Python virtual environments and dependency isolation

Need to isolate project dependencies and avoid conflicts between different Python projects.

Modern Python dependency management:

# Built-in venv (recommended for most cases)
python3 -m venv .venv
source .venv/bin/activate  # Linux/macOS
# .venv\Scripts\activate   # Windows

# Install dependencies
pip install -r requirements.txt

# Freeze current state
pip freeze > requirements.txt

# Better: use pip-tools for reproducible builds
pip install pip-tools

# requirements.in (what you want)
django>=4.2
celery
redis

# Compile to locked versions
pip-compile requirements.in  # -> requirements.txt with exact versions
pip-sync requirements.txt    # Install exactly what's specified

# Modern alternative: uv (much faster)
pip install uv
uv venv
uv pip install -r requirements.txt
# Or use uv directly without activating:
uv run python script.py

Project structure:

project/
  .venv/                  # Never commit this
  requirements.in         # Direct dependencies
  requirements.txt        # Locked (compiled) dependencies
  requirements-dev.txt    # Dev-only dependencies
  pyproject.toml         # Modern project metadata

Add to .gitignore: .venv/, __pycache__/, *.pyc

Without virtual environments, all projects share the global Python, causing version conflicts between projects.

Python project setup and dependency management