Skip to content

Latest commit

 

History

History
407 lines (329 loc) · 18.9 KB

CONTRIBUTING.md

File metadata and controls

407 lines (329 loc) · 18.9 KB
Raw

Contributing to Reliabot

🙏Thanks in advance for your contributions ❤️

There are many ways to contribute – one of the best is just using Reliabot for your projects – and no contribution is too small or too large.

See the Table of contents for different ways to help and details about how to contribute to Reliabot. Please read the relevant section before contributing. This makes the process easier and improves the experience for everyone.

👀 Looking forward to your contributions 🎉

If you use or like the project, but don't have time to contribute, that's OK. Here are some easy ways to support Reliabot and show your appreciation.

  • Star the Reliabot GitHub project
  • Refer to Reliabot in your project's README
  • Link to Reliabot in social media

Table of contents

Asking a question

💡 If you have a Reliabot question, take a look at the README and its FAQ to see if it's answered there.

Before asking a question, search for existing issues that are similar. If you find a relevant issue but still need clarification, ask your question in that issue. A few internet search queries may also turn up an answer quickly.

If you don't find an answer or an existing issue, follow these steps to ask your question:

  • Create a new issue, labeling it as a 🟣 question.
  • Provide as much context as you can for your question.
    • Reliabot version
    • Whether you're running Reliabot with pre-commit or on the command line
    • Operating System (distribution) and Python versions
  • Explain what you're trying to do, and what you don't understand.

Someone should respond to the issue as soon as possible. Be patient. 🕗😊

Reporting problems

Before submitting a problem report

A good problem report shouldn't force others to contact you for critical missing information. Please investigate carefully, find examples and counter-examples, and describe the problem in detail. Completing the following steps in advance helps to fix any potential bug as fast as possible.

  • To see if other users have experienced (and perhaps already solved) your problem, search the bug tracker for similar issues or error messages.
  • Searching the internet (especially Stack Overflow) can help you find any users of other tools, or in other forums, who have run into a similar problem.
  • Make sure you are using the latest Reliabot version.
  • Confirm that your problem is really a bug and not an error on your part, such as using Reliabot with unsupported environments or versions. Reading the documentation can help. If you need help, consider just asking a question first.
  • Collect information about the problem:
    • Specific Reliabot command line and context.
    • Reliabot output, including error messages, warnings, and stack traces. Setting PYTHONWARNINGS=default in the environment can show extra diagnostics.
    • Your .pre-commit-config.yml, dependabot.yml, and .gitignore files.
    • If your GitHub repository is private, output from tree or ls -R.
    • Versions of Reliabot, Python, and your Operating System (distribution).
    • Output from reliabot.py --self-test (if relevant).
    • Can you reliably reproduce the problem? And can you also reproduce it with older Reliabot versions?

How to submit a good problem report

⚠️Don't report security related issues or vulnerabilities, or include sensitive information in reports on the issue tracker, or elsewhere in public. Instead, open a draft security advisory and provide the vulnerability information there. See the Reliabot security policy for more details.

The Reliabot project uses GitHub issues for tracking bugs. If you found a problem using Reliabot:

  • Create an issue using the problem template. It may not be clear yet whether it's a bug, so please don't label the issue.
  • Describe the behavior you expect and the differences in the actual behavior.
  • Please provide as much context as possible and describe the reproduction steps that others can follow to recreate the issue. This may include your GitHub repository, probably its pre-commit-config.yaml, and definitely your dependabot.yml file.
  • For the best problem reports and fastest fixes, try to isolate the problem and create a reduced test case.
  • Provide the information you collected in the previous section.
  • If you would like to develop a fix for the problem, or already have a proposed change, please mention this in the problem report.

Once you have created the issue:

  • Someone reviews it, labeling the issue with one of these:
    • 🔴 bug – if it's really a Reliabot bug or error
    • ⚪️ duplicate – if there's already an issue for this
    • 🔵 enhancement – if Reliabot works as intended, but could work better
    • 🔵 help wanted – if they aren't sure what to label it
    • ⚪️ invalid if this is spam or nonsense or not even a question
    • 🟣 question – if it's not a bug, and you should do something differently
    • ⚪️ wontfix – if Reliabot is working as intended, and won't change
  • They try to reproduce the problem using your provided steps. If there are no steps or obvious way to reproduce it, they may ask you for those steps and mark the issue as help wanted or invalid. Until someone can reproduce the problem, they won't try to fix it.
  • Once they can reproduce the problem, they'll assign it to someone to implement a fix. If you indicate interest in working on a fix, or that you already have a proposed fix, they may assign it to you.

Suggesting enhancements

This section shows you how to submit an enhancement suggestion for Reliabot, including completely new features and minor improvements to existing capabilities. Following these guidelines helps others understand your suggestion and how it relates to other approaches.

Before suggesting an enhancement

  • Make sure you are using the latest Reliabot version.
  • Read the documentation to see if Reliabot can already do what you want.
  • Search the issues to see if somebody already suggested this enhancement. If so, add a comment to that existing issue instead of opening a new one.
  • Find out whether your idea fits with the scope and aims of the project. It's up to you to make a strong case for the merits of this feature. Keep in mind that features should be useful to many users and not just a few of them.

Writing a good enhancement suggestion

The Reliabot project tracks enhancement suggestions with GitHub issues. To suggest an enhancement for Reliabot:

  • Create an issue using the enhancement template, labeling it as an 🔵 enhancement.
  • Use a clear and descriptive title to identify the enhancement.
  • Describe any problem the enhancement would solve, and explain how.
  • Provide a clear and concise description of the suggested enhancement.
  • Describe the current behavior and explain the behavior you'd like to see instead. You can also describe alternative solutions that aren't as good, and explain why.
  • Explain why this enhancement would be useful to many reliabot users.

Becoming a contributor

Legal Notice

By contributing code or documentation to this project, you are certifying, per the Developer Certificate of Origin version 1.1 or later, that you have the necessary rights to the content you contribute, and that they allow distribution under the Reliabot project license,

You do this by "signing off" your commits. "Signing off" isn't cryptographic, it's just a line like "Signed-off-by: your name your.email@example.org" at the end of a commit message. You can add it manually when squash merging a PR, or by giving the -s or --signoff option to git commit.

First contributions

Unsure where to start contributing? Start by looking at Reliabot issues labeled good first issue or help wanted.

  • good first issue – should only require a few lines of code and a test.
  • help wanted – more complex, requiring code design and Python experience.

Pre-commit setup

Reliabot uses pre-commit to perform lint and style checks and fixes. Whether you are contributing code or documentation, install pre-commit on your system, and then enable it for your local copy of the Reliabot repository:

$ cd reliabot && pre-commit install
pre-commit installed at .git/hooks/pre-commit

If you don't (or can't) install pre-commit on your system (say, if you're using the GitHub file editor in the web UI to fix a small typo), once you create a PR, the pre-commit.ci service runs the pre-commit checks for you and pushes fixes to your PRs for many style / formatting issues. Run git pull on your PR branch after the pre-commit.ci - pr checks are no longer pending, and show either success or failure.

Contributing code and bug fixes

When contributing new code or fixes for existing code, please add or enhance the Reliabot code tests to cover the proposed changes. For a bug fix, add a test that checks for the bug and only passes with the fix. For new features, add new or extend existing tests to provide coverage and prevent regressions from future code changes. Tests should fail with existing code, and pass with your proposed changes.

Testing

The reliabot.py script integrates tests within itself as doctests in module and function docstrings, and the __test__ map at the end of the script. Put tests of bug fixes or corner cases for more thorough coverage in the __test__ map, keeping the function docstrings focused on explaining typical use cases.

You can run these doctests with any Reliabot installation by running it with the --self-test argument:

reliabot$ reliabot/reliabot.py --self-test 2>/dev/null | tail -5
   1 tests in __main__.usage
   4 tests in __main__.validate_dependabot_config
147 tests in 41 items.
147 passed and 0 failed.
Test passed.

The complete Reliabot test suite of Reliabot runs with tox. You can run them with make tests, and install tox with make tox if you can't or don't want to use your system's package manager to install it.

Linting and pre-commit checks

Reliabot uses an extensive set of linters and checkers to identify potential problems with Python code. While developing code, you can run these at any time with pre-commit. You can also run just one specific tool with a command like pre-commit run ruff or pre-commit run prospector. The fast Ruff linter runs most of the checks, but Reliabot also uses the much slower Prospector to run Pylint checks, since many of those still aren't implemented by Ruff.

Checks aren't always 100% accurate; there are cases where code needs to do possibly questionable things. Most of the checkers allow you to suppress checks for specific lines of code with comments like # noqa: B123 (for Ruff) or # pylint: disable=some-code-smell (for Pylint), and sometimes both may be necessary for a single line.

Improving the documentation

The documentation for Reliabot currently consists of a number of Markdown files in standard locations (README.md, CONTRIBUTING.md, SECURITY.md).

Reliabot's pre-commit configuration includes Markdown linters and formatters, as well as a hook for Vale to review prose style and recommend changes. Vale generates style errors, warnings, and suggestions. Only errors cause pre-commit failures, but if there are errors, pre-commit also shows warnings and suggestions.

To run it directly, use make checks, which always shows warnings and suggestions. The Makefile rule for checks expects you to have already installed Vale with brew, choco, or a Linux package manager. Alternately, you can use make vale to install Vale with pipx and add its directory to your PATH. You may need to run rehash or start a new shell or login to run Vale after doing that the first time.

When contributing changes to the documentation, you must address any Vale errors, and it's strongly recommended to address warnings as well. You don't need to address suggestions, and any changes you make to do so shouldn't sacrifice clarity or conciseness.

If Vale doesn't recognize a correctly spelled word, just add it to styles/config/vocabularies/Reliabot/accept.txt. That file uses case-sensitive regular expressions to accept words.

Code tests in Markdown documentation

A script wrapper for doctest-cli tests fenced code blocks with a console language tag in Markdown files. It executes command lines following a prompt (lines starting with non-whitespace followed by $ and whitespace) and if their output doesn't match the following lines in the code block, it fails. When that happens, fix either the code or the documentation to match the other. If you make changes to console fenced code blocks in the documentation, you should run these tests using tox -e doctest-cli.

Style guides

Pre-commit checks cover most of the style guides below, and often auto-fix style issues, either as local changes in your copy of the Reliabot repository, or as commits that pre-commit.ci adds to your PR.

Commit messages

  • Limit the first line to 72 or fewer characters.
  • Start the first line with a conventional commit type, like these examples: feat: or ci(pre-commit):.
    • You can use commitizen or similar tools to help manage this.
  • Use the present tense ("Add feature" not "Added feature").
  • Use the imperative mood ("Move cursor to…" not "Moves cursor to…").
  • Reference issues and pull requests liberally after the first line.

Python code

Reliabot code follows Black style formatting. Pre-commit checks enforce this, along with various PEP code styles, using the much faster Ruff formatter.

Shell code

Reliabot pre-commit hooks format shell scripts with shfmt -i 2 -ci, and lint them with shellcheck.

Markdown documentation

Reliabot documentation uses GitHub Flavored Markdown, following the Markdown style guide implemented by Executable Books mdformat, with some non-default mdformat style settings:

  • "Consecutive" numbering for ordered lists for compatibility with the markdownlint checker, like this: 
    1. first
2. second
3. etcetera
  • Word wrapping at 79 columns.
  • A pre-commit hook based on a Python script from a decade-old blog post rewrites all reference links with numeric tags ordered by appearance in the text. This doesn't affect inline links like [shell](#shell-code), used for any references to anchors in the same file but not anything else.

While these settings don't minimize Git diffs in line-by-line diff mode, they do maximize readability of the Markdown source files, and the --word-diff=color option for git diff generates usable diffs in most cases.

Reliabot uses several mdformat plugins to auto-generate tables of contents, format Markdown tables, and to format code blocks with shell, Python, and configuration language tags.

Configuration files

Reliabot uses Prettier to automatically format CFG, INI, JSON, TOML, and YAML configuration files. The only non-standard setting is to use single quotes in preference to double quotes for formats that allow this.

Attribution

The contributing-gen generator created the initial version of this file. Make your own CONTRIBUTING.md 📝