在此查询中指定第三个不同的列。我们称之为拉尔夫。这显然不起作用：

SELECT id, Title, DISTINCT Ralph,
COUNT(*) AS NumOccurrences
FROM Table
GROUP BY id, Title
HAVING ( COUNT(*) > 1 )

那会是什么？

Answer 1

select * from (
    SELECT id, Title, COUNT(*) AS NumOccurrences
    FROM Table t
    GROUP BY id, Title
    HAVING ( COUNT(*) > 1 )
) t
cross apply (
    select distinct Ralph
    from Table
    where id = t.id and Title = t.Title
) t2

Answer 2

您可以将COUNT（*）与OVER（）子句

一起使用

 ;WITH cte AS
 (       
  SELECT id, Title, Ralph, COUNT(*) OVER (PARTITION BY id, Title) AS cnt
  FROM dbo.test11
  GROUP BY id, Title, Ralph
  ) 
  SELECT *
  FROM cte
  WHERE cnt > 1

SQLFiddle上的演示

查找具有两个相似列和一个不同列的重复项

2 个答案: