Python virtual environment best practices

Need consistent, reproducible Python environments across development, CI, and production.

Python virtual environment workflow:

# Create venv (built-in, no extra install needed)
python3 -m venv .venv

# Activate
source .venv/bin/activate  # Linux/macOS
# .venv\Scripts\activate   # Windows

# Verify you're in the venv
which python3  # Should show .venv/bin/python3

# Install dependencies
pip install -r requirements.txt

# Pin exact versions for reproducibility
pip freeze > requirements.txt

# Better: Use separate files
# requirements.in    - direct dependencies (what you need)
# requirements.txt   - pinned versions (what you install)

# Using pip-tools (recommended)
pip install pip-tools

# requirements.in:
# flask>=2.0
# sqlalchemy>=2.0
# redis

pip-compile requirements.in   # Creates pinned requirements.txt
pip-sync requirements.txt     # Install exactly what's pinned

# Dev dependencies separate
pip-compile requirements-dev.in
pip-sync requirements.txt requirements-dev.txt

# .gitignore
.venv/
*.pyc
__pycache__/

Modern alternatives:

# uv (fast Rust-based installer)
pip install uv
uv venv .venv
uv pip install -r requirements.txt

# Poetry
poetry init
poetry add flask sqlalchemy
poetry install
poetry shell

# PDM
pdm init
pdm add flask
pdm install

Key rules:

• Always use a venv (never install globally)
• Pin versions in production
• Add .venv/ to .gitignore
• Use python3 -m pip (not just pip) to avoid path issues

Virtual environments isolate project dependencies, preventing conflicts between projects and ensuring reproducible builds across machines.

Python project setup and dependency management