activesupport 3.0.0.beta4 → 3.0.pre

data/lib/active_support/json/backends/jsongem.rb

@@ -23,18 +23,15 @@ module ActiveSupport
       private
         def convert_dates_from(data)
           case data
-          when nil
-            nil
-          when DATE_REGEX
-            DateTime.parse(data)
-          when Array
-            data.map! { |d| convert_dates_from(d) }
-          when Hash
-            data.each do |key, value|
-              data[key] = convert_dates_from(value)
-            end
-          else
-            data
+            when DATE_REGEX
+              DateTime.parse(data)
+            when Array
+              data.map! { |d| convert_dates_from(d) }
+            when Hash
+              data.each do |key, value|
+                data[key] = convert_dates_from(value)
+              end
+            else data
           end
         end
       end

data/lib/active_support/json/decoding.rb

@@ -6,15 +6,12 @@ module ActiveSupport
   mattr_accessor :parse_json_times
 
   module JSON
-    # Listed in order of preference.
-    DECODERS = %w(Yajl Yaml)
-
     class << self
       attr_reader :parse_error
       delegate :decode, :to => :backend
 
       def backend
-        set_default_backend unless defined?(@backend)
+        self.backend = "Yaml" unless defined?(@backend)
         @backend
       end
 
@@ -34,18 +31,6 @@ module ActiveSupport
       ensure
         self.backend = old_backend
       end
-
-      def set_default_backend
-        DECODERS.find do |name|
-          begin
-            self.backend = name
-            true
-          rescue LoadError
-            # Try next decoder.
-            false
-          end
-        end
-      end
     end
   end
 end

data/lib/active_support/json/encoding.rb

@@ -1,14 +1,15 @@
 # encoding: utf-8
-require 'bigdecimal'
 require 'active_support/core_ext/array/wrap'
-require 'active_support/core_ext/big_decimal/conversions' # for #to_s
 require 'active_support/core_ext/hash/except'
 require 'active_support/core_ext/hash/slice'
 require 'active_support/core_ext/module/delegation'
 require 'active_support/core_ext/object/instance_variables'
 require 'active_support/deprecation'
 
-require 'active_support/time'
+require 'active_support/core_ext/date_time/conversions'
+require 'active_support/core_ext/time/conversions'
+require 'active_support/time_with_zone'
+require 'active_support/values/time_zone'
 
 # Hack to load json gem first so we can overwrite its to_json.
 begin
@@ -67,15 +68,6 @@ module ActiveSupport
 
 
       ESCAPED_CHARS = {
-        "\x00" => '\u0000', "\x01" => '\u0001', "\x02" => '\u0002',
-        "\x03" => '\u0003', "\x04" => '\u0004', "\x05" => '\u0005',
-        "\x06" => '\u0006', "\x07" => '\u0007', "\x0B" => '\u000B',
-        "\x0E" => '\u000E', "\x0F" => '\u000F', "\x10" => '\u0010',
-        "\x11" => '\u0011', "\x12" => '\u0012', "\x13" => '\u0013',
-        "\x14" => '\u0014', "\x15" => '\u0015', "\x16" => '\u0016',
-        "\x17" => '\u0017', "\x18" => '\u0018', "\x19" => '\u0019',
-        "\x1A" => '\u001A', "\x1B" => '\u001B', "\x1C" => '\u001C',
-        "\x1D" => '\u001D', "\x1E" => '\u001E', "\x1F" => '\u001F',
         "\010" =>  '\b',
         "\f"   =>  '\f',
         "\n"   =>  '\n',
@@ -97,16 +89,14 @@ module ActiveSupport
         def escape_html_entities_in_json=(value)
           self.escape_regex = \
             if @escape_html_entities_in_json = value
-              /[\x00-\x1F"\\><&]/
+              /[\010\f\n\r\t"\\><&]/
             else
-              /[\x00-\x1F"\\]/
+              /[\010\f\n\r\t"\\]/
             end
         end
 
         def escape(string)
-          if string.respond_to?(:force_encoding)
-            string = string.encode(::Encoding::UTF_8, :undef => :replace).force_encoding(::Encoding::BINARY)
-          end
+          string = string.dup.force_encoding(::Encoding::BINARY) if string.respond_to?(:force_encoding)
           json = string.
             gsub(escape_regex) { |s| ESCAPED_CHARS[s] }.
             gsub(/([\xC0-\xDF][\x80-\xBF]|
@@ -114,14 +104,11 @@ module ActiveSupport
                    [\xF0-\xF7][\x80-\xBF]{3})+/nx) { |s|
             s.unpack("U*").pack("n*").unpack("H*")[0].gsub(/.{4}/n, '\\\\u\&')
           }
-          json = %("#{json}")
-          json.force_encoding(::Encoding::UTF_8) if json.respond_to?(:force_encoding)
-          json
+          %("#{json}")
         end
       end
 
-      self.use_standard_json_time_format = true
-      self.escape_html_entities_in_json  = false
+      self.escape_html_entities_in_json = true
     end
 
     CircularReferenceError = Deprecation::DeprecatedConstantProxy.new('ActiveSupport::JSON::CircularReferenceError', Encoding::CircularReferenceError)
@@ -134,13 +121,7 @@ class Object
     ActiveSupport::JSON.encode(self, options)
   end
 
-  def as_json(options = nil) #:nodoc:
-    if respond_to?(:to_hash)
-      to_hash
-    else
-      instance_values
-    end
-  end
+  def as_json(options = nil) instance_values end #:nodoc:
 end
 
 # A string that returns itself as its JSON-encoded form.
@@ -178,20 +159,9 @@ class Numeric
   def encode_json(encoder) to_s end #:nodoc:
 end
 
-class BigDecimal
-  # A BigDecimal would be naturally represented as a JSON number. Most libraries,
-  # however, parse non-integer JSON numbers directly as floats. Clients using
-  # those libraries would get in general a wrong number and no way to recover
-  # other than manually inspecting the string with the JSON code itself.
-  #
-  # That's why a JSON string is returned. The JSON literal is not numeric, but if
-  # the other end knows by contract that the data is supposed to be a BigDecimal,
-  # it still has the chance to post-process the string and get the real value.
-  def as_json(options = nil) to_s end #:nodoc:
-end
-
 class Regexp
-  def as_json(options = nil) to_s end #:nodoc:
+  def as_json(options = nil) self end #:nodoc:
+  def encode_json(encoder) inspect end #:nodoc:
 end
 
 module Enumerable

data/lib/active_support/locale/en.yml

@@ -15,10 +15,7 @@ en:
     month_names: [~, January, February, March, April, May, June, July, August, September, October, November, December]
     abbr_month_names: [~, Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec]
     # Used in date_select and datime_select.
-    order:
-      - :year
-      - :month
-      - :day
+    order: [ :year, :month, :day ]
 
   time:
     formats:

data/lib/active_support/memoizable.rb

@@ -1,4 +1,4 @@
-require 'active_support/core_ext/kernel/singleton_class'
+require 'active_support/core_ext/object/metaclass'
 require 'active_support/core_ext/module/aliasing'
 
 module ActiveSupport

data/lib/active_support/message_encryptor.rb

@@ -1,5 +1,4 @@
 require 'openssl'
-require 'active_support/base64'
 
 module ActiveSupport
   # MessageEncryptor is a simple way to encrypt values which get stored somewhere 

data/lib/active_support/message_verifier.rb

@@ -1,5 +1,4 @@
-require 'active_support/base64'
-require 'active_support/core_ext/object/blank'
+require 'active_support/core_ext/string/bytesize'
 
 module ActiveSupport
   # MessageVerifier makes it easy to generate and verify messages which are signed
@@ -47,11 +46,11 @@ module ActiveSupport
       def secure_compare(a, b)
         return false unless a.bytesize == b.bytesize
 
-        l = a.unpack "C*"
+        l = a.unpack "C#{a.bytesize}"
 
-        res = true
-        b.each_byte { |byte| res = (byte == l.shift) && res }
-        res
+        res = 0
+        b.each_byte { |byte| res |= byte ^ l.shift }
+        res == 0
       end
 
       def generate_digest(data)

data/lib/active_support/multibyte.rb

@@ -1,12 +1,24 @@
 # encoding: utf-8
+
 require 'active_support/core_ext/module/attribute_accessors'
 
 module ActiveSupport #:nodoc:
   module Multibyte
-    autoload :EncodingError, 'active_support/multibyte/exceptions'
-    autoload :Chars, 'active_support/multibyte/chars'
-    autoload :Unicode, 'active_support/multibyte/unicode'
-    
+    # A list of all available normalization forms. See http://www.unicode.org/reports/tr15/tr15-29.html for more
+    # information about normalization.
+    NORMALIZATION_FORMS = [:c, :kc, :d, :kd]
+
+    # The Unicode version that is supported by the implementation
+    UNICODE_VERSION = '5.1.0'
+
+    # The default normalization used for operations that require normalization. It can be set to any of the
+    # normalizations in NORMALIZATION_FORMS.
+    #
+    # Example:
+    #   ActiveSupport::Multibyte.default_normalization_form = :c
+    mattr_accessor :default_normalization_form
+    self.default_normalization_form = :kc
+
     # The proxy class returned when calling mb_chars. You can use this accessor to configure your own proxy
     # class so you can support other encodings. See the ActiveSupport::Multibyte::Chars implementation for
     # an example how to do this.
@@ -35,10 +47,13 @@ module ActiveSupport #:nodoc:
                   \xf4        [\x80-\x8f] [\x80-\xbf] [\x80-\xbf])\z /xn,
       # Quick check for valid Shift-JIS characters, disregards the odd-even pairing
       'Shift_JIS' => /\A(?:
-                  [\x00-\x7e\xa1-\xdf]                                     |
-                  [\x81-\x9f\xe0-\xef] [\x40-\x7e\x80-\x9e\x9f-\xfc])\z /xn
+                  [\x00-\x7e \xa1-\xdf]                                     |
+                  [\x81-\x9f \xe0-\xef] [\x40-\x7e \x80-\x9e \x9f-\xfc])\z /xn
     }
   end
 end
 
-require 'active_support/multibyte/utils'
+require 'active_support/multibyte/chars'
+require 'active_support/multibyte/exceptions'
+require 'active_support/multibyte/unicode_database'
+require 'active_support/multibyte/utils'

data/lib/active_support/multibyte/chars.rb

@@ -1,5 +1,4 @@
 # encoding: utf-8
-require 'active_support/core_ext/string/access'
 require 'active_support/core_ext/string/behavior'
 
 module ActiveSupport #:nodoc:
@@ -19,7 +18,7 @@ module ActiveSupport #:nodoc:
     #   bad.explicit_checking_method "T".mb_chars.downcase.to_s
     #
     # The default Chars implementation assumes that the encoding of the string is UTF-8, if you want to handle different
-    # encodings you can write your own multibyte string handler and configure it through
+    # encodings you can write your own multibyte string handler and configure it through 
     # ActiveSupport::Multibyte.proxy_class.
     #
     #   class CharsForUTF32
@@ -34,19 +33,59 @@ module ActiveSupport #:nodoc:
     #
     #   ActiveSupport::Multibyte.proxy_class = CharsForUTF32
     class Chars
+      # Hangul character boundaries and properties
+      HANGUL_SBASE = 0xAC00
+      HANGUL_LBASE = 0x1100
+      HANGUL_VBASE = 0x1161
+      HANGUL_TBASE = 0x11A7
+      HANGUL_LCOUNT = 19
+      HANGUL_VCOUNT = 21
+      HANGUL_TCOUNT = 28
+      HANGUL_NCOUNT = HANGUL_VCOUNT * HANGUL_TCOUNT
+      HANGUL_SCOUNT = 11172
+      HANGUL_SLAST = HANGUL_SBASE + HANGUL_SCOUNT
+      HANGUL_JAMO_FIRST = 0x1100
+      HANGUL_JAMO_LAST = 0x11FF
+
+      # All the unicode whitespace
+      UNICODE_WHITESPACE = [
+        (0x0009..0x000D).to_a, # White_Space # Cc   [5] <control-0009>..<control-000D>
+        0x0020,                # White_Space # Zs       SPACE
+        0x0085,                # White_Space # Cc       <control-0085>
+        0x00A0,                # White_Space # Zs       NO-BREAK SPACE
+        0x1680,                # White_Space # Zs       OGHAM SPACE MARK
+        0x180E,                # White_Space # Zs       MONGOLIAN VOWEL SEPARATOR
+        (0x2000..0x200A).to_a, # White_Space # Zs  [11] EN QUAD..HAIR SPACE
+        0x2028,                # White_Space # Zl       LINE SEPARATOR
+        0x2029,                # White_Space # Zp       PARAGRAPH SEPARATOR
+        0x202F,                # White_Space # Zs       NARROW NO-BREAK SPACE
+        0x205F,                # White_Space # Zs       MEDIUM MATHEMATICAL SPACE
+        0x3000,                # White_Space # Zs       IDEOGRAPHIC SPACE
+      ].flatten.freeze
+
+      # BOM (byte order mark) can also be seen as whitespace, it's a non-rendering character used to distinguish
+      # between little and big endian. This is not an issue in utf-8, so it must be ignored.
+      UNICODE_LEADERS_AND_TRAILERS = UNICODE_WHITESPACE + [65279] # ZERO-WIDTH NO-BREAK SPACE aka BOM
+
+      # Returns a regular expression pattern that matches the passed Unicode codepoints
+      def self.codepoints_to_pattern(array_of_codepoints) #:nodoc:
+        array_of_codepoints.collect{ |e| [e].pack 'U*' }.join('|')
+      end
+      UNICODE_TRAILERS_PAT = /(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+\Z/
+      UNICODE_LEADERS_PAT = /\A(#{codepoints_to_pattern(UNICODE_LEADERS_AND_TRAILERS)})+/
+
+      UTF8_PAT = ActiveSupport::Multibyte::VALID_CHARACTER['UTF-8']
 
       attr_reader :wrapped_string
       alias to_s wrapped_string
       alias to_str wrapped_string
 
-      if RUBY_VERSION >= "1.9"
+      if '1.9'.respond_to?(:force_encoding)
         # Creates a new Chars instance by wrapping _string_.
         def initialize(string)
           @wrapped_string = string
           @wrapped_string.force_encoding(Encoding::UTF_8) unless @wrapped_string.frozen?
         end
-
-        undef <=>
       else
         def initialize(string) #:nodoc:
           @wrapped_string = string
@@ -75,6 +114,12 @@ module ActiveSupport #:nodoc:
         true
       end
 
+      # Returns +true+ if the Chars class can and should act as a proxy for the string _string_. Returns
+      # +false+ otherwise.
+      def self.wants?(string)
+        $KCODE == 'UTF8' && consumes?(string)
+      end
+
       # Returns +true+ when the proxy class can handle the string. Returns +false+ otherwise.
       def self.consumes?(string)
         # Unpack is a little bit faster than regular expressions.
@@ -86,122 +131,30 @@ module ActiveSupport #:nodoc:
 
       include Comparable
 
-      if RUBY_VERSION < "1.9"
-        # Returns +true+ if the Chars class can and should act as a proxy for the string _string_. Returns
-        # +false+ otherwise.
-        def self.wants?(string)
-          $KCODE == 'UTF8' && consumes?(string)
-        end
-
-        # Returns <tt>-1</tt>, <tt>0</tt> or <tt>+1</tt> depending on whether the Chars object is to be sorted before,
-        # equal or after the object on the right side of the operation. It accepts any object that implements +to_s+.
-        # See <tt>String#<=></tt> for more details.
-        #
-        # Example:
-        #   'é'.mb_chars <=> 'ü'.mb_chars #=> -1
-        def <=>(other)
-          @wrapped_string <=> other.to_s
-        end
-
-        # Returns a new Chars object containing the _other_ object concatenated to the string.
-        #
-        # Example:
-        #   ('Café'.mb_chars + ' périferôl').to_s #=> "Café périferôl"
-        def +(other)
-          self << other
-        end
-
-        # Like <tt>String#=~</tt> only it returns the character offset (in codepoints) instead of the byte offset.
-        #
-        # Example:
-        #   'Café périferôl'.mb_chars =~ /ô/ #=> 12
-        def =~(other)
-          translate_offset(@wrapped_string =~ other)
-        end
-
-        # Inserts the passed string at specified codepoint offsets.
-        #
-        # Example:
-        #   'Café'.mb_chars.insert(4, ' périferôl').to_s #=> "Café périferôl"
-        def insert(offset, fragment)
-          unpacked = Unicode.u_unpack(@wrapped_string)
-          unless offset > unpacked.length
-            @wrapped_string.replace(
-              Unicode.u_unpack(@wrapped_string).insert(offset, *Unicode.u_unpack(fragment)).pack('U*')
-            )
-          else
-            raise IndexError, "index #{offset} out of string"
-          end
-          self
-        end
-
-        # Returns +true+ if contained string contains _other_. Returns +false+ otherwise.
-        #
-        # Example:
-        #   'Café'.mb_chars.include?('é') #=> true
-        def include?(other)
-          # We have to redefine this method because Enumerable defines it.
-          @wrapped_string.include?(other)
-        end
-
-        # Returns the position _needle_ in the string, counting in codepoints. Returns +nil+ if _needle_ isn't found.
-        #
-        # Example:
-        #   'Café périferôl'.mb_chars.index('ô') #=> 12
-        #   'Café périferôl'.mb_chars.index(/\w/u) #=> 0
-        def index(needle, offset=0)
-          wrapped_offset = first(offset).wrapped_string.length
-          index = @wrapped_string.index(needle, wrapped_offset)
-          index ? (Unicode.u_unpack(@wrapped_string.slice(0...index)).size) : nil
-        end
-
-        # Returns the position _needle_ in the string, counting in
-        # codepoints, searching backward from _offset_ or the end of the
-        # string. Returns +nil+ if _needle_ isn't found.
-        #
-        # Example:
-        #   'Café périferôl'.mb_chars.rindex('é') #=> 6
-        #   'Café périferôl'.mb_chars.rindex(/\w/u) #=> 13
-        def rindex(needle, offset=nil)
-          offset ||= length
-          wrapped_offset = first(offset).wrapped_string.length
-          index = @wrapped_string.rindex(needle, wrapped_offset)
-          index ? (Unicode.u_unpack(@wrapped_string.slice(0...index)).size) : nil
-        end
-
-        # Returns the number of codepoints in the string
-        def size
-          Unicode.u_unpack(@wrapped_string).size
-        end
-        alias_method :length, :size
-
-        # Strips entire range of Unicode whitespace from the right of the string.
-        def rstrip
-          chars(@wrapped_string.gsub(Unicode::TRAILERS_PAT, ''))
-        end
-
-        # Strips entire range of Unicode whitespace from the left of the string.
-        def lstrip
-          chars(@wrapped_string.gsub(Unicode::LEADERS_PAT, ''))
-        end
-
-        # Strips entire range of Unicode whitespace from the right and left of the string.
-        def strip
-          rstrip.lstrip
-        end
+      # Returns <tt>-1</tt>, <tt>0</tt> or <tt>+1</tt> depending on whether the Chars object is to be sorted before,
+      # equal or after the object on the right side of the operation. It accepts any object that implements +to_s+.
+      # See <tt>String#<=></tt> for more details.
+      #
+      # Example:
+      #   'é'.mb_chars <=> 'ü'.mb_chars #=> -1
+      def <=>(other)
+        @wrapped_string <=> other.to_s
+      end
 
-        # Returns the codepoint of the first character in the string.
-        #
-        # Example:
-        #   'こんにちは'.mb_chars.ord #=> 12371
-        def ord
-          Unicode.u_unpack(@wrapped_string)[0]
-        end
+      # Returns a new Chars object containing the _other_ object concatenated to the string.
+      #
+      # Example:
+      #   ('Café'.mb_chars + ' périferôl').to_s #=> "Café périferôl"
+      def +(other)
+        self << other
+      end
 
-      else
-        def =~(other)
-          @wrapped_string =~ other
-        end
+      # Like <tt>String#=~</tt> only it returns the character offset (in codepoints) instead of the byte offset.
+      #
+      # Example:
+      #   'Café périferôl'.mb_chars =~ /ô/ #=> 12
+      def =~(other)
+        translate_offset(@wrapped_string =~ other)
       end
 
       # Works just like <tt>String#split</tt>, with the exception that the items in the resulting list are Chars
@@ -213,6 +166,56 @@ module ActiveSupport #:nodoc:
         @wrapped_string.split(*args).map { |i| i.mb_chars }
       end
 
+      # Inserts the passed string at specified codepoint offsets.
+      #
+      # Example:
+      #   'Café'.mb_chars.insert(4, ' périferôl').to_s #=> "Café périferôl"
+      def insert(offset, fragment)
+        unpacked = self.class.u_unpack(@wrapped_string)
+        unless offset > unpacked.length
+          @wrapped_string.replace(
+            self.class.u_unpack(@wrapped_string).insert(offset, *self.class.u_unpack(fragment)).pack('U*')
+          )
+        else
+          raise IndexError, "index #{offset} out of string"
+        end
+        self
+      end
+
+      # Returns +true+ if contained string contains _other_. Returns +false+ otherwise.
+      #
+      # Example:
+      #   'Café'.mb_chars.include?('é') #=> true
+      def include?(other)
+        # We have to redefine this method because Enumerable defines it.
+        @wrapped_string.include?(other)
+      end
+
+      # Returns the position _needle_ in the string, counting in codepoints. Returns +nil+ if _needle_ isn't found.
+      #
+      # Example:
+      #   'Café périferôl'.mb_chars.index('ô') #=> 12
+      #   'Café périferôl'.mb_chars.index(/\w/u) #=> 0
+      def index(needle, offset=0)
+        wrapped_offset = self.first(offset).wrapped_string.length
+        index = @wrapped_string.index(needle, wrapped_offset)
+        index ? (self.class.u_unpack(@wrapped_string.slice(0...index)).size) : nil
+      end
+
+      # Returns the position _needle_ in the string, counting in
+      # codepoints, searching backward from _offset_ or the end of the
+      # string. Returns +nil+ if _needle_ isn't found.
+      #
+      # Example:
+      #   'Café périferôl'.mb_chars.rindex('é') #=> 6
+      #   'Café périferôl'.mb_chars.rindex(/\w/u) #=> 13
+      def rindex(needle, offset=nil)
+        offset ||= length
+        wrapped_offset = self.first(offset).wrapped_string.length
+        index = @wrapped_string.rindex(needle, wrapped_offset)
+        index ? (self.class.u_unpack(@wrapped_string.slice(0...index)).size) : nil
+      end
+
       # Like <tt>String#[]=</tt>, except instead of byte offsets you specify character offsets.
       #
       # Example:
@@ -232,7 +235,7 @@ module ActiveSupport #:nodoc:
         if args.first.is_a?(Regexp)
           @wrapped_string[*args] = replace_by
         else
-          result = Unicode.u_unpack(@wrapped_string)
+          result = self.class.u_unpack(@wrapped_string)
           if args[0].is_a?(Fixnum)
             raise IndexError, "index #{args[0]} out of string" if args[0] >= result.length
             min = args[0]
@@ -245,10 +248,10 @@ module ActiveSupport #:nodoc:
           else
             needle = args[0].to_s
             min = index(needle)
-            max = min + Unicode.u_unpack(needle).length - 1
+            max = min + self.class.u_unpack(needle).length - 1
             range = Range.new(min, max)
           end
-          result[range] = Unicode.u_unpack(replace_by)
+          result[range] = self.class.u_unpack(replace_by)
           @wrapped_string.replace(result.pack('U*'))
         end
       end
@@ -292,13 +295,33 @@ module ActiveSupport #:nodoc:
         justify(integer, :center, padstr)
       end
 
+      # Strips entire range of Unicode whitespace from the right of the string.
+      def rstrip
+        chars(@wrapped_string.gsub(UNICODE_TRAILERS_PAT, ''))
+      end
+
+      # Strips entire range of Unicode whitespace from the left of the string.
+      def lstrip
+        chars(@wrapped_string.gsub(UNICODE_LEADERS_PAT, ''))
+      end
+
+      # Strips entire range of Unicode whitespace from the right and left of the string.
+      def strip
+        rstrip.lstrip
+      end
+
+      # Returns the number of codepoints in the string
+      def size
+        self.class.u_unpack(@wrapped_string).size
+      end
+      alias_method :length, :size
 
       # Reverses all characters in the string.
       #
       # Example:
       #   'Café'.mb_chars.reverse.to_s #=> 'éfaC'
       def reverse
-        chars(Unicode.g_unpack(@wrapped_string).reverse.flatten.pack('U*'))
+        chars(self.class.u_unpack(@wrapped_string).reverse.pack('U*'))
       end
 
       # Implements Unicode-aware slice with codepoints. Slicing on one point returns the codepoints for that
@@ -314,15 +337,15 @@ module ActiveSupport #:nodoc:
         elsif (args.size == 2 && !args[1].is_a?(Numeric))
           raise TypeError, "cannot convert #{args[1].class} into Integer" # Do as if we were native
         elsif args[0].kind_of? Range
-          cps = Unicode.u_unpack(@wrapped_string).slice(*args)
+          cps = self.class.u_unpack(@wrapped_string).slice(*args)
           result = cps.nil? ? nil : cps.pack('U*')
         elsif args[0].kind_of? Regexp
           result = @wrapped_string.slice(*args)
         elsif args.size == 1 && args[0].kind_of?(Numeric)
-          character = Unicode.u_unpack(@wrapped_string)[args[0]]
+          character = self.class.u_unpack(@wrapped_string)[args[0]]
           result = character.nil? ? nil : [character].pack('U')
         else
-          result = Unicode.u_unpack(@wrapped_string).slice(*args).pack('U*')
+          result = self.class.u_unpack(@wrapped_string).slice(*args).pack('U*')
         end
         result.nil? ? nil : chars(result)
       end
@@ -340,22 +363,20 @@ module ActiveSupport #:nodoc:
         slice
       end
 
-      # Limit the byte size of the string to a number of bytes without breaking characters. Usable
-      # when the storage for a string is limited for some reason.
+      # Returns the codepoint of the first character in the string.
       #
       # Example:
-      #   s = 'こんにちは'
-      #   s.mb_chars.limit(7) #=> "こに"
-      def limit(limit)
-        slice(0...translate_offset(limit))
+      #   'こんにちは'.mb_chars.ord #=> 12371
+      def ord
+        self.class.u_unpack(@wrapped_string)[0]
       end
 
       # Convert characters in the string to uppercase.
       #
       # Example:
-      #   'Laurent, où sont les tests ?'.mb_chars.upcase.to_s #=> "LAURENT, OÙ SONT LES TESTS ?"
+      #   'Laurent, òu sont les tests?'.mb_chars.upcase.to_s #=> "LAURENT, ÒU SONT LES TESTS?"
       def upcase
-        chars(Unicode.apply_mapping @wrapped_string, :uppercase_mapping)
+        apply_mapping :uppercase_mapping
       end
 
       # Convert characters in the string to lowercase.
@@ -363,7 +384,7 @@ module ActiveSupport #:nodoc:
       # Example:
       #   'VĚDA A VÝZKUM'.mb_chars.downcase.to_s #=> "věda a výzkum"
       def downcase
-        chars(Unicode.apply_mapping @wrapped_string, :lowercase_mapping)
+        apply_mapping :lowercase_mapping
       end
 
       # Converts the first character to uppercase and the remainder to lowercase.
@@ -377,11 +398,25 @@ module ActiveSupport #:nodoc:
       # Returns the KC normalization of the string by default. NFKC is considered the best normalization form for
       # passing strings to databases and validations.
       #
+      # * <tt>str</tt> - The string to perform normalization on.
       # * <tt>form</tt> - The form you want to normalize in. Should be one of the following:
       #   <tt>:c</tt>, <tt>:kc</tt>, <tt>:d</tt>, or <tt>:kd</tt>. Default is
-      #   ActiveSupport::Multibyte::Unicode.default_normalization_form
-      def normalize(form = nil)
-        chars(Unicode.normalize(@wrapped_string, form))
+      #   ActiveSupport::Multibyte.default_normalization_form
+      def normalize(form=ActiveSupport::Multibyte.default_normalization_form)
+        # See http://www.unicode.org/reports/tr15, Table 1
+        codepoints = self.class.u_unpack(@wrapped_string)
+        chars(case form
+          when :d
+            self.class.reorder_characters(self.class.decompose_codepoints(:canonical, codepoints))
+          when :c
+            self.class.compose_codepoints(self.class.reorder_characters(self.class.decompose_codepoints(:canonical, codepoints)))
+          when :kd
+            self.class.reorder_characters(self.class.decompose_codepoints(:compatability, codepoints))
+          when :kc
+            self.class.compose_codepoints(self.class.reorder_characters(self.class.decompose_codepoints(:compatability, codepoints)))
+          else
+            raise ArgumentError, "#{form} is not a valid normalization variant", caller
+        end.pack('U*'))
       end
 
       # Performs canonical decomposition on all the characters.
@@ -390,7 +425,7 @@ module ActiveSupport #:nodoc:
       #   'é'.length #=> 2
       #   'é'.mb_chars.decompose.to_s.length #=> 3
       def decompose
-        chars(Unicode.decompose_codepoints(:canonical, Unicode.u_unpack(@wrapped_string)).pack('U*'))
+        chars(self.class.decompose_codepoints(:canonical, self.class.u_unpack(@wrapped_string)).pack('U*'))
       end
 
       # Performs composition on all the characters.
@@ -399,7 +434,7 @@ module ActiveSupport #:nodoc:
       #   'é'.length #=> 3
       #   'é'.mb_chars.compose.to_s.length #=> 2
       def compose
-        chars(Unicode.compose_codepoints(Unicode.u_unpack(@wrapped_string)).pack('U*'))
+        chars(self.class.compose_codepoints(self.class.u_unpack(@wrapped_string)).pack('U*'))
       end
 
       # Returns the number of grapheme clusters in the string.
@@ -408,14 +443,12 @@ module ActiveSupport #:nodoc:
       #   'क्षि'.mb_chars.length #=> 4
       #   'क्षि'.mb_chars.g_length #=> 3
       def g_length
-        Unicode.g_unpack(@wrapped_string).length
+        self.class.g_unpack(@wrapped_string).length
       end
 
       # Replaces all ISO-8859-1 or CP1252 characters by their UTF-8 equivalent resulting in a valid UTF-8 string.
-      #
-      # Passing +true+ will forcibly tidy all bytes, assuming that the string's encoding is entirely CP1252 or ISO-8859-1.
-      def tidy_bytes(force = false)
-        chars(Unicode.tidy_bytes(@wrapped_string, force))
+      def tidy_bytes
+        chars(self.class.tidy_bytes(@wrapped_string))
       end
 
       %w(lstrip rstrip strip reverse upcase downcase tidy_bytes capitalize).each do |method|
@@ -429,21 +462,213 @@ module ActiveSupport #:nodoc:
         end
       end
 
+      class << self
+
+        # Unpack the string at codepoints boundaries. Raises an EncodingError when the encoding of the string isn't
+        # valid UTF-8.
+        #
+        # Example:
+        #   Chars.u_unpack('Café') #=> [67, 97, 102, 233]
+        def u_unpack(string)
+          begin
+            string.unpack 'U*'
+          rescue ArgumentError
+            raise EncodingError, 'malformed UTF-8 character'
+          end
+        end
+
+        # Detect whether the codepoint is in a certain character class. Returns +true+ when it's in the specified
+        # character class and +false+ otherwise. Valid character classes are: <tt>:cr</tt>, <tt>:lf</tt>, <tt>:l</tt>,
+        # <tt>:v</tt>, <tt>:lv</tt>, <tt>:lvt</tt> and <tt>:t</tt>.
+        #
+        # Primarily used by the grapheme cluster support.
+        def in_char_class?(codepoint, classes)
+          classes.detect { |c| UCD.boundary[c] === codepoint } ? true : false
+        end
+
+        # Unpack the string at grapheme boundaries. Returns a list of character lists.
+        #
+        # Example:
+        #   Chars.g_unpack('क्षि') #=> [[2325, 2381], [2359], [2367]]
+        #   Chars.g_unpack('Café') #=> [[67], [97], [102], [233]]
+        def g_unpack(string)
+          codepoints = u_unpack(string)
+          unpacked = []
+          pos = 0
+          marker = 0
+          eoc = codepoints.length
+          while(pos < eoc)
+            pos += 1
+            previous = codepoints[pos-1]
+            current = codepoints[pos]
+            if (
+                # CR X LF
+                one = ( previous == UCD.boundary[:cr] and current == UCD.boundary[:lf] ) or
+                # L X (L|V|LV|LVT)
+                two = ( UCD.boundary[:l] === previous and in_char_class?(current, [:l,:v,:lv,:lvt]) ) or
+                # (LV|V) X (V|T)
+                three = ( in_char_class?(previous, [:lv,:v]) and in_char_class?(current, [:v,:t]) ) or
+                # (LVT|T) X (T)
+                four = ( in_char_class?(previous, [:lvt,:t]) and UCD.boundary[:t] === current ) or
+                # X Extend
+                five = (UCD.boundary[:extend] === current)
+              )
+            else
+              unpacked << codepoints[marker..pos-1]
+              marker = pos
+            end
+          end 
+          unpacked
+        end
+
+        # Reverse operation of g_unpack.
+        #
+        # Example:
+        #   Chars.g_pack(Chars.g_unpack('क्षि')) #=> 'क्षि'
+        def g_pack(unpacked)
+          (unpacked.flatten).pack('U*')
+        end
+
+        def padding(padsize, padstr=' ') #:nodoc:
+          if padsize != 0
+            new(padstr * ((padsize / u_unpack(padstr).size) + 1)).slice(0, padsize)
+          else
+            ''
+          end
+        end
+
+        # Re-order codepoints so the string becomes canonical.
+        def reorder_characters(codepoints)
+          length = codepoints.length- 1
+          pos = 0
+          while pos < length do
+            cp1, cp2 = UCD.codepoints[codepoints[pos]], UCD.codepoints[codepoints[pos+1]]
+            if (cp1.combining_class > cp2.combining_class) && (cp2.combining_class > 0)
+              codepoints[pos..pos+1] = cp2.code, cp1.code
+              pos += (pos > 0 ? -1 : 1)
+            else
+              pos += 1
+            end
+          end
+          codepoints
+        end
+
+        # Decompose composed characters to the decomposed form.
+        def decompose_codepoints(type, codepoints)
+          codepoints.inject([]) do |decomposed, cp|
+            # if it's a hangul syllable starter character
+            if HANGUL_SBASE <= cp and cp < HANGUL_SLAST
+              sindex = cp - HANGUL_SBASE
+              ncp = [] # new codepoints
+              ncp << HANGUL_LBASE + sindex / HANGUL_NCOUNT
+              ncp << HANGUL_VBASE + (sindex % HANGUL_NCOUNT) / HANGUL_TCOUNT
+              tindex = sindex % HANGUL_TCOUNT
+              ncp << (HANGUL_TBASE + tindex) unless tindex == 0
+              decomposed.concat ncp
+            # if the codepoint is decomposable in with the current decomposition type
+            elsif (ncp = UCD.codepoints[cp].decomp_mapping) and (!UCD.codepoints[cp].decomp_type || type == :compatability)
+              decomposed.concat decompose_codepoints(type, ncp.dup)
+            else
+              decomposed << cp
+            end
+          end
+        end
+
+        # Compose decomposed characters to the composed form.
+        def compose_codepoints(codepoints)
+          pos = 0
+          eoa = codepoints.length - 1
+          starter_pos = 0
+          starter_char = codepoints[0]
+          previous_combining_class = -1
+          while pos < eoa
+            pos += 1
+            lindex = starter_char - HANGUL_LBASE
+            # -- Hangul
+            if 0 <= lindex and lindex < HANGUL_LCOUNT
+              vindex = codepoints[starter_pos+1] - HANGUL_VBASE rescue vindex = -1
+              if 0 <= vindex and vindex < HANGUL_VCOUNT
+                tindex = codepoints[starter_pos+2] - HANGUL_TBASE rescue tindex = -1
+                if 0 <= tindex and tindex < HANGUL_TCOUNT
+                  j = starter_pos + 2
+                  eoa -= 2
+                else
+                  tindex = 0
+                  j = starter_pos + 1
+                  eoa -= 1
+                end
+                codepoints[starter_pos..j] = (lindex * HANGUL_VCOUNT + vindex) * HANGUL_TCOUNT + tindex + HANGUL_SBASE
+              end
+              starter_pos += 1
+              starter_char = codepoints[starter_pos]
+            # -- Other characters
+            else
+              current_char = codepoints[pos]
+              current = UCD.codepoints[current_char]
+              if current.combining_class > previous_combining_class
+                if ref = UCD.composition_map[starter_char]
+                  composition = ref[current_char]
+                else
+                  composition = nil
+                end
+                unless composition.nil?
+                  codepoints[starter_pos] = composition
+                  starter_char = composition
+                  codepoints.delete_at pos
+                  eoa -= 1
+                  pos -= 1
+                  previous_combining_class = -1
+                else
+                  previous_combining_class = current.combining_class
+                end
+              else
+                previous_combining_class = current.combining_class
+              end
+              if current.combining_class == 0
+                starter_pos = pos
+                starter_char = codepoints[pos]
+              end
+            end
+          end
+          codepoints
+        end
+
+        # Replaces all ISO-8859-1 or CP1252 characters by their UTF-8 equivalent resulting in a valid UTF-8 string.
+        def tidy_bytes(string)
+          string.split(//u).map do |c|
+            c.force_encoding(Encoding::ASCII) if c.respond_to?(:force_encoding)
+
+            if !ActiveSupport::Multibyte::VALID_CHARACTER['UTF-8'].match(c)
+              n = c.unpack('C')[0]
+              n < 128 ? n.chr :
+              n < 160 ? [UCD.cp1252[n] || n].pack('U') :
+              n < 192 ? "\xC2" + n.chr : "\xC3" + (n-64).chr
+            else
+              c
+            end
+          end.join
+        end
+      end
+
       protected
 
         def translate_offset(byte_offset) #:nodoc:
           return nil if byte_offset.nil?
           return 0   if @wrapped_string == ''
-
-          if @wrapped_string.respond_to?(:force_encoding)
-            @wrapped_string = @wrapped_string.dup.force_encoding(Encoding::ASCII_8BIT)
-          end
-
+          chunk = @wrapped_string[0..byte_offset]
           begin
-            @wrapped_string[0...byte_offset].unpack('U*').length
-          rescue ArgumentError => e
-            byte_offset -= 1
-            retry
+            begin
+              chunk.unpack('U*').length - 1
+            rescue ArgumentError => e
+              chunk = @wrapped_string[0..(byte_offset+=1)]
+              # Stop retrying at the end of the string
+              raise e unless byte_offset < chunk.length 
+              # We damaged a character, retry
+              retry
+            end
+          # Catch the ArgumentError so we can throw our own
+          rescue ArgumentError 
+            raise EncodingError, 'malformed UTF-8 character'
           end
         end
 
@@ -453,23 +678,26 @@ module ActiveSupport #:nodoc:
           padsize = padsize > 0 ? padsize : 0
           case way
           when :right
-            result = @wrapped_string.dup.insert(0, padding(padsize, padstr))
+            result = @wrapped_string.dup.insert(0, self.class.padding(padsize, padstr))
           when :left
-            result = @wrapped_string.dup.insert(-1, padding(padsize, padstr))
+            result = @wrapped_string.dup.insert(-1, self.class.padding(padsize, padstr))
           when :center
-            lpad = padding((padsize / 2.0).floor, padstr)
-            rpad = padding((padsize / 2.0).ceil, padstr)
+            lpad = self.class.padding((padsize / 2.0).floor, padstr)
+            rpad = self.class.padding((padsize / 2.0).ceil, padstr)
             result = @wrapped_string.dup.insert(0, lpad).insert(-1, rpad)
           end
           chars(result)
         end
 
-        def padding(padsize, padstr=' ') #:nodoc:
-          if padsize != 0
-            chars(padstr * ((padsize / Unicode.u_unpack(padstr).size) + 1)).slice(0, padsize)
-          else
-            ''
-          end
+        def apply_mapping(mapping) #:nodoc:
+          chars(self.class.u_unpack(@wrapped_string).map do |codepoint|
+            cp = UCD.codepoints[codepoint]
+            if cp and (ncp = cp.send(mapping)) and ncp > 0
+              ncp
+            else
+              codepoint
+            end
+          end.pack('U*'))
         end
 
         def chars(string) #:nodoc: