Question

在表格中，我有4列，ID和V1，V2和V3。如何选择Value1和Value2重复的行的ID？例如：

ID | V1 | V2 | V3  
1    A    B    X
2    A    C    Y
3    A    B    Z  
4    A    B    W
5    F    B    K
6    A    B    N
7    H    B    N
8    O    C    N

我期待结果：1,3,4和6

我已经尝试了什么

select MovimentacaoContaId from MovimentacaoConta
group by NumeroDocumento,PessoaId,DataEmissaoReal,ValorTotalReal
having count(NumeroDocumento)>1 and count(PessoaId)>1 and count(DataEmissaoReal)>1 and count(ValorTotalReal)>1

select MovimentacaoContaId from MovimentacaoConta
group by NumeroDocumento,PessoaId,DataEmissaoReal,ValorTotalReal
having count(NumeroDocumento,PessoaId,DataEmissaoReal,ValorTotalReal)>1 //obvious error...

select distinct MovimentacaoContaId from MovimentacaoConta
group by NumeroDocumento,PessoaId,DataEmissaoReal,ValorTotalReal
having count(NumeroDocumento)>1 and count(PessoaId)>1 and count(DataEmissaoReal)>1 and count(ValorTotalReal)>1

PS：我想要这些特定的列，因为V1和V2应该是一个compisite PK，但是之前的开发人员没有这样做，并且在它之上插入了几个错误，而不是更新记录，插入... < / p>

Answer 1

您需要以某种方式自我加入，因为您必须将聚合与每一行相关联。

SELECT
    ID
FROM
    MovimentacaoConta mc
INNER JOIN (
    SELECT
        V1,
        V2,
        COUNT(*) AS row_count
    FROM
        MovimentacaoConta
    GROUP BY
        V1,
        V2
) mc0
ON
    mc.v1 = mc0.v1
AND
    mc.v2 = mc0.v2
WHERE
    mc0.row_count > 1

Answer 2

select table.* 
  from table 
  join (select V1, V2 
          from table
         group by V1, V2 
        having count(*) > 1) V1V2
    on table.V1 = V1V2.V1 
   and table.V2 = V1V2.V2

Answer 3

这是我最喜欢的方式：

;with cte as 
(
    select ROW_NUMBER() over (partition by v1,v2 order by v1,v2) as rn, *
    from your_tbl
)
select * from cte where rn > 1

从重复值中选择PK

3 个答案: