Question

我有这些ID，我总是以这种格式获取它们：

538693,538693,541616,541616,541616,541620,541116,541116 ，538639,538639,538720,538720,538720,541608,541608， 541608,538675,538675,538675,538675

然后我查询哪些返回我的一些，例如返回我最后3个ID。现在我需要从第一个记录集中选择剩余的ID而不是最后一个。

这只是示例，我有超过200行的这个ID ..

Answer 1

这是一个可能会给你一个想法的例子。看看。

SQL> with
  2  -- a long list of (duplicate?) IDs
  3  all_ids (col) as
  4    (select '538693 , 538693 , 541616 , 541616 , 541616 , 541620 , 541116 , 541116 , 538639 , 538639' from dual
  5    ),
  6  -- that's what you get from your query
  7  do_query (col) as
  8    (select '541116 , 538639 , 538639' from dual
  9    ),
 10  -- split ALL_IDs to rows
 11  all_1 as
 12    (select regexp_substr(col, '[^ , ]+', 1, level) col
 13     from all_ids
 14     connect by level <= regexp_count(col, ',') + 1
 15    ),
 16  -- split IDs from DO_QUERY to rows as well
 17  do_1 as
 18    (select regexp_substr(col, '[^ , ]+', 1, level) col
 19     from do_query
 20     connect by level <= regexp_count(col, ',') + 1
 21    ),
 22  -- MINUS set operator will return IDs from ALL_IDS that aren't in DO_QUERY IDs
 23  minus_me as
 24    (select col From all_1
 25     minus
 26     select col from do_1
 27    )
 28  -- finally, compose the remaining **unique** IDs back
 29  select listagg(col, ' , ') within group (order by col) result
 30  from minus_me;

RESULT
--------------------------------------------------------------------------------
538693 , 541616 , 541620

SQL>

Answer 2

使用xmltable代码

更简单

SELECT *
FROM (
    SELECT to_number(column_value) AS n
    FROM XMLTABLE('538693 , 538693 , 541616 , 541616 , 541616 , 541620 , 541116 ,
                   541116 , 538639 , 538639 , 538720 , 538720 , 538720 , 541608 , 
                   541608 , 541608 , 538675 , 538675 , 538675 , 538675')
    ) o
WHERE n NOT IN (
     select to_number(column_value) 
  FROM xmltable(
        '538675
        ,538675
        ,538675')
        );

Demo

Answer 3

我没有oracle试用，但可能会有效。

EG。如果您有ID 1,2,3 ... 10，则下面的查询将返回1,2,3，... 7

SELECT ID FROM TableName WHERE ID NOT IN
(SELECT ID FROM TableName WHERE ROWNUM < 3)

不确定语法，只需尝试逻辑：）

Answer 4

这是另一种可能的解决方案w / out将输出转换为字符串。看看你的情况是否可以接受。这会将您的行转换为列。然后很容易消除最后3行：

select distinct column_value from table(sys.odcinumberlist(1,1,2,3,3,4,4,5));

oracle选择第二个选择中不存在的位置

4 个答案: