子查询的列过多

Question

I have two tables with same structure: tmp_grn and grn .
I have to delete rows from table tmp_grn which already exists in table grn .
Problem is I don't have a unique or primary key but I can determine a unique row with the combination of two columns. Let's say column names are grn_code and item_skucode .

My query:

DELETE FROM tmp_grn 
  WHERE grn_code AND item_skucode IN 
    (SELECT grn_code , item_skucode FROM grn);

I am getting this error:

ERROR: subquery has too many columns

What should be the right way to do this?

Answer 1

If you want to combine two columns, you need to put them into parenthesis:

DELETE FROM tmp_grn 
  WHERE (grn_code, item_skucode) IN (SELECT grn_code, item_skucode 
                                     FROM grn);

But suslov's answer using an exists is most probably faster - you need to check the execution plan to verify that.

Answer 2

You can use exists ( if you want to check the pair of values ):

delete from tmp_grn t
where exists ( select * 
               from grn
               where grn_code = t.grn_code
                 and item_skucode = t.item_skucode);

Answer 3

delete * from tmp_grn intersect select * from grn