Thank you for considering contributing to Kedro! It's people like you that make Kedro such a great tool. We welcome contributions in the form of pull requests (PRs), issues or code reviews. You can add to code, documentation, or simply send us spelling and grammar fixes or extra tests. Contribute anything that you think improves the community for us all!
The following sections describe our vision and the contribution process.
There is some work ahead, but Kedro aims to become the standard for developing production-ready data pipelines. To be production-ready, a data pipeline needs to be monitored, scheduled, scalable, versioned, testable and reproducible. Currently, Kedro helps you develop data pipelines that are testable, versioned, reproducible and we'll be extending our capability to cover the full set of characteristics for data pipelines over time.
The Kedro team pledges to foster and maintain a welcoming and friendly community in all of our spaces. All members of our community are expected to follow our Code of Conduct and we will do our best to enforce those principles and build a happy environment where everyone is treated with respect and dignity.
We use GitHub Issues to keep track of known bugs. We keep a close eye on them and try to make it clear when we have an internal fix in progress. Before reporting a new issue, please do your best to ensure your problem hasn't already been reported. If so, it's often better to just leave a comment on an existing issue, rather than create a new one. Old issues also can often include helpful tips and solutions to common problems.
If you are looking for help with your code, and the FAQs in our documentation haven't helped you, please consider posting a question on Stack Overflow. If you tag it kedro and python, more people will see it and may be able to help. We are unable to provide individual support via email. In the interest of community engagement we also believe that help is much more valuable if it's shared publicly, so that more people can benefit from it.
If you're over on Stack Overflow and want to boost your points, take a look at the kedro tag and see if you can help others out by sharing your knowledge. It's another great way to contribute.
If you have already checked the existing issues in GitHub issues and are still convinced that you have found odd or erroneous behaviour then please file an issue. We have a template that helps you provide the necessary information we'll need in order to address your query.
If you have new ideas for Kedro functionality then please open a GitHub issue with the label Type: Enhancement. You can submit an issue here which describes the feature you would like to see, why you need it, and how it should work.
If you're unsure where to begin contributing to Kedro, please start by looking through the good first issues and help wanted issues on GitHub.
We focus on three areas for contribution: core, contrib or plugin:
corerefers to the primary Kedro librarycontribrefers to features that could be added tocorethat do not introduce too many depencies or require new Kedro CLI commands to be created e.g. adding a new dataset to theiodata management modulepluginrefers to new functionality that requires a Kedro CLI command e.g. adding in Airflow functionality
Typically, we only accept small contributions for the core Kedro library but accept new features as plugins or additions to the contrib module. We regularly review contrib and may migrate modules to core if they prove to be essential for the functioning of the framework or if we believe that they are used by most projects.
Working on your first pull request? You can learn how from these resources:
- Aim for cross-platform compatibility on Windows, macOS and Linux
- We use Anaconda as a preferred virtual environment
- We use SemVer for versioning
Our code is designed to be compatible with Python 3.5 onwards and our style guidelines are (in cascading order):
- PEP 8 conventions for all Python code
- Google docstrings for code comments
- PEP 484 type hints for all user-facing functions / class methods e.g.
def count_truthy(elements: List[Any]) -> int:
return sum(1 for elem in elements if element)
Note: We only accept contributions under the Apache 2.0 license and you should have permission to share the submitted code.
We use a branching model that helps us keep track of branches in a logical, consistent way. All branches should have the hyphen-separated convention of: <type-of-change>/<short-description-of-change> e.g. contrib/io-dataset
| Types of changes | Description |
|---|---|
contrib |
Changes under contrib/ and has no side-effects to other contrib/ modules |
docs |
Changes to the documentation under docs/source/ |
feature |
Non-breaking change which adds functionality |
fix |
Non-breaking change which fixes an issue |
tests |
Changes to project unit tests/ and / or integration features/ tests |
Small contributions are accepted for the core library:
- Fork the project
- Develop your contribution in a new branch and open a PR against the
developbranch - Make sure the CI builds are green (have a look at the section Running checks locally below)
- Update the PR according to the reviewer's comments
You can add new work to contrib if you do not need to create a new Kedro CLI command:
- Create an issue describing your contribution
- Fork the project and work in
contrib - Develop your contribution in a new branch and open a PR against the
developbranch - Make sure the CI builds are green (have a look at the section Running checks locally below)
- Include a
README.mdwith instructions on how to use your contribution - Update the PR according to the reviewer's comments
See the plugin development documentation for guidance on how to design and develop a Kedro plugin.
To run E2E tests you need to install the test requirements which includes behave, do this using the following command:
pip install -r test_requirements.txtAll checks run by our CI / CD servers can be run locally on your computer.
make lintmake testNote: We place conftest.py files in some test directories to make fixtures reusable by any tests in that directory. If you need to see which test fixtures are available and where they come from, you can issue:
pytest --fixtures path/to/the/test/location.pybehaveOur CI / CD also checks that kedro installs cleanly on a fresh Python virtual environment, a task which depends on successfully building the docs:
make build-docsThis command will only work on Unix-like systems and requires pandoc to be installed.
❗ Running
make build-docsin a Python 3.5 environment may sometimes yield multiple warning messages like the following:MemoryDataSet.md: WARNING: document isn't included in any toctree. You can simply ignore them or switch to Python 3.6+ when building documentation.