Modules & Packages

Modules: familiar concept, simpler syntax

Python’s module system is conceptually identical to ESM: every .py file is a module, you import from other files or installed packages, and a package manager (pip) handles third-party dependencies. The syntax is cleaner — no export keyword, no default vs named export distinction.

Importing modules

// TypeScript (ESM)
import * as fs from "fs";
import { join, dirname } from "path";
import axios from "axios";  // default export from npm

// Named import with alias
import { readFileSync as readFile } from "fs";

console.log(join("src", "main.ts"));

# Python
import os                          # import whole module
from os.path import join, dirname  # import specific names
import math as m                   # import with alias
from math import sqrt, pi          # import names directly

print(join("src", "main.py"))
print(m.floor(3.7))
print(sqrt(16))

Packages vs modules

In Python:

A module is a single .py file.
A package is a directory containing an __init__.py file (Python 3.3+ also supports “namespace packages” without it).

// TypeScript project structure
// src/
//   utils/
//     index.ts      ← barrel export
//     string.ts
//     number.ts
//   main.ts

// main.ts
import { formatDate } from "./utils";
import { capitalize } from "./utils/string";

# Python package structure
# src/
#   utils/
#     __init__.py   ← marks it as a package
#     string_utils.py
#     number_utils.py
#   main.py

# main.py
from utils import format_date        # from __init__.py
from utils.string_utils import capitalize

pip — Python’s npm

pip is the standard package installer. The Python ecosystem equivalent of package.json is requirements.txt (simple) or pyproject.toml (modern).

# npm / Node.js workflow
# package.json declares dependencies

npm install           # install all deps
npm install express   # add a package
npm install -D jest   # add dev dependency

# import in code
import express from "express";

# pip / Python workflow
# requirements.txt declares dependencies

pip install -r requirements.txt  # install all
pip install fastapi               # add a package
pip install pytest --dev          # dev dependency

# requirements.txt
# fastapi==0.110.0
# uvicorn==0.27.0
# pytest==7.4.0

# Modern: use pyproject.toml + pip install -e .

The standard library — batteries included

Python ships a rich standard library. Many things you reach for npm to do are built-in:

Python stdlib	npm equivalent
`json`	`JSON.parse` / `JSON.stringify`
`os`, `pathlib`	`path`, `fs`
`datetime`	`date-fns`, `dayjs`
`re`	`regexp` patterns
`collections`	lodash data structures
`itertools`	lodash iteration helpers
`http.server`	express (very basic)
`unittest`	jest (basic)

Try it

Open in an online REPL

import os
import math
from collections import Counter

# os — path manipulation
path = os.path.join("src", "utils", "main.py")
print(f"Path: {path}")

# math — built-in math functions
print(f"sqrt(144) = {math.sqrt(144)}")
print(f"pi = {math.pi:.4f}")
print(f"ceil(2.3) = {math.ceil(2.3)}")

# collections.Counter — frequency counting
words = ["apple", "banana", "apple", "cherry", "banana", "apple"]
freq = Counter(words)
print(f"Word counts: {dict(freq)}")
print(f"Most common: {freq.most_common(2)}")