parser.js

var regexp = LSD.RegExp({
  'fn_tail': '\\.',
  'fn_arguments': '\\g<inside_round>',
  'fn_name': '\\g<unicode>',
  'fn': '\\g<fn_tail>?\\g<fn_name>\\(\\g<fn_arguments>\\)',
  
  'block_arguments': '\\g<inside_curly>',
  'block_body': '\\g<inside_curly>',
  'block': '\\{\\g<block_arguments>?\\g<block_body>\\}',
  
  'double_string': '"((?:[^"]|\\")*)"/',
  'single_string': "'((?:[^']|\\')*)'",
  'string': '\\g<double_string>|\\g<single_string>'
  
  'whitespace': '\\s+',
  'comma': ',',
  'semicolon': ';',
  'separator': '(\\g<whitespace>|\\g<comma>|\\g<semicolon>)'
  
  'index': '\\[\\s*\\g<inside_square>\\s*\\]',
  
  'token_tail': '\\.',
  'token_name': '\\g<unicode>',
  'token': '\\g<token_tail>?\\g<token_name>',
  
  'length_number': '[-+]?(?:\\d+\\.\\d*|\\d*\\.\\d+)'
  'length_unit': 'em|px|pt|%|fr|deg|(?=$|[^a-zA-Z0-9.])'
  'length': '\\g<length_number>\\g<length_unit>',
  
  'operator': '([-+]|[\\/%^~=><*\\^!|&$]+)'
}, {
  fn: function(tail, name, args) {

  },

  block: function(args, body) {

  },

  index: function(content) {

  },

  token: function(name) {

  },

  length: function(number, unit) {

  },

  operator: function(name) {

  },
  
  separator: function(token) {
    switch (token) {
      case ';':
        break;
      case ',':
        break;
    }
  }
})

var results = regexp.exec(string);

This is what it would look like using 'Subexp call ("Tanaka Akira special")'.
Cf. http://www.geocities.jp/kosako3/oniguruma/doc/RE.txt #9

Yeah, perhaps it makes sense to try to follow someone else's syntax. Although \g really adds some mess around to what already is hard to read.
Thanks, i'll keep it in mind and change the subgroups to square brackets thing