Python

Setup

Install Python 3
- On a Unix-based OS, the system's default Python installation is normally Python 2. Check this by running python --version on a terminal. If this is so, install Python 3 but DO NOT REMOVE / OVERWRITE / UNINSTALL the old Python 2. The system uses Python 2 for its internal scripts and removing it may break the OS installation.
- For Linux, either install it from apt or from source
  - How to install Python3.7 on Ubuntu18.04?
- For Mac, use Homebrew
  - Installing Python 3 on Mac OS X
  - How is Python handled by Homebrew
- Verify the Python installation by running
```
$ which python3
$ python3 --version
```
Setup and use a virtual environment
- There are many different modules for creating a virtual environment
- If Python 2 support is needed, use the virtualenv module
- If Python 3 only, use the built-in venv or poetry
  - For venv
    - Set VENV_DIR to directory for virtual environments (ex. $HOME/.venvs)
    - Copy the bash aliases for using venv
      - mkvenv PROJECT_NAME
      - upvenv PROJECT_NAME
      - dnvenv
      - rmvenv PROJECT_NAME
      - lsvenv
  - For poetry
    - Set poetry config cache-dir
For each Python project
- For virtualenv and venv
  - Create/Activate a virtual environment
```
$ mkvenv NAME /path/to/python
```
  - Create a requirements.txt file
```
$ pip freeze > requirements.txt
```
  - Commit the requirements.txt file
  - When switching environments:
```
$ pip install -r requirements.txt
```
- For poetry
  - Specify the Python version to use
```
$ cd <project>
$ poetry env use python3
$ poetry env use python3.X # 3.7, 3.8, 3.9, 3.10, 3.11
```
  - Initialize the project
```
$ poetry init
```
  - Install packages
```
$ poetry add <package>
$ pipenv add --dev PACKAGE
```
  - Commit the generated pyproject.toml and poetry.lock
  - Activate the environment
```
$ poetry shell
```

FAQ

Coding Conventions

Package Management

Managing Python packages the right way

Notable Modules

Linters

bandit - security issues
flake8 - PEP8 issues
mypy - typing issues
mccabe - complexity checker

Formatters

black - no-config, opinioniated formatter

Terminal

ipython - Enhanced interactive Python shell

CLI

colorama - Cross-platform colorizer for console logs
click - For creating CLI-based applications
typer - For creating CLI-based applications with type-hinting support
tqdm - For better progress bars and loop tracking

Web Application

flask - Lightweight web application framework
fastapi - Modern framework with async support, type-hints, and auto-generated documentation

Data Validation

pydantic - Data validation through type hints

Data Visualization

bokeh - Interactive and embeddable images for modern web browsers