The rule detects division expressions where the two operands are/promote to integers, and the result is assigned to a float or double. It is recommended to include a double literal (1.0) as an operand (in case the programmer wants to enforce a double division) or to change the type of the variable to integer (if an integer result is expected).