Skip to content

Latest commit

 

History

History
118 lines (93 loc) · 7.23 KB

CONTRIBUTING.md

File metadata and controls

118 lines (93 loc) · 7.23 KB
Raw

Contributing Guidelines

There are two main ways to contribute to the project — submitting issues and submitting fixes/changes/improvements via pull requests.

Submitting issues

Both bug reports and feature requests are welcome. Submit issues here.

  • Search for existing issues to avoid reporting duplicates.
  • When submitting a bug report:
    • Test it against the most recently released version. It might have already been fixed.
    • Include the code reproducing the problem or attach the link to the repository with the project that fully reproduces the problem.
    • However, don't put off reporting any weird or rarely appearing issues just because you cannot consistently reproduce them.
    • If the bug is in behavior, then explain what behavior you've expected and what you've got.
  • When submitting a feature request:
    • Explain why you need the feature &mdash, your use case, and your domain.
    • Explaining the problem you face is more important than suggesting a solution. Report your issue even if you don't have any proposed solution.
    • If there is an alternative way to do what you need, show the alternative's code.

Submitting PRs

We love PRs. Submit PRs here. However, please keep in mind that maintainers will have to support the resulting code of the project, so do familiarize yourself with the following guidelines.

  • All development (both new features and bug fixes) is performed in the master branch.
    • Base PRs against the master branch.
    • PR should be linked with the issue, excluding minor documentation changes, adding unit tests, and fixing typos.
  • If you make any code changes:
  • If you fix a bug:
    • Write the test that reproduces the bug.
    • Fixes without tests are accepted only in exceptional circumstances if it can be shown that writing the corresponding test is too hard or otherwise impractical.
  • If you introduce any new public APIs:
    • All new APIs must come with documentation and tests.
    • If you plan API additions, please start by submitting an issue with the proposed API design to gather community feedback.
    • Contact the maintainers to coordinate any great work in advance via submitting an issue.
  • If you fix documentation:
    • If you plan extensive rewrites/additions to the docs, then please contact the maintainers to coordinate the work in advance.
    • Also, we have a special simple guide how to contribute in the documentation.

PR workflow

  1. The contributor builds the library locally and runs all unit tests via the Gradle task dataframe:test -Pkotlin.dataframe.debug=true (see the "Building" chapter).
  2. The contributor submits the PR if the local build is successful and the tests are green.
  3. The reviewer puts their name in the "Reviewers" section of the proposed PR at the start of the review process.
  4. The reviewer leaves comments or marks the PR with the abbreviation "LGTM" (Looks good to me).
  5. The contributor answers the comments or fixes the proposed PR.
  6. The reviewer marks the PR with the word "LGTM."
  7. The maintainer could suggest merging the master branch to the PR branch a few times due to changes in the master branch.
  8. If the PR influences generated code/samples, a bot will inform about this in the PR comments.
  9. The maintainer runs TeamCity builds (unit tests and examples as integration tests).
  10. TeamCity writes the result (passed or not passed) to the PR checks at the bottom of the proposed PR.
  11. If it is possible, maintainers share the details of the failed build with the contributor.
  12. The maintainer merges the PR if all checks are successful and there is no conflict with the master branch.
  13. The maintainer closes the PR and the issue linked to it.
  14. If the PR influences generated code, a bot will auto-commit the newly generated code into the master branch.

How to fix an existing issue

  • If you are going to work on the existing issue:
    • Comment on the existing issue if you want to work on it.
    • Wait till it is assigned to you by maintainers.
    • Ensure that the issue describes a problem and a solution that has received positive feedback. Propose a solution if there isn't any.
  • If you are going to submit your first PR in this project:
    • Find tickets with the label "good first issue" which are not assigned to somebody.
    • Learn the examples module. Submit an interesting new example or improve documentation for one of them.
  • If you are ready to participate in library design and new experiments, find tickets with the label "research" or join our discussions.

Environment requirements

  • JDK >= 11 referred to by the JAVA_HOME environment variable.

    • Note, any version above 11 should work in theory, but JDK 11 is the only version we test with, so it is the recommended version.

  • We recommend using IntelliJ IDEA as the IDE. This has the best support for Kotlin, compiler plugins, Gradle, and Kotlin Notebook of course.

  • We recommend using the Ktlint plugin for IntelliJ IDEA. It is able to read the .editorconfig file and apply the same formatting rules as Ktlint in the CI.

  • Check out the KDoc Preprocessor guide to understand how to work with the KDoc preprocessor.

Building

This library is built with Gradle.

  • Run ./gradlew build to build. It also runs all the tests and checks the linter.
  • Run ./gradlew <module>:test to test the module you are looking at to speed things up during development.
  • Make sure to pass the extra parameter -Pkotlin.dataframe.debug=true to enable debug mode. This flag will make sure some extra checks are run, which are important but too heavy for production.

You can import this project into IDEA, but you have to delegate the build actions to Gradle (in Preferences -> Build, Execution, Deployment -> Build Tools -> Gradle -> Runner)

Contacting maintainers

  • If something cannot be done or doesn't work conveniently — submit an issue.
  • To attract attention to your problem, raise a question, or make a new comment, mention one of us on Github: @koperagen @Jolanrensen @zaleslaw @ileasile
  • Discussions and general inquiries — use #datascience channel in KotlinLang Slack.