在Oracle SQL Developer中在PL / SQL中更新多行时忽略重复项

Question

I'm trying to update all the rows of a table (A) when the ids with another table (B) match. The problem is that i'm getting the following error :

unable to get a stable set of rows in the source tables

I've made my research and I know that the cause may be duplicated rows in one of the tables. Only table B has duplicated rows, i've tried in may ways to ignore them with some queries, but unsuccessfully.

 merge into A x 
    using B y
    on (x.id= y.id)
    when matched then
      UPDATE SET 
      x.apples= y.apples,
      x.bananas= y.bananas,
      x.grapes= y.grapes;

Can someone help?

Thanks in advance

Answer 1

Duplicates can be on the side where you are doing the updates. But you can not have a duplicate on the source side.
Think about it they sql engine does not know which of the multiple records to use on the update. You need to either fix the duplicate issue. Or do some sort of max or min to get a unique data set to be used in the update.

Answer 2

I was able to solve this problem, here's the solution i made:

 merge into A x 
        using (select distinct id from B) y
        on (x.id= y.id)
        when matched then
          UPDATE SET 
          x.apples= (select apples from B where id = 
(select distinct id from B where id = x.id) and rownum = '1'),
          x.bananas= (select bananas from B where id = 
(select distinct id from B where id = x.id) and rownum = '1'),
          x.grapes= (select grapes from B where id = 
(select distinct id from B where id = x.id ) and rownum = '1');