Project Setup and Tooling¶

Python project setup involves choosing an IDE, managing dependencies with virtual environments, and following PEP 8 style conventions. Modern tooling includes pip, venv, poetry, and type checkers like mypy.

Key Facts¶

Always use virtual environments to isolate project dependencies
PEP 8: 4-space indent, snake_case for functions/variables, PascalCase for classes
requirements.txt pins exact versions for reproducibility
pyproject.toml is the modern project configuration standard
Three IDE tiers: text editors (VS Code), full IDEs (PyCharm), notebooks (Jupyter)
python -m venv is built-in; poetry and pipenv are higher-level alternatives

Patterns¶

Virtual Environments¶

python -m venv myenv              # create
source myenv/bin/activate         # activate (Linux/Mac)
myenv\Scripts\activate            # activate (Windows)
pip install package               # install in venv
deactivate                        # exit

pip freeze > requirements.txt     # save dependencies
pip install -r requirements.txt   # install from file

Package Management¶

# pip (standard)
pip install fastapi uvicorn sqlalchemy
pip install -r requirements.txt

# poetry (modern)
poetry init
poetry add fastapi uvicorn
poetry install

# pipenv
pipenv install fastapi
pipenv shell

PEP 8 Style Guide¶

# Naming
my_variable = 42             # snake_case for variables/functions
MY_CONSTANT = 3.14           # UPPER_CASE for constants
class MyClass:               # PascalCase for classes
    pass

# Formatting
x = 1                        # spaces around operators
some_list[0]                 # no space before brackets
my_dict['key']               # no space before brackets

# Line length: max 79 (PEP 8) or 120 (modern projects)
# Imports: one per line, at top of file
# Order: stdlib, third-party, local
import os
import sys

import requests
from sqlalchemy import select

from myapp.models import User

Running Python Code¶

# Script
python my_script.py

# Module
python -m pytest
python -m http.server 8000

# Interactive REPL
python

Development Environments¶

Tool	Type	Best For
VS Code	Editor + extensions	General development, lightweight
PyCharm	Full IDE	Python-specific features, debugging
Jupyter	Notebook	Data exploration, learning, prototyping

Dependency Management Tools¶

Tool	Lock file	Resolver	Virtual env
pip + venv	requirements.txt	Basic	Manual
pipenv	Pipfile.lock	Yes	Automatic
poetry	poetry.lock	Yes	Automatic
uv	uv.lock	Yes	Automatic

Python Installation¶

Anaconda/Miniconda - includes scientific packages (NumPy, pandas, Jupyter)
python.org - standard installer
pyenv - manage multiple Python versions

Gotchas¶

Without virtual env, pip install affects system Python - can break OS tools
pip freeze captures ALL installed packages including transitive deps - use pip-compile for clean dependency management
Default Python on macOS/Linux may be Python 2 - use python3 explicitly
requirements.txt doesn't capture dependency tree - poetry.lock/Pipfile.lock do
VS Code needs the Python extension installed for IntelliSense, linting, debugging