# the remainig is not easy to solve: float_var = float_cast(extpression) # This cannot be fixed by simply casting again to float because some # compilers may ignore the double casting.