Master Python Imports: Best Practices for Clean and Safe Code

1. Import the whole module

Import the entire module instead of individual functions or classes, then access its contents via the module name.

import math
result = math.sqrt(16)

2. Use an alias

When a module name is long or may clash with other names, assign it a short alias.

import numpy as np
array = np.array([1, 2, 3])

Aliases also help disambiguate functions with the same name from different modules.

from html import escape as html_escape
from xml.sax import saxutils as xml_escape

3. Import only what you need

Import specific functions or classes rather than using from module import *, reducing namespace pollution.

from datetime import datetime, timedelta
now = datetime.now()
tomorrow = now + timedelta(days=1)

4. Avoid from module import *

Wildcard imports pull all public names into the current namespace, increasing the risk of name collisions; use them only in interactive sessions for quick testing.

# Not recommended
from os import *
from sys import *

5. Check for existing names

Before importing a new name, verify that the same identifier does not already exist in the current scope.

if 'name' not in locals():
    from some_module import name
else:
    print("Name 'name' already exists")

6. Use the built‑in __all__ mechanism

If you maintain a module, define an __all__ list to explicitly specify which names are exported when from module import * is used.

# some_module.py
__all__ = ['function_a', 'function_b']

def function_a():
    pass

def function_b():
    pass

def internal_function():
    pass

Example of avoiding name conflicts by importing a module with an alias and preserving the built‑in open function.

import os as os_module
my_open = open  # Save built‑in open

def open(filename):
    """Custom file‑open wrapper"""
    return my_open(filename, 'r')

file_descriptor = os_module.open('path/to/file', os_module.O_RDONLY)