The following are the unary operator method functions:

def __abs__(self) -> 'FixedPoint':
    return FixedPoint(abs(self.value), self.scale)

def __float__(self) -> float:
    return self.value / self.scale

def __int__(self) -> int:
    return int(self.value / self.scale)

def __trunc__(self) -> int:
    return int(math.trunc(self.value / self.scale))

def __ceil__(self) -> int:
    return int(math.ceil(self.value / self.scale))

def __floor__(self) -> int:
    return int(math.floor(self.value / self.scale))

def __round__(self, ndigits: Optional[int] = 0) -> Any:
    return FixedPoint(round(self.value / self.scale, ndigits=ndigits), self.scale)

def __neg__(self) -> 'FixedPoint':
    return FixedPoint(-self.value, self.scale)

def __pos__(self) -> 'FixedPoint':
    return self

For the __round__(), __trunc__(), __ceil__(), and __floor__() operators, we've delegated the work to a Python library function. There are some potential optimizations, but we've taken the lazy route of creating a float approximation and used that to create the desired result. This suite of methods ensures that our FixedPoint objects will work with a number of arithmetic functions. Yes, there are a lot of operators in Python. This isn't the entire suite. We haven't provided implementations for the comparison or bit-kicking operators. The comparisons are generally similar to arithmetic operations, and are left as an exercise for the reader. The bit-wise operators (&, |, ^, and ~) don't have a clear meaning outside the domains, like values or sets, so we shouldn't implement them.

In the next section, we'll see how to implement FixedPoint reflected operators.