hive/mcp-server/node_modules/linkify-it/lib/re.mjs

import { Any, Cc, Z, P } from 'uc.micro'

export default function (opts) {
  const re = {}
  opts = opts || {}

  re.src_Any = Any.source
  re.src_Cc = Cc.source
  re.src_Z = Z.source
  re.src_P = P.source

  // \p{\Z\P\Cc\CF} (white spaces + control + format + punctuation)
  re.src_ZPCc = [re.src_Z, re.src_P, re.src_Cc].join('|')

  // \p{\Z\Cc} (white spaces + control)
  re.src_ZCc = [re.src_Z, re.src_Cc].join('|')

  // Experimental. List of chars, completely prohibited in links
  // because can separate it from other part of text
  const text_separators = '[><\uff5c]'

  // All possible word characters (everything without punctuation, spaces & controls)
  // Defined via punctuation & spaces to save space
  // Should be something like \p{\L\N\S\M} (\w but without `_`)
  re.src_pseudo_letter = '(?:(?!' + text_separators + '|' + re.src_ZPCc + ')' + re.src_Any + ')'
  // The same as abothe but without [0-9]
  // var src_pseudo_letter_non_d = '(?:(?![0-9]|' + src_ZPCc + ')' + src_Any + ')';

  re.src_ip4 =

    '(?:(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)\\.){3}(25[0-5]|2[0-4][0-9]|[01]?[0-9][0-9]?)'

  // Prohibit any of "@/[]()" in user/pass to avoid wrong domain fetch.
  re.src_auth = '(?:(?:(?!' + re.src_ZCc + '|[@/\\[\\]()]).)+@)?'

  re.src_port =

    '(?::(?:6(?:[0-4]\\d{3}|5(?:[0-4]\\d{2}|5(?:[0-2]\\d|3[0-5])))|[1-5]?\\d{1,4}))?'

  re.src_host_terminator =

    '(?=$|' + text_separators + '|' + re.src_ZPCc + ')' +
    '(?!' + (opts['---'] ? '-(?!--)|' : '-|') + '_|:\\d|\\.-|\\.(?!$|' + re.src_ZPCc + '))'

  re.src_path =

    '(?:' +
      '[/?#]' +
        '(?:' +
          '(?!' + re.src_ZCc + '|' + text_separators + '|[()[\\]{}.,"\'?!\\-;]).|' +
          '\\[(?:(?!' + re.src_ZCc + '|\\]).)*\\]|' +
          '\\((?:(?!' + re.src_ZCc + '|[)]).)*\\)|' +
          '\\{(?:(?!' + re.src_ZCc + '|[}]).)*\\}|' +
          '\\"(?:(?!' + re.src_ZCc + '|["]).)+\\"|' +
          "\\'(?:(?!" + re.src_ZCc + "|[']).)+\\'|" +

          // allow `I'm_king` if no pair found
          "\\'(?=" + re.src_pseudo_letter + '|[-])|' +

          // google has many dots in "google search" links (#66, #81).
          // github has ... in commit range links,
          // Restrict to
          // - english
          // - percent-encoded
          // - parts of file path
          // - params separator
          // until more examples found.
          '\\.{2,}[a-zA-Z0-9%/&]|' +

          '\\.(?!' + re.src_ZCc + '|[.]|$)|' +
          (opts['---']
            ? '\\-(?!--(?:[^-]|$))(?:-*)|' // `---` => long dash, terminate
            : '\\-+|'
          ) +
          // allow `,,,` in paths
          ',(?!' + re.src_ZCc + '|$)|' +

          // allow `;` if not followed by space-like char
          ';(?!' + re.src_ZCc + '|$)|' +

          // allow `!!!` in paths, but not at the end
          '\\!+(?!' + re.src_ZCc + '|[!]|$)|' +

          '\\?(?!' + re.src_ZCc + '|[?]|$)' +
        ')+' +
      '|\\/' +
    ')?'

  // Allow anything in markdown spec, forbid quote (") at the first position
  // because emails enclosed in quotes are far more common
  re.src_email_name =

    '[\\-;:&=\\+\\$,\\.a-zA-Z0-9_][\\-;:&=\\+\\$,\\"\\.a-zA-Z0-9_]*'

  re.src_xn =

    'xn--[a-z0-9\\-]{1,59}'

  // More to read about domain names
  // http://serverfault.com/questions/638260/

  re.src_domain_root =

    // Allow letters & digits (http://test1)
    '(?:' +
      re.src_xn +
      '|' +
      re.src_pseudo_letter + '{1,63}' +
    ')'

  re.src_domain =

    '(?:' +
      re.src_xn +
      '|' +
      '(?:' + re.src_pseudo_letter + ')' +
      '|' +
      '(?:' + re.src_pseudo_letter + '(?:-|' + re.src_pseudo_letter + '){0,61}' + re.src_pseudo_letter + ')' +
    ')'

  re.src_host =

    '(?:' +
    // Don't need IP check, because digits are already allowed in normal domain names
    //   src_ip4 +
    // '|' +
      '(?:(?:(?:' + re.src_domain + ')\\.)*' + re.src_domain/* _root */ + ')' +
    ')'

  re.tpl_host_fuzzy =

    '(?:' +
      re.src_ip4 +
    '|' +
      '(?:(?:(?:' + re.src_domain + ')\\.)+(?:%TLDS%))' +
    ')'

  re.tpl_host_no_ip_fuzzy =

    '(?:(?:(?:' + re.src_domain + ')\\.)+(?:%TLDS%))'

  re.src_host_strict =

    re.src_host + re.src_host_terminator

  re.tpl_host_fuzzy_strict =

    re.tpl_host_fuzzy + re.src_host_terminator

  re.src_host_port_strict =

    re.src_host + re.src_port + re.src_host_terminator

  re.tpl_host_port_fuzzy_strict =

    re.tpl_host_fuzzy + re.src_port + re.src_host_terminator

  re.tpl_host_port_no_ip_fuzzy_strict =

    re.tpl_host_no_ip_fuzzy + re.src_port + re.src_host_terminator

  //
  // Main rules
  //

  // Rude test fuzzy links by host, for quick deny
  re.tpl_host_fuzzy_test =

    'localhost|www\\.|\\.\\d{1,3}\\.|(?:\\.(?:%TLDS%)(?:' + re.src_ZPCc + '|>|$))'

  re.tpl_email_fuzzy =

      '(^|' + text_separators + '|"|\\(|' + re.src_ZCc + ')' +
      '(' + re.src_email_name + '@' + re.tpl_host_fuzzy_strict + ')'

  re.tpl_link_fuzzy =
      // Fuzzy link can't be prepended with .:/\- and non punctuation.
      // but can start with > (markdown blockquote)
      '(^|(?![.:/\\-_@])(?:[$+<=>^`|\uff5c]|' + re.src_ZPCc + '))' +
      '((?![$+<=>^`|\uff5c])' + re.tpl_host_port_fuzzy_strict + re.src_path + ')'

  re.tpl_link_no_ip_fuzzy =
      // Fuzzy link can't be prepended with .:/\- and non punctuation.
      // but can start with > (markdown blockquote)
      '(^|(?![.:/\\-_@])(?:[$+<=>^`|\uff5c]|' + re.src_ZPCc + '))' +
      '((?![$+<=>^`|\uff5c])' + re.tpl_host_port_no_ip_fuzzy_strict + re.src_path + ')'

  return re
}