Why does multiplication by a constant cause an overflow in PL/SQL?

Question

Oracle

Here are three seemingly identical expressions, but on the latest Oracle 12c PL/SQL crashes with the error "ORA-01426: numeric overflow. Evaluation of an value expression causes an overflow/underflow". Why is that?

declare
    n number(38,0);
    mul number(38,0) := 100;
begin
    n := 2255543300; -- ok
    n := 22555433 * mul; -- ok
    n := 22555433 * 100; -- numeric overflow
end;

Answer 1

n := 22555433 * 100;
A calculation with two PLS_INTEGER values ​​that is out of range of PLS_INTEGER throws an overflow exception, even if you assign the result to the NUMBER data type.
Here Tom Kyte analyzes this issue in detail.