Hive MERGE: "when matched then update" allows invalid column names.

create table target (
  id int, val int
)
CLUSTERED BY (id) INTO 2 BUCKETS STORED AS ORC TBLPROPERTIES ("transactional"="true");

create table source2 (
  id int, val int
);
insert into source2 values (2, 25), (3, 35), (4, 45);

merge into target
using source2 sub on sub.id = target.id
when matched then update set invalid = sub.val;