Question

我在那篇文章之前经常搜索过，这是我的问题：

表：

table1 
table2
table3
table4
table5
table6
table7
table8
table9
table10

表1有：

id (primary key), cola, colb, colc

在表1中，可乐重复了 cola 列值：

abc
abc

在表2中：

wid (it's id from table1), cola, colb, colc

表3：

wid (it's id from table1), cola, colb, colc, cold

等等，所有表格内容都是wid。

我要做的是删除重复行依赖于来自table1的列 cola 并仅保留1.然后从表table2，table3，table4，table5中删除行，...与table1中已删除的重复行的id相同。

Answer 1

让我们看看能否提供帮助。

要查找欺骗，您可以使用分区并添加一个rownumber：

WITH FindDupes AS (
  SELECT[[id],  
     row_number() OVER(PARTITION BY cola ORDER BY id) AS [rn]
  FROM Table1
)
SELECT * FROM FindDupes WHERE [rn] > 1

通过这种方式，您可以找到所有重复的行ID。

您还可以展开查询以删除所有表格中的ID。

WITH DeleteDupes AS (
  SELECT[[id],  
     row_number() OVER(PARTITION BY cola ORDER BY id) AS [rn]
  FROM Table1
)
DELETE t1, t2, t3, t4, t5, t6, t7, t8, t9, t10 FROM
     table1 as t1
     INNER JOIN table2 as t2 on t1.id = t2.wid
     INNER JOIN table3 as t3 on t1.id = t3.wid
     INNER JOIN table4 as t4 on t1.id = t4.wid
     INNER JOIN table5 as t5 on t1.id = t5.wid
     INNER JOIN table6 as t6 on t1.id = t6.wid
     INNER JOIN table7 as t7 on t1.id = t7.wid
     INNER JOIN table8 as t8 on t1.id = t8.wid
     INNER JOIN table9 as t9 on t1.id = t9.wid
     INNER JOIN table10 as t10 on t1.id = t10.wid
WHERE t1.id in (SELECT id FROM DeleteDupes WHERE [rn] > 1)

这会留下一行重复项，并从所有10个表中删除所有其他项。您没有指定应保留哪些重复项。如果要修改它，可以通过更改ORDER BY子句来完成此操作。要保留的应该是首先导致ORDER BY子句。

sql从具有重复列值的多表中删除行

1 个答案: