Skip to content

Latest commit

 

History

History
65 lines (52 loc) · 2.38 KB

CONTRIBUTING.md

File metadata and controls

65 lines (52 loc) · 2.38 KB
Raw

Contributing

Code Style

  • Tab indent code.

  • Keep code within 80 columns.

  • Use the function keyword for functions.

  • Quote all String variables.

  • Prefer single-line expressions where appropriate:

      [[ -n "$foo" ]] && other command

  if   [[ "$foo" == "bar" ]]; then command
  elif [[ "$foo" == "baz" ]]; then other_command
  fi

  case "$foo" in
         bar) command ;;
     baz) other_command ;;
  esac

Pull Request Guidelines

  • Additional features should go in separate files within share/chruby/.
  • All new code must have shunit2 unit-tests.
  • If a feature is out of scope or does more than switches Rubies, it should become its own project. Simply copy the generic Makefile for shell script projects.

What Will Not Be Accepted

  • Completion: tab completion is not necessary for fuzzy matching.
  • Gemset functionality: there is already chgems and others, which provides gemset like behavior.
  • Sub-commands: this tends to promote bloat. It was first popularized by RVM and then copied by rbenv.
  • Bash or Zsh only features: chruby must fully support both bash and zsh.
  • New .files: we should strive to prevent dotfile proliferation, and instead support common dotfiles which other Ruby Switchers could support.
  • New environment variables: much like with dotfile proliferation, we should keep environment variable pollution to a minimum.
  • Ruby/OS specific workarounds: we should not tolerate the existence of bugs in specific Rubies or Operating Systems. Instead we should strive to resolve these bugs, to the benefit of all users.
  • Any RVM/rbenv specific feature: we should not blindly copy the features of RVM or rbenv in the hopes of making new users comfortable. Instead, we should question these features and rethink (or unthink) them.
    • Are some of these features even necessary?
    • Could they be implemented differently?
    • Do they over-optimize rarely performed tasks?
    • Do they make assumptions about workflow?
    • Do they add additional complexity?
    • Do they introduce possible Security risks?