Python project structure foundation or template
5.6K
This repository is meant to be used as a minimal, yet opinionated baseline for Python software projects. It includes:
The intended use is to add this repository as a VCS remote for your project. Thus developers can merge changes from this repository as we make changes related to Python project structure and tooling. As we add structure specific to certain types of projects (e.g. CLI scripts, web development, etc.), frameworks (e.g. Flask, Pyramid, Django, etc.), libraries and such, branches will be used for each such variation such that structure common to different variations can be merged back into the branches for those specific variations.
This is a rough guide to applying this project template to your project. This is not thoroughly tested as such tests would be so meta as to be extremely wasteful of developer time to create and maintain. So report any issues you have or better yet figure it out and submit a PR with corrections to this section.
Choose the right branch to use:
Is your project a CLI utility? A web application? Which project hosting provider and/or CI/CD platform will you use? Choose the appropriate branch for your project:
dist:
Basic Python distribution with build, tests, linters, code formatting and release publishing from local developer checkouts.
cli:
The above plus support for project's that provide an executable CLI.
docker:
The dist branch plus Docker containers for both development
and end-users/deployments.
ci:
The above plus GitLab CI/CD pipelines that run tests and linters
as CI and publish releases from develop and main as CD.
ci-cli:
The above plus the cli branch.
etc.
Do not use the develop or main branches in your project as those
branches are used to test the CI/CD automatic releases process and
as such contain bumped versions, release notes, and other release
artifacts that shouldn't be merged into real projects.
Reconcile VCS history:
If starting a fresh project:
$ git clone --origin "template" --branch "ci-cli" \
"https://gitlab.com/rpatterson/python-project-structure.git" "./foo-project"
$ cd "./foo-project"
$ git config remote.template.tagOpt --no-tags
$ git remote add "origin" "[email protected]:foo-username/foo-project.git"
$ git config remote.template.tagOpt --no-tags
$ git switch -C "main" --track "origin/main"
If merging into an existing project:
$ git remote add "template" \
"https://gitlab.com/rpatterson/python-project-structure.git"
$ git config remote.template.tagOpt --no-tags
$ git merge --allow-unrelated-histories "template/ci-cli"
Rename file and directory paths derived from the project name:
$ git ls-files | grep -iE 'python.?project.?structure'
Rename strings derived from the project name and template author identity in project files:
$ git grep -iE 'python.?project.?structure|ross|Patterson'
Examine # TEMPLATE: comments and change as appropriate:
These are the bits that need the developer's attention and reasoning to take the correct action. So read the comments and address them with care and attention:
$ git grep "TEMPLATE"
Finally, remove this section from this ./README.rst and update the
rest of it's content as appropriate for your project. As fixes and
features are added to the upstream template, you can merge them into
your project and repeat steps 3-5 above as needed.
This template publishes pre-releases on all pushes to the develop
branch and final releases on all pushes to the main branch. Project
owners may decide which types of changes should go through pre-release
before final release and which types of changes should go straight to
final release. For example they may decide that:
develop. See the ./CONTRIBUTING.rst
file
for such an example public contributions policy and workflow.main and, after passing all tests and checks, merged back into
main to publish final releases immediately.main as above for bug fixes.Install and use either via a local, native installation or a Docker container image:
Install using any tool for installing standard Python 3 distributions such as pip:
$ pip3 install --user python-project-structure
Optional shell tab completion is available via argcomplete.
The recommended way to use the Docker container image is via Docker Compose. See the example ./docker-compose.yml file for an example configuration. Once you have your configuration, you can create and run the container:
$ docker compose up
Alternatively, you make use the image directly. Pull the Docker image:
$ docker pull "registry.gitlab.org/rpatterson/python-project-structure"
And then use the image to create and run a container:
$ docker run --rm -it "registry.gitlab.org/rpatterson/python-project-structure" ...
Images variant tags are published for the Python version, branch, and
major/minor versions so that users can control when they get new images
over time, e.g.
registry.gitlab.org/merpatterson/python-project-structure:py310-main.
The canonical Python version is 3.10 which is the version used in tags
without py###, e.g.
registry.gitlab.org/merpatterson/python-project-structure:main.
Pre-releases are from develop and final releases are from main which
is also the default for tags without a branch, e.g.
registry.gitlab.org/merpatterson/python-project-structure:py310. The
major/minor version tags are only applied to the final release images
and without the corresponding main branch tag, e.g.
registry.gitlab.org/merpatterson/python-project-structure:py310-v0.8.
Multi-platform Docker images are published containing images for the
following platforms or architectures in the Python 3.10 py310 variant:
linux/amd64linux/arm64linux/arm/v7See the command-line help for details on options and arguments:
$ python-project-structure --help
usage: python-project-structure [-h]
Python project structure foundation or template, top-level package.
optional arguments:
-h, --help show this help message and exit
If using the Docker container image, the container can be run from the command-line as well:
$ docker compose run "python-project-structure" python-project-structure --help
usage: python-project-structure [-h]
Python project structure foundation or template, top-level package.
optional arguments:
-h, --help show this help message and exit
NOTE: This project is hosted on GitLab. There's a mirror on GitHub but please use GitLab for reporting issues, submitting PRs/MRs and any other development or maintenance activity.
See the ./CONTRIBUTING.rst file for more details on how to get started with development.
There are many other Python project templates so why make another? I've been doing Python development since 1998, so I've had plenty of time to develop plenty of opinions of my own.
What I want in a template is complete tooling (e.g. test coverage, linting, formatting, CI/CD, etc.) but minimal dependencies, structure, and opinion beyond complete tooling (e.g. some non-Python build/task system, structure for frameworks/libraries not necessarily being used, etc.). I couldn't find a template that manages that balance so here we are.
I also find it hard to discern from other templates why they made what choices the did. As such, I also use this template as a way to try out various different options in the Python development world and evaluate them for myself. You can learn about my findings and the reasons the choices I've made in the commit history.
Most importantly, however, I've never found a satisfactory approach to keeping project structure up to date over time. So the primary motivation is to use this repository as a remote from which we can merge structure updates over the life of projects using the template.
Content type
Image
Digest
sha256:2ffea34c3…
Size
336.8 MB
Last updated
over 3 years ago
docker pull merpatterson/python-project-structure