Question

这是screnario，我有2个表table1和table2，我需要使用table2更新table1，因为table2中有很多重复，但需要使用table2中的where子句选择一个

Table1(unique table)
    Email_Address|Country|Region|Source|PP|PS
    a@a.com IN AP SS ES PP
    b@g.com US AM ES ST PO

Table2(duplicate values)
    Email_Address|Country|Region|Source|PP|PS
    a@a.com IN AP SS ES PP
    a@a.com ES AP SS ES SST
    a@a.com IN AP MS ES P
    a@a.com IN AP NS ST PP
    b@g.com US AM ES ST PO
    b@g.com IN AP NS ST PP

现在我需要从Table2获取不同的值，其中源是SS并且需要针对Table1电子邮件进行映射。

我使用过这样的东西但是没有用，

Update Table1
Set Table1.Country=Table2.Country,Table1.PP=Table2.PP
Where Email_Address In (Select Distinct Email_Address 
From Table2 
where Table2.Source = 'SS')

任何帮助都是有帮助的

Answer 1

update   t1
set      Source = 'SS'
,        PP = t2.MaxPP
from     Table1 t1
join     (
         select  Email_address
         ,       max(PP) as MaxPP -- Pick max PP from the available group
         from    Table2
         where   Source = 'SS'
         group by
                 Email_address
         ) t2
on       t2.Email_address = t1.Email_address

如果PP中有多个匹配的Table2值，则必须选择。上面的例子选择了最大值。如果那不是你想要的，请澄清你的问题。

Answer 2

对于此类更新，您可以使用from子句。您要做的是为源为“SS”的每个电子邮件地址选择一行。以下选项使用row_number()和newid()选择一个randome行：

Update t1
    Set Source = t2.Source,
        PP = t2.PP
    from table1 t1 join
         (select t2.*,
                 row_number() over (partition by email_address order by newid()) as seqnum
          from table2 t2
          where t2.Source = 'SS'
         ) t2
         on t1.email_address = t2.email_address and seqnum = 1;

Sql使用distinct和where子句从另一个表更新表值

2 个答案: