PL/SQL 查找并让行明智的重复

Question

我有一个有 10 列的表。看起来像这样

NULL 1 1 NULL NULL NULL NULL NULL 2 NULL
1 NULL NULL 2 NULL 1 NULL NULL 2 NULL
NULL NULL NULL NULL NULL NULL NULL NULL NULL NULL
NULL NULL NULL NULL NULL NULL NULL NULL 2 NULL

我需要一个返回以下内容的 SQL 查询：（不同的行）

1 2
1 2
NULL
2

Answer 1

您可以按如下方式使用hiearchy query：

SELECT LISTAGG(E, ' ') WITHIN GROUP( ORDER BY E )
FROM
    ( SELECT DISTINCT
            ROWID AS RID,
            DECODE(COLUMN_VALUE, 1, COL1, 2, COL2, 3, COL3, 4, COL4 ..., 10, COL10) AS E
        FROM YOUR_TABLE
             CROSS JOIN TABLE ( CAST(MULTISET(
                SELECT LEVEL
                  FROM DUAL
                CONNECT BY LEVEL <= 10 ) AS SYS.ODCINUMBERLIST) )
    )
GROUP BY RID;

Answer 2

您可以将UNPIVOT 列转换为行（并保留NULL 值），然后过滤以保留DISTINCT 行，然后将行聚合成一个字符串，忽略NULL 值，除非它们都是@ 987654326@:

SELECT LISTAGG( value, ' ' ) WITHIN GROUP ( ORDER BY value ) AS vals
FROM   (
  SELECT DISTINCT
         rn,
         value,
         COUNT(value) OVER ( PARTITION BY rn ) AS num_values
  FROM   (
    SELECT ROWNUM AS rn,
           t.*
    FROM   table_name t
  )
  UNPIVOT INCLUDE NULLS (
    value FOR key IN ( col1, col2, col3, col4, col5, col6, col7, col8, col9, col10 )
  )
)
WHERE num_values = 0
OR    value IS NOT NULL
GROUP BY rn

因此，对于您的示例数据：

CREATE TABLE table_name (
  col1 NUMBER,
  col2 NUMBER,
  col3 NUMBER,
  col4 NUMBER,
  col5 NUMBER,
  col6 NUMBER,
  col7 NUMBER,
  col8 NUMBER,
  col9 NUMBER,
  col10 NUMBER
);

INSERT INTO table_name
SELECT NULL,    1,    1, NULL, NULL, NULL, NULL, NULL,    2, NULL FROM DUAL UNION ALL
SELECT    1, NULL, NULL,    2, NULL,    1, NULL, NULL,    2, NULL FROM DUAL UNION ALL
SELECT NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL FROM DUAL UNION ALL
SELECT NULL, NULL, NULL, NULL, NULL, NULL, NULL, NULL,    2, NULL FROM DUAL;

这个输出：

|瓦尔斯 | | :--- | | 1 2 | | 1 2 | | 空 | | 2 |

db小提琴here

Answer 3

如果这有助于尝试将列取消旋转为 1，请尝试以下操作，以消除空值，在取消旋转然后通过分组聚合之前为每个先前的选择保留 rowid

    Select op from (Select rn, 
     listagg(distinct col, 
       ',') within group (order by 
         col) as  op from
   (  Select *, row_number() over(
    Partition by col1, col2.. Col10 order by
   ..) rn
     from table
    UnPIVOT 
  (COL, num) IN   ( (COL1, COL2,... 
     Col10), rn)) 

   ) where col is not null group by 
  rn)