not only that, but the answer they give you is WRONG!!!!!!!!!!!!!!

so what if when they counted (which is really slow), there were 500 rows. By the time the count finished, maybe there were 490, or 510 or 0 or whatever.

They should have just: select dbms_random.value into :x from dual;

it would be approximately as accurate and meaningful!!!!



begin
delete from t where a = :a and b = :b and c = :c ....whatever... ;
:x := sql%rowcount;
end;


tell them to use that block of code - that'll work in any language, all languages give you a way to find the number of rows affected by your last DML as far as I know, but you didn't tell me what language you use. This approach works in all of them though.

Or you could ask them to read the documentation and learn the API they are using, to avoid wasteful, incorrect approaches to a rather simple problem!