Question

我创建了一个查询以删除表中的重复行。我尝试这样做，但是输出是“太多值”

DELETE FROM Employeetest 
WHERE employeeid IN (SELECT 
                         employeetest.*, 
                         ROW_NUMBER() OVER (PARTITION BY employeeid ORDER BY employeeid) AS rownumber 
                     FROM
                         employeetest 
                     HAVING
                         (rownumber > 1));

我也是

SELECT
    employeetest.*, 
    ROW_NUMBER() OVER (PARTITION BY employeeid ORDER BY employeeid) AS rownumber 
FROM
    employeetest

然后

DELETE * FROM employeetest;

它不起作用

Answer 1

您标记为plsql的我知道您的DB是Oracle。因此，您可以按以下方式使用rowid pseudocolumn：

delete Employeetest t1
where rowid <
(
select max(rowid)
  from Employeetest t2 
 where t2.employeeid = t1.employeeid 
);

如果目标是删除employeeid个值的所有重复项。

P.S。无法以这种方式Delete * from employeetest where ...进行删除，但是可以使用Delete from employeetest where ...或Delete employeetest where ...。

Answer 2

我总是这样使用：

delete employeetest
where  rowid in
       ( select lag(rowid) over (partition by employeeid order by null) 
         from employeetest )

Answer 3

此逻辑通常也写为：

delete Employeetest 
where rowid in (select max(rowid)
                from Employeetest e2
                group by e2.employeeid
               );

在SQL中复制行会输出太多值

3 个答案: