Question

我有一个数据集如下：

SELECT '  1234 ' ID,NULL TAG,' AB' CODE FROM DUAL
UNION ALL 
SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
UNION ALL
SELECT '1234' ID,NULL TAG,' AB' CODE FROM DUAL)

我试图通过以下查询获得重复项。输出应该是第1行和第3行，但我只得到一行作为1或3的修剪输出。

SELECT TRIM(ID),TRIM(TAG),TRIM(CODE) FROM 
(SELECT '  1234 ' ID,NULL TAG,' AB' CODE FROM DUAL
UNION ALL 
SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
UNION ALL
SELECT '1234' ID,NULL TAG,' AB' CODE FROM DUAL)
WHERE (TRIM(ID),TRIM(TAG),TRIM(CODE)) IN 
(
SELECT TRIM(ID),TRIM(TAG),TRIM(CODE) FROM
(SELECT '  1234 ' ID,NULL TAG,' AB' CODE FROM DUAL
UNION ALL 
SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
UNION ALL
SELECT '1234' ID,NULL TAG,' AB' CODE FROM DUAL)
GROUP BY TRIM(ID),TRIM(TAG),TRIM(CODE)
HAVING COUNT(*) >1
)

Answer 1

我刚跑了这个，它返回了第1行和第3行：

SELECT TRIM(ID),TRIM(TAG),TRIM(CODE) 
FROM 
(
    SELECT '  1234 ' ID,NULL TAG,' AB' CODE FROM DUAL
    UNION ALL 
    SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
    UNION ALL
    SELECT '1234' ID,NULL TAG,' AB' CODE FROM DUAL
)
WHERE TRIM(ID) IN 
(
    SELECT TRIM(ID)
    FROM
    (
        SELECT '  1234 ' ID,NULL TAG,' AB' CODE FROM DUAL
        UNION ALL 
        SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
        UNION ALL
        SELECT '1234' ID,NULL TAG,' AB' CODE FROM DUAL
    )
   GROUP BY TRIM(ID),TRIM(TAG),TRIM(CODE)
   HAVING COUNT(*) >1
)

我将您的WHERE更改为仅引用TRIM(ID)而不是所有3个值。

编辑＃1 ，部分问题在于您要将null与null进行比较，而这是您无法做到的。因此，您可以对列进行null检查，如果是null，则替换它。我用null包裹了nvl(null, 'na')列，因此它有一个要比较的值：

SELECT TRIM(ID) id,TRIM(TAG) tag,TRIM(CODE) code 
FROM 
(
    SELECT '  1234 ' ID, nvl(null, 'na') TAG,' AB' CODE FROM DUAL
    UNION ALL 
    SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
    UNION ALL
    SELECT '  1234 ' ID,nvl(null, 'na') TAG,' AB' CODE FROM DUAL
    UNION ALL 
    SELECT '  1234 ' ID,nvl(null, 'na') TAG,' AC' CODE FROM DUAL
)
WHERE (TRIM(ID),TRIM(TAG),TRIM(CODE)) IN 
(
    SELECT TRIM(ID),TRIM(TAG),TRIM(CODE) 
    FROM
    (
        SELECT '  1234 ' ID,nvl(null, 'na') TAG,' AB' CODE FROM DUAL
        UNION ALL 
        SELECT '453' ID,'GEN' TAG,'AB' CODE FROM DUAL
        UNION ALL
        SELECT '  1234 ' ID,nvl(null, 'na') TAG,' AB' CODE FROM DUAL
        UNION ALL 
        SELECT '  1234 ' ID,nvl(null, 'na') TAG,' AC' CODE FROM DUAL
    )
   GROUP BY TRIM(ID), TRIM(CODE), TRIM(TAG)
   HAVING COUNT(*) >1
)

请参阅SQL Fiddle with Demo

带空格的重复行：

1 个答案: