Contributing

Before anything else, thank you for taking some of your precious time to help this project move forward. ❤️

If you're new to open source and feeling a bit nervous, we understand! We recommend watching this excellent guide to give you a grounding in some of the basic concepts. We want you to feel safe to make mistakes, and ask questions.

If anything in this guide or anywhere else in the codebase doesn't make sense to you, please let us know! It's through your feedback that we can make this codebase more welcoming, so we'll be glad to hear thoughts.

You can chat with us in the #committers channel in our Community Discord, or feel free to raise an issue or start a discussion if you're experiencing any friction trying to make your contribution.

About to contribute?

Great! Before making a significant contribution, consider discussing the outline of your solution first. We don't want to waste your or our time working on something we might not accept.

Please also read this guide all the way through.

AI Usage

Cucumber is developed and maintained by humans. We read every discusison, issue and pull request. As such we appreciate it, if you would take time to communicate with us.

When you use AI, we require that you do the following:

• All AI usage in any form must be disclosed If you have used AI, you must state to what extent the work was AI-assisted. For example, did you an LLM to translate your pull request description? Or did the LLM generate the code for you which you then edited? Ect, ect.
• You must fully understand all code. You should be able to explain what your changes do and how they interact with the greater ecosystem without the aid of AI tools.
• You must be in the loop. Any content generated with AI must have been reviewed and edited by you (a human), before you submit it to us.

Project organization

Note: This is the catch-all contributing guide and not all repositories are organized the same. So this guide won't help you with anything specific. The repository might have a dedicated CONTRIBUTING.md. If so, you'll want to consult that instead of the next two sections.

In general there are two types. Polyglot and monoglot.

Polyglot

Polyglot repositories contains near identical implementations of a component in different languages. It typically looks like this:

.
├── README.md
├── CHANGELOG.md
├── CONTRIBUTING.md
├── LICENSE
├── dotnet          # the dotnet implementation
├── java            # the java implementation
├── javascript      # the javascript implementation
├── python          # the python implementation
├── ruby            # the ruby implementation
└── testdata        # test data for acceptance tests

Generally speaking, each implementation can be built independently of the others using the tools typical for that language. If you don't know what those are .github/workflows/test-<language>.yaml would be a good place to start.

Monoglot

If the repository you're looking at doesn't look like a polyglot repo it probably is for a single language. In this case it can probably be built using the tools typical for that language. If you don't know what those are, then again .github/workflows/test-<language>.yaml would be a good place to start.