1  # encoding: utf-8
   2 
   3  module ActiveSupport #:nodoc:
   4    module Multibyte
   5      # A list of all available normalization forms. See http://www.unicode.org/reports/tr15/tr15-29.html for more
   6      # information about normalization.
   7      NORMALIZATION_FORMS = [:c, :kc, :d, :kd]
   8 
   9      # The Unicode version that is supported by the implementation
  10      UNICODE_VERSION = '5.1.0'
  11 
  12      # The default normalization used for operations that require normalization. It can be set to any of the
  13      # normalizations in NORMALIZATION_FORMS.
  14      #
  15      # Example:
  16      #   ActiveSupport::Multibyte.default_normalization_form = :c
  17      mattr_accessor :default_normalization_form
  18      self.default_normalization_form = :kc
  19 
  20      # The proxy class returned when calling mb_chars. You can use this accessor to configure your own proxy
  21      # class so you can support other encodings. See the ActiveSupport::Multibyte::Chars implementation for
  22      # an example how to do this.
  23      #
  24      # Example:
  25      #   ActiveSupport::Multibyte.proxy_class = CharsForUTF32
  26      def self.proxy_class=(klass)
  27        @proxy_class = klass
  28      end
  29 
  30      # Returns the currect proxy class
  31      def self.proxy_class
  32        @proxy_class ||= ActiveSupport::Multibyte::Chars
  33      end
  34 
  35      # Regular expressions that describe valid byte sequences for a character
  36      VALID_CHARACTER = {
  37        # Borrowed from the Kconv library by Shinji KONO - (also as seen on the W3C site)
  38        'UTF-8' => /\A(?:
  39                    [\x00-\x7f]                                         |
  40                    [\xc2-\xdf] [\x80-\xbf]                             |
  41                    \xe0        [\xa0-\xbf] [\x80-\xbf]                 |
  42                    [\xe1-\xef] [\x80-\xbf] [\x80-\xbf]                 |
  43                    \xf0        [\x90-\xbf] [\x80-\xbf] [\x80-\xbf]     |
  44                    [\xf1-\xf3] [\x80-\xbf] [\x80-\xbf] [\x80-\xbf]     |
  45                    \xf4        [\x80-\x8f] [\x80-\xbf] [\x80-\xbf])\z /xn,
  46        # Quick check for valid Shift-JIS characters, disregards the odd-even pairing
  47        'Shift_JIS' => /\A(?:
  48                    [\x00-\x7e \xa1-\xdf]                                     |
  49                    [\x81-\x9f \xe0-\xef] [\x40-\x7e \x80-\x9e \x9f-\xfc])\z /xn
  50      }
  51    end
  52  end
  53 
  54  require 'active_support/multibyte/chars'
  55  require 'active_support/multibyte/exceptions'
  56  require 'active_support/multibyte/unicode_database'
  57  require 'active_support/multibyte/utils'