Why can methods in Python standard library be called as functions?

I found a strange phenomenon that some methods in Python standard library can be called as functions. For example, there is a method randint() defined in class Random in module random.py. As I understand, when we wanted to call it, we should import the module random first, then declare a instance of class Random so that method randint() can be called such as

import random as rd
aaa = rd.Random()
aaa.randint()

But I just found that randint() can be called as a function and it is unnecessary to do as above:

import random as rd
rd.randint()

So I wonder how does this illogical phenomenon happen?

Why can methods in Python standard library be called as functions?

Solution

Looking into random.py, we see that in fact an instance is created and its methods exported as module-level functions. There's nothing magic about it.

The comment there describes it beautifully:

# Create one instance, seeded from current time, and export its methods
# as module-level functions.  The functions share state across all uses
# (both in the user's code and in the Python libraries), but that's fine
# for most programs and is easier for the casual user than making them
# instantiate their own Random() instance.

_inst = Random()
# .. SNIP ..
randint = _inst.randint

Other standard library modules take a different but similar approach to export module-level functions. E.g. the re module uses an internal function _compile() to create a local Pattern instance and let it do the work (docstring omitted for brevity):

def search(pattern, string, flags=0):
    return _compile(pattern, flags).search(string)

We can conclude that there's no general mechanism to elevate methods to module-level functions. It's just good interface design on the part of the individual modules.