highlight.js/es/languages/ruby.js

/*
Language: Ruby
Description: Ruby is a dynamic, open source programming language with a focus on simplicity and productivity.
Website: https://www.ruby-lang.org/
Author: Anton Kovalyov <anton@kovalyov.net>
Contributors: Peter Leonov <gojpeg@yandex.ru>, Vasily Polovnyov <vast@whiteants.net>, Loren Segal <lsegal@soen.ca>, Pascal Hurni <phi@ruby-reactive.org>, Cedric Sohrauer <sohrauer@googlemail.com>
Category: common
*/

function ruby(hljs) {
  const regex = hljs.regex;
  const RUBY_METHOD_RE = '([a-zA-Z_]\\w*[!?=]?|[-+~]@|<<|>>|=~|===?|<=>|[<>]=?|\\*\\*|[-/+%^&*~`|]|\\[\\]=?)';
  const RUBY_KEYWORDS = {
    keyword:
      'and then defined module in return redo if BEGIN retry end for self when ' +
      'next until do begin unless END rescue else break undef not super class case ' +
      'require yield alias while ensure elsif or include attr_reader attr_writer attr_accessor ' +
      '__FILE__',
    built_in: 'proc lambda',
    literal:
      'true false nil'
  };
  const YARDOCTAG = {
    className: 'doctag',
    begin: '@[A-Za-z]+'
  };
  const IRB_OBJECT = {
    begin: '#<',
    end: '>'
  };
  const COMMENT_MODES = [
    hljs.COMMENT(
      '#',
      '$',
      {
        contains: [ YARDOCTAG ]
      }
    ),
    hljs.COMMENT(
      '^=begin',
      '^=end',
      {
        contains: [ YARDOCTAG ],
        relevance: 10
      }
    ),
    hljs.COMMENT('^__END__', '\\n$')
  ];
  const SUBST = {
    className: 'subst',
    begin: /#\{/,
    end: /\}/,
    keywords: RUBY_KEYWORDS
  };
  const STRING = {
    className: 'string',
    contains: [
      hljs.BACKSLASH_ESCAPE,
      SUBST
    ],
    variants: [
      {
        begin: /'/,
        end: /'/
      },
      {
        begin: /"/,
        end: /"/
      },
      {
        begin: /`/,
        end: /`/
      },
      {
        begin: /%[qQwWx]?\(/,
        end: /\)/
      },
      {
        begin: /%[qQwWx]?\[/,
        end: /\]/
      },
      {
        begin: /%[qQwWx]?\{/,
        end: /\}/
      },
      {
        begin: /%[qQwWx]?</,
        end: />/
      },
      {
        begin: /%[qQwWx]?\//,
        end: /\//
      },
      {
        begin: /%[qQwWx]?%/,
        end: /%/
      },
      {
        begin: /%[qQwWx]?-/,
        end: /-/
      },
      {
        begin: /%[qQwWx]?\|/,
        end: /\|/
      },
      // in the following expressions, \B in the beginning suppresses recognition of ?-sequences
      // where ? is the last character of a preceding identifier, as in: `func?4`
      {
        begin: /\B\?(\\\d{1,3})/
      },
      {
        begin: /\B\?(\\x[A-Fa-f0-9]{1,2})/
      },
      {
        begin: /\B\?(\\u\{?[A-Fa-f0-9]{1,6}\}?)/
      },
      {
        begin: /\B\?(\\M-\\C-|\\M-\\c|\\c\\M-|\\M-|\\C-\\M-)[\x20-\x7e]/
      },
      {
        begin: /\B\?\\(c|C-)[\x20-\x7e]/
      },
      {
        begin: /\B\?\\?\S/
      },
      // heredocs
      {
        // this guard makes sure that we have an entire heredoc and not a false
        // positive (auto-detect, etc.)
        begin: regex.concat(
          /<<[-~]?'?/,
          regex.lookahead(/(\w+)(?=\W)[^\n]*\n(?:[^\n]*\n)*?\s*\1\b/)
        ),
        contains: [
          hljs.END_SAME_AS_BEGIN({
            begin: /(\w+)/,
            end: /(\w+)/,
            contains: [
              hljs.BACKSLASH_ESCAPE,
              SUBST
            ]
          })
        ]
      }
    ]
  };

  // Ruby syntax is underdocumented, but this grammar seems to be accurate
  // as of version 2.7.2 (confirmed with (irb and `Ripper.sexp(...)`)
  // https://docs.ruby-lang.org/en/2.7.0/doc/syntax/literals_rdoc.html#label-Numbers
  const decimal = '[1-9](_?[0-9])*|0';
  const digits = '[0-9](_?[0-9])*';
  const NUMBER = {
    className: 'number',
    relevance: 0,
    variants: [
      // decimal integer/float, optionally exponential or rational, optionally imaginary
      {
        begin: `\\b(${decimal})(\\.(${digits}))?([eE][+-]?(${digits})|r)?i?\\b`
      },

      // explicit decimal/binary/octal/hexadecimal integer,
      // optionally rational and/or imaginary
      {
        begin: "\\b0[dD][0-9](_?[0-9])*r?i?\\b"
      },
      {
        begin: "\\b0[bB][0-1](_?[0-1])*r?i?\\b"
      },
      {
        begin: "\\b0[oO][0-7](_?[0-7])*r?i?\\b"
      },
      {
        begin: "\\b0[xX][0-9a-fA-F](_?[0-9a-fA-F])*r?i?\\b"
      },

      // 0-prefixed implicit octal integer, optionally rational and/or imaginary
      {
        begin: "\\b0(_?[0-7])+r?i?\\b"
      }
    ]
  };

  const PARAMS = {
    className: 'params',
    begin: '\\(',
    end: '\\)',
    endsParent: true,
    keywords: RUBY_KEYWORDS
  };

  const RUBY_DEFAULT_CONTAINS = [
    STRING,
    {
      className: 'class',
      beginKeywords: 'class module',
      end: '$|;',
      illegal: /=/,
      contains: [
        hljs.inherit(hljs.TITLE_MODE, {
          begin: '[A-Za-z_]\\w*(::\\w+)*(\\?|!)?'
        }),
        {
          begin: '<\\s*',
          contains: [
            {
              begin: '(' + hljs.IDENT_RE + '::)?' + hljs.IDENT_RE,
              // we already get points for <, we don't need poitns
              // for the name also
              relevance: 0
            }
          ]
        }
      ].concat(COMMENT_MODES)
    },
    {
      className: 'function',
      // def method_name(
      // def method_name;
      // def method_name (end of line)
      begin: regex.concat(/def\s+/, regex.lookahead(RUBY_METHOD_RE + "\\s*(\\(|;|$)")),
      relevance: 0, // relevance comes from kewords
      keywords: "def",
      end: '$|;',
      contains: [
        hljs.inherit(hljs.TITLE_MODE, {
          begin: RUBY_METHOD_RE
        }),
        PARAMS
      ].concat(COMMENT_MODES)
    },
    {
      // swallow namespace qualifiers before symbols
      begin: hljs.IDENT_RE + '::'
    },
    {
      className: 'symbol',
      begin: hljs.UNDERSCORE_IDENT_RE + '(!|\\?)?:',
      relevance: 0
    },
    {
      className: 'symbol',
      begin: ':(?!\\s)',
      contains: [
        STRING,
        {
          begin: RUBY_METHOD_RE
        }
      ],
      relevance: 0
    },
    NUMBER,
    {
      // negative-look forward attempts to prevent false matches like:
      // @ident@ or $ident$ that might indicate this is not ruby at all
      className: "variable",
      begin: '(\\$\\W)|((\\$|@@?)(\\w+))(?=[^@$?])' + `(?![A-Za-z])(?![@$?'])`
    },
    {
      className: 'params',
      begin: /\|/,
      end: /\|/,
      relevance: 0, // this could be a lot of things (in other languages) other than params
      keywords: RUBY_KEYWORDS
    },
    { // regexp container
      begin: '(' + hljs.RE_STARTERS_RE + '|unless)\\s*',
      keywords: 'unless',
      contains: [
        {
          className: 'regexp',
          contains: [
            hljs.BACKSLASH_ESCAPE,
            SUBST
          ],
          illegal: /\n/,
          variants: [
            {
              begin: '/',
              end: '/[a-z]*'
            },
            {
              begin: /%r\{/,
              end: /\}[a-z]*/
            },
            {
              begin: '%r\\(',
              end: '\\)[a-z]*'
            },
            {
              begin: '%r!',
              end: '![a-z]*'
            },
            {
              begin: '%r\\[',
              end: '\\][a-z]*'
            }
          ]
        }
      ].concat(IRB_OBJECT, COMMENT_MODES),
      relevance: 0
    }
  ].concat(IRB_OBJECT, COMMENT_MODES);

  SUBST.contains = RUBY_DEFAULT_CONTAINS;
  PARAMS.contains = RUBY_DEFAULT_CONTAINS;

  // >>
  // ?>
  const SIMPLE_PROMPT = "[>?]>";
  // irb(main):001:0>
  const DEFAULT_PROMPT = "[\\w#]+\\(\\w+\\):\\d+:\\d+>";
  const RVM_PROMPT = "(\\w+-)?\\d+\\.\\d+\\.\\d+(p\\d+)?[^\\d][^>]+>";

  const IRB_DEFAULT = [
    {
      begin: /^\s*=>/,
      starts: {
        end: '$',
        contains: RUBY_DEFAULT_CONTAINS
      }
    },
    {
      className: 'meta',
      begin: '^(' + SIMPLE_PROMPT + "|" + DEFAULT_PROMPT + '|' + RVM_PROMPT + ')(?=[ ])',
      starts: {
        end: '$',
        contains: RUBY_DEFAULT_CONTAINS
      }
    }
  ];

  COMMENT_MODES.unshift(IRB_OBJECT);

  return {
    name: 'Ruby',
    aliases: [
      'rb',
      'gemspec',
      'podspec',
      'thor',
      'irb'
    ],
    keywords: RUBY_KEYWORDS,
    illegal: /\/\*/,
    contains: [
      hljs.SHEBANG({
        binary: "ruby"
      })
    ]
      .concat(IRB_DEFAULT)
      .concat(COMMENT_MODES)
      .concat(RUBY_DEFAULT_CONTAINS)
  };
}

export { ruby as default };