Discover Python 3.8’s Game‑Changing Features: Walrus Operator, Audit Hooks & Beyond

Assignment Expressions (Walrus Operator)

PEP 572 adds the assignment expression, also known as the "Walrus Operator" ( :=), allowing assignment inside expressions.

fib = lambda n: (lambda f, t, i, a, b: print(f"fib({i}) = {b}") or t == i or f)(lambda f, t, i, a, b: f, 1, 0, 1)

Positional‑Only Parameters

PEP 570 enforces the use of positional‑only parameters, causing a TypeError when keyword arguments are supplied to functions like divmod().

Runtime Audit Hooks

Python 3.8 allows adding audit hooks to monitor events such as network requests.

import sys

def audit_hook(event, args):
    if event in ['urllib.Request']:
        print(f"Network {event=} {args=}")

sys.addaudithook(audit_hook)

import urllib.request
urllib.request.urlopen('https://httpbin.org/get?a=1')

Cross‑Process Shared Memory

Shared memory objects can be accessed across processes using multiprocessing.shared_memory.

from multiprocessing import shared_memory

a = shared_memory.ShareableList([1, 'a', 0.1])
# In another process
b = shared_memory.ShareableList(name=a.shm.name)
print(b)  # Output: ShareableList([1, 'a', 0.1])

Importlib.metadata for Package Introspection

The new importlib.metadata module can retrieve package version, dependencies, and other metadata.

from importlib.metadata import version, files, requires, distribution
print(version('flask'))          # e.g., '1.1.1'
print(requires('requests'))      # list of requirements
print(distribution('celery').metadata['License'])

Cached Property

The cached_property decorator is now part of the standard library, providing a memoized attribute.

lru_cache Without Parentheses

From Python 3.8, functools.lru_cache can be used directly as a decorator without calling it.

@lru_cache
def add(a, b):
    return a + b

Asyncio REPL

Python now includes an interactive asyncio REPL for experimenting with asynchronous code.

F‑strings Debug

A new debugging feature for f‑strings (though its practical usefulness is limited).

Async Mock

The unittest.mock module adds AsyncMock for mocking asynchronous functions.

import asyncio
from unittest.mock import AsyncMock

mock = AsyncMock(return_value={'json': 123})
result = asyncio.run(mock())
print(result)  # {'json': 123}

Iterable Unpacking Fix

Python 3.8 resolves issues with iterable unpacking in comprehensions and other contexts.