Skip to content

Latest commit

 

History

History
177 lines (132 loc) · 6.73 KB

README.md

File metadata and controls

177 lines (132 loc) · 6.73 KB

Python Project Template

This is an example Python project, using my preferred project structure and development environment setup. The instructions below are tailored for macOS.

Python Development Environment

Approach:

  • Use Conda as an environment manager
  • Use Poetry as the package/dependency manager

Benefits of this approach include:

  • Better dependency management (often faster than Conda's dependency resolver)
  • Having most package configurations (e.g., pytest, coverage , bump2version, etc.) in a single file
  • The ability to install a Conda package if you have to (this should be a last resort!)
  • Poetry can automatically add new packages to the pyproject.toml file
  • Poetry can show the list of library dependencies of individual packages
  • Build a Python package and publishing to PyPI is as easy as running two commands
  • No need to have separate environment files for production and development environments

Resources and references:

One-Time Steps Per Machine

Run these steps once per machine.

  1. Install Homebrew or ensure it's up-to-date by running:

     $ brew update

  2. Install Xcode or ensure it's up-to-date by checking the Mac App Store

  3. Install Xcode command line tools:

     $ xcode-select --install

  4. Install Conda:

     $ brew install miniconda

  5. Configure Conda for your shell:

     $ conda init "$(basename "${SHELL}")"

Project Steps

Run these steps whenever creating a new Python project (or, just clone this template project which already includes all of these steps and in that case, skip to the next section).

Naming conventions:

  • Project names and directory names, except for package names, are all lowercase with words separated by single dashes (-)
  • Package names and file names are all lowercase, with words separated by single underscores (_)
  • Python variable names are all lowercase, with words separated by single underscores (_)
  • Python constant variable names are all uppercase, with words separated by single underscores (_)
  • Python class names are CamelCase
  • Line lengths are 79
  • All other rules of PEP8 are followed

  1. Create an environment.yaml file that specifies the desired Python environment; this will be used by Conda to create an isolated Python environment. A default environment.yaml is provided below. At a minimum, specify the project name in name. (Note: the poetry dependency is important; this will be used later on in the setup as the dependency manager.)

    name: python-project-template
dependencies:
  - python=3.9
  - poetry=1.4.0

  2. Create the Conda environment from environment.yaml:

     $ conda env create -f environment.yaml

  3. Activate the environment by specifying the project name to conda activate (in this example, python-project-template):

     $ conda activate python-project-template

  4. Verify the environment was activated correctly:

     $ conda env list         # should output all environments with a * next to the active env
 $ poetry --version       # should output Poetry version X.Y.Z

  5. To deactivate the environment (and return to the base environment):

     $ conda deactivate

  6. Add required Python dependencies for your application (this updates pyproject.toml). For example, add the flask package:

     $ poetry add flask

  7. When done adding required dependencies, install them into the project environment:

     $ poetry install

    This generates poetry.lock, and also installs the current project. Be sure to commit this file after generating it.

  8. To get the latest versions of the dependencies and to update poetry.lock, run:

     $ poetry update

  9. In PyCharm, be sure to right-click on the src directory and select Mark Directory as > Sources Root. If it is already set as the Sources Root, this option will not be available so no action is required.

  10. To run tests (make sure you've run poetry install at least once so the current project is installed):

    $ cd tests 
$ pytest .

  11. To run tests with coverage, first add and install the pytest-cov dependency:

    $ poetry add --dev pytest-cov

  12. Run tests with coverage (replace python_project_template with your project package name):

    $ pytest --cov=python_project_template tests/

  13. Add pre-commit to dependencies:

    $ poetry add pre-commit --dev

  14. In PyCharm, add black to External Tools to be able to format with it. Go to PyCharm Preferences > External Tools, create a new External Tool, and enter the following. Note the Program path may differ slightly on your machine; to confirm, run which black from within the project's Python environment. When adding the program to PyCharm, be sure to replace the environment name in the path with $ProjectName.

    1. Name: Black

    Program: /usr/local/Caskroom/miniconda/base/envs/$ProjectName$/bin/black 3. Arguments: "$FilePath$" 4. Working Directory: $ProjectFileDir$

Starting From This Template

  1. After cloning the project, edit environment.yaml and set the project name.

  2. Run the following to create the Conda environment for this project:

     $ conda env create -f environment.yaml

  3. Activate the environment by specifying the project name to conda activate:

     $ conda activate <name>

  4. Verify the environment was activated correctly:

     $ conda env list         # should output all environments with a * next to the active env
 $ poetry --version       # should output Poetry version X.Y.Z

  5. Install all Python dependencies:

     $ poetry install

  6. Add new dependencies if necessary:

     $ poetry add <package_name>    # be sure to commit the updated pyproject.toml and poetry.lock files!

  7. Run the following commands to activate the required git pre-commit hooks:

     $ pre-commit install

  8. Run the project test suite to verify everything is working:

     $ cd tests
 $ pytest .