Modules, Packages, and Imports - Python Mastery

Modules

A module is any Python file. When you import it, Python executes the file and makes its contents available. Modules are how you organize code into reusable units.

# mymodule.py
PI = 3.14159

def circle_area(r):
    return PI * r ** 2

class Circle:
    def __init__(self, radius):
        self.radius = radius

# main.py — importing the module
import mymodule
print(mymodule.PI)
print(mymodule.circle_area(5))

# Import specific names
from mymodule import circle_area, PI
print(circle_area(5))   # No module prefix needed

# Import with alias
import mymodule as mm
from mymodule import circle_area as area
print(area(5))

Import Styles

# Good: explicit imports
import os
import math
from pathlib import Path
from collections import defaultdict, Counter

# Wildcard import — avoid except in interactive use
from math import *   # Pollutes namespace, unclear what's imported

# Relative imports (inside packages)
from . import sibling_module           # Same package
from .sibling_module import some_func  # Specific from sibling
from .. import parent_module           # Parent package

The Standard Library

Python ships with a huge standard library. You don't need to install these:

import os            # OS interaction, environment variables, paths
import sys           # Python interpreter, sys.argv, sys.exit
import re            # Regular expressions
import json          # JSON parsing
import csv           # CSV files
import math          # Math functions
import random        # Random numbers
import datetime      # Dates and times
import time          # Time functions, sleep
import urllib        # URL handling
import http          # HTTP client/server
import socket        # Network sockets
import threading     # Threads
import multiprocessing  # Multiple processes
import subprocess    # Run external commands
import hashlib       # Cryptographic hashes
import base64        # Encoding
import logging       # Logging framework
import unittest      # Unit testing
import argparse      # Command-line argument parsing
import itertools     # Iterator tools
import functools     # Function tools
import operator      # Operator functions
import copy          # Shallow and deep copy
import io            # In-memory file objects
import struct        # Binary data packing

# Quick examples
import os
print(os.getcwd())
print(os.environ.get("HOME"))
os.path.join("folder", "file.txt")   # Cross-platform paths

import datetime
now = datetime.datetime.now()
print(now.strftime("%Y-%m-%d %H:%M:%S"))

Creating Packages

A package is a directory containing an __init__.py file (can be empty) and Python modules.

# Project structure
myapp/
    __init__.py       # Makes myapp a package
    models.py
    utils.py
    config.py
    api/
        __init__.py   # Makes api a subpackage
        routes.py
        auth.py

# Import from package
from myapp import models
from myapp.models import User
from myapp.api.routes import app
from myapp.api import auth

init.py

# myapp/__init__.py
# Controls what 'from myapp import *' gives
__all__ = ["models", "utils"]

# Can run initialization code
print("myapp loaded")

# Common pattern: expose key classes at package level
from .models import User, Product
from .utils import format_date, validate_email

# Now users can: from myapp import User (not from myapp.models import User)

if name == "main"

# utils.py
def add(a, b):
    return a + b

def main():
    print(add(3, 4))

if __name__ == "__main__":
    # This runs ONLY when script is run directly:
    # python3 utils.py
    # NOT when imported:
    # import utils
    main()

Key Takeaways

Every .py file is a module: import it to use its contents
Packages = directories with __init__.py
Standard library is huge: check it before installing third-party packages
if __name__ == "__main__": code that runs only as a script, not when imported
Avoid wildcard imports: explicit imports prevent namespace pollution

Practice Exercises

Create a math_utils.py module with functions for factorial, fibonacci, and is_prime. Import and test from another file.
Create a package called shapes with modules for different shape types. Expose key classes from __init__.py.
Write a script using argparse that takes command-line arguments and does something useful.

← Context Managers pip and Virtual Environments →