pract6 0.1.0

data/.gem/ruby/2.3.0/gems/activesupport-5.1.4/lib/active_support/message_verifier.rb

@@ -0,0 +1,134 @@
+require "base64"
+require "active_support/core_ext/object/blank"
+require "active_support/security_utils"
+
+module ActiveSupport
+  # +MessageVerifier+ makes it easy to generate and verify messages which are
+  # signed to prevent tampering.
+  #
+  # This is useful for cases like remember-me tokens and auto-unsubscribe links
+  # where the session store isn't suitable or available.
+  #
+  # Remember Me:
+  #   cookies[:remember_me] = @verifier.generate([@user.id, 2.weeks.from_now])
+  #
+  # In the authentication filter:
+  #
+  #   id, time = @verifier.verify(cookies[:remember_me])
+  #   if Time.now < time
+  #     self.current_user = User.find(id)
+  #   end
+  #
+  # By default it uses Marshal to serialize the message. If you want to use
+  # another serialization method, you can set the serializer in the options
+  # hash upon initialization:
+  #
+  #   @verifier = ActiveSupport::MessageVerifier.new('s3Krit', serializer: YAML)
+  #
+  # +MessageVerifier+ creates HMAC signatures using SHA1 hash algorithm by default.
+  # If you want to use a different hash algorithm, you can change it by providing
+  # `:digest` key as an option while initializing the verifier:
+  #
+  #   @verifier = ActiveSupport::MessageVerifier.new('s3Krit', digest: 'SHA256')
+  class MessageVerifier
+    class InvalidSignature < StandardError; end
+
+    def initialize(secret, options = {})
+      raise ArgumentError, "Secret should not be nil." unless secret
+      @secret = secret
+      @digest = options[:digest] || "SHA1"
+      @serializer = options[:serializer] || Marshal
+    end
+
+    # Checks if a signed message could have been generated by signing an object
+    # with the +MessageVerifier+'s secret.
+    #
+    #   verifier = ActiveSupport::MessageVerifier.new 's3Krit'
+    #   signed_message = verifier.generate 'a private message'
+    #   verifier.valid_message?(signed_message) # => true
+    #
+    #   tampered_message = signed_message.chop # editing the message invalidates the signature
+    #   verifier.valid_message?(tampered_message) # => false
+    def valid_message?(signed_message)
+      return if signed_message.nil? || !signed_message.valid_encoding? || signed_message.blank?
+
+      data, digest = signed_message.split("--".freeze)
+      data.present? && digest.present? && ActiveSupport::SecurityUtils.secure_compare(digest, generate_digest(data))
+    end
+
+    # Decodes the signed message using the +MessageVerifier+'s secret.
+    #
+    #   verifier = ActiveSupport::MessageVerifier.new 's3Krit'
+    #
+    #   signed_message = verifier.generate 'a private message'
+    #   verifier.verified(signed_message) # => 'a private message'
+    #
+    # Returns +nil+ if the message was not signed with the same secret.
+    #
+    #   other_verifier = ActiveSupport::MessageVerifier.new 'd1ff3r3nt-s3Krit'
+    #   other_verifier.verified(signed_message) # => nil
+    #
+    # Returns +nil+ if the message is not Base64-encoded.
+    #
+    #   invalid_message = "f--46a0120593880c733a53b6dad75b42ddc1c8996d"
+    #   verifier.verified(invalid_message) # => nil
+    #
+    # Raises any error raised while decoding the signed message.
+    #
+    #   incompatible_message = "test--dad7b06c94abba8d46a15fafaef56c327665d5ff"
+    #   verifier.verified(incompatible_message) # => TypeError: incompatible marshal file format
+    def verified(signed_message)
+      if valid_message?(signed_message)
+        begin
+          data = signed_message.split("--".freeze)[0]
+          @serializer.load(decode(data))
+        rescue ArgumentError => argument_error
+          return if argument_error.message.include?("invalid base64")
+          raise
+        end
+      end
+    end
+
+    # Decodes the signed message using the +MessageVerifier+'s secret.
+    #
+    #   verifier = ActiveSupport::MessageVerifier.new 's3Krit'
+    #   signed_message = verifier.generate 'a private message'
+    #
+    #   verifier.verify(signed_message) # => 'a private message'
+    #
+    # Raises +InvalidSignature+ if the message was not signed with the same
+    # secret or was not Base64-encoded.
+    #
+    #   other_verifier = ActiveSupport::MessageVerifier.new 'd1ff3r3nt-s3Krit'
+    #   other_verifier.verify(signed_message) # => ActiveSupport::MessageVerifier::InvalidSignature
+    def verify(signed_message)
+      verified(signed_message) || raise(InvalidSignature)
+    end
+
+    # Generates a signed message for the provided value.
+    #
+    # The message is signed with the +MessageVerifier+'s secret. Without knowing
+    # the secret, the original value cannot be extracted from the message.
+    #
+    #   verifier = ActiveSupport::MessageVerifier.new 's3Krit'
+    #   verifier.generate 'a private message' # => "BAhJIhRwcml2YXRlLW1lc3NhZ2UGOgZFVA==--e2d724331ebdee96a10fb99b089508d1c72bd772"
+    def generate(value)
+      data = encode(@serializer.dump(value))
+      "#{data}--#{generate_digest(data)}"
+    end
+
+    private
+      def encode(data)
+        ::Base64.strict_encode64(data)
+      end
+
+      def decode(data)
+        ::Base64.strict_decode64(data)
+      end
+
+      def generate_digest(data)
+        require "openssl" unless defined?(OpenSSL)
+        OpenSSL::HMAC.hexdigest(OpenSSL::Digest.const_get(@digest).new, @secret, data)
+      end
+  end
+end

data/.gem/ruby/2.3.0/gems/activesupport-5.1.4/lib/active_support/multibyte.rb

@@ -0,0 +1,21 @@
+module ActiveSupport #:nodoc:
+  module Multibyte
+    autoload :Chars, "active_support/multibyte/chars"
+    autoload :Unicode, "active_support/multibyte/unicode"
+
+    # 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.
+    #
+    #   ActiveSupport::Multibyte.proxy_class = CharsForUTF32
+    def self.proxy_class=(klass)
+      @proxy_class = klass
+    end
+
+    # Returns the current proxy class.
+    def self.proxy_class
+      @proxy_class ||= ActiveSupport::Multibyte::Chars
+    end
+  end
+end

data/.gem/ruby/2.3.0/gems/activesupport-5.1.4/lib/active_support/multibyte/chars.rb

@@ -0,0 +1,233 @@
+require "active_support/json"
+require "active_support/core_ext/string/access"
+require "active_support/core_ext/string/behavior"
+require "active_support/core_ext/module/delegation"
+require "active_support/core_ext/regexp"
+
+module ActiveSupport #:nodoc:
+  module Multibyte #:nodoc:
+    # Chars enables you to work transparently with UTF-8 encoding in the Ruby
+    # String class without having extensive knowledge about the encoding. A
+    # Chars object accepts a string upon initialization and proxies String
+    # methods in an encoding safe manner. All the normal String methods are also
+    # implemented on the proxy.
+    #
+    # String methods are proxied through the Chars object, and can be accessed
+    # through the +mb_chars+ method. Methods which would normally return a
+    # String object now return a Chars object so methods can be chained.
+    #
+    #   'The Perfect String  '.mb_chars.downcase.strip.normalize
+    #   # => #<ActiveSupport::Multibyte::Chars:0x007fdc434ccc10 @wrapped_string="the perfect string">
+    #
+    # Chars objects are perfectly interchangeable with String objects as long as
+    # no explicit class checks are made. If certain methods do explicitly check
+    # the class, call +to_s+ before you pass chars objects to them.
+    #
+    #   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
+    # ActiveSupport::Multibyte.proxy_class.
+    #
+    #   class CharsForUTF32
+    #     def size
+    #       @wrapped_string.size / 4
+    #     end
+    #
+    #     def self.accepts?(string)
+    #       string.length % 4 == 0
+    #     end
+    #   end
+    #
+    #   ActiveSupport::Multibyte.proxy_class = CharsForUTF32
+    class Chars
+      include Comparable
+      attr_reader :wrapped_string
+      alias to_s wrapped_string
+      alias to_str wrapped_string
+
+      delegate :<=>, :=~, :acts_like_string?, to: :wrapped_string
+
+      # 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
+
+      # Forward all undefined methods to the wrapped string.
+      def method_missing(method, *args, &block)
+        result = @wrapped_string.__send__(method, *args, &block)
+        if /!$/.match?(method)
+          self if result
+        else
+          result.kind_of?(String) ? chars(result) : result
+        end
+      end
+
+      # Returns +true+ if _obj_ responds to the given method. Private methods
+      # are included in the search only if the optional second parameter
+      # evaluates to +true+.
+      def respond_to_missing?(method, include_private)
+        @wrapped_string.respond_to?(method, include_private)
+      end
+
+      # Returns +true+ when the proxy class can handle the string. Returns
+      # +false+ otherwise.
+      def self.consumes?(string)
+        string.encoding == Encoding::UTF_8
+      end
+
+      # Works just like <tt>String#split</tt>, with the exception that the items
+      # in the resulting list are Chars instances instead of String. This makes
+      # chaining methods easier.
+      #
+      #   'Café périferôl'.mb_chars.split(/é/).map { |part| part.upcase.to_s } # => ["CAF", " P", "RIFERÔL"]
+      def split(*args)
+        @wrapped_string.split(*args).map { |i| self.class.new(i) }
+      end
+
+      # Works like <tt>String#slice!</tt>, but returns an instance of
+      # Chars, or +nil+ if the string was not modified. The string will not be
+      # modified if the range given is out of bounds
+      #
+      #   string = 'Welcome'
+      #   string.mb_chars.slice!(3)    # => #<ActiveSupport::Multibyte::Chars:0x000000038109b8 @wrapped_string="c">
+      #   string # => 'Welome'
+      #   string.mb_chars.slice!(0..3) # => #<ActiveSupport::Multibyte::Chars:0x00000002eb80a0 @wrapped_string="Welo">
+      #   string # => 'me'
+      def slice!(*args)
+        string_sliced = @wrapped_string.slice!(*args)
+        if string_sliced
+          chars(string_sliced)
+        end
+      end
+
+      # Reverses all characters in the string.
+      #
+      #   'Café'.mb_chars.reverse.to_s # => 'éfaC'
+      def reverse
+        chars(Unicode.unpack_graphemes(@wrapped_string).reverse.flatten.pack("U*"))
+      end
+
+      # Limits 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.
+      #
+      #   'こんにちは'.mb_chars.limit(7).to_s # => "こん"
+      def limit(limit)
+        slice(0...translate_offset(limit))
+      end
+
+      # Converts characters in the string to uppercase.
+      #
+      #   'Laurent, où sont les tests ?'.mb_chars.upcase.to_s # => "LAURENT, OÙ SONT LES TESTS ?"
+      def upcase
+        chars Unicode.upcase(@wrapped_string)
+      end
+
+      # Converts characters in the string to lowercase.
+      #
+      #   'VĚDA A VÝZKUM'.mb_chars.downcase.to_s # => "věda a výzkum"
+      def downcase
+        chars Unicode.downcase(@wrapped_string)
+      end
+
+      # Converts characters in the string to the opposite case.
+      #
+      #    'El Cañón'.mb_chars.swapcase.to_s # => "eL cAÑÓN"
+      def swapcase
+        chars Unicode.swapcase(@wrapped_string)
+      end
+
+      # Converts the first character to uppercase and the remainder to lowercase.
+      #
+      #  'über'.mb_chars.capitalize.to_s # => "Über"
+      def capitalize
+        (slice(0) || chars("")).upcase + (slice(1..-1) || chars("")).downcase
+      end
+
+      # Capitalizes the first letter of every word, when possible.
+      #
+      #   "ÉL QUE SE ENTERÓ".mb_chars.titleize.to_s    # => "Él Que Se Enteró"
+      #   "日本語".mb_chars.titleize.to_s               # => "日本語"
+      def titleize
+        chars(downcase.to_s.gsub(/\b('?\S)/u) { Unicode.upcase($1) })
+      end
+      alias_method :titlecase, :titleize
+
+      # Returns the KC normalization of the string by default. NFKC is
+      # considered the best normalization form for passing strings to databases
+      # and validations.
+      #
+      # * <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))
+      end
+
+      # Performs canonical decomposition on all the characters.
+      #
+      #   'é'.length                         # => 2
+      #   'é'.mb_chars.decompose.to_s.length # => 3
+      def decompose
+        chars(Unicode.decompose(:canonical, @wrapped_string.codepoints.to_a).pack("U*"))
+      end
+
+      # Performs composition on all the characters.
+      #
+      #   'é'.length                       # => 3
+      #   'é'.mb_chars.compose.to_s.length # => 2
+      def compose
+        chars(Unicode.compose(@wrapped_string.codepoints.to_a).pack("U*"))
+      end
+
+      # Returns the number of grapheme clusters in the string.
+      #
+      #   'क्षि'.mb_chars.length   # => 4
+      #   'क्षि'.mb_chars.grapheme_length # => 3
+      def grapheme_length
+        Unicode.unpack_graphemes(@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))
+      end
+
+      def as_json(options = nil) #:nodoc:
+        to_s.as_json(options)
+      end
+
+      %w(capitalize downcase reverse tidy_bytes upcase).each do |method|
+        define_method("#{method}!") do |*args|
+          @wrapped_string = send(method, *args).to_s
+          self
+        end
+      end
+
+      private
+
+        def translate_offset(byte_offset)
+          return nil if byte_offset.nil?
+          return 0   if @wrapped_string == ""
+
+          begin
+            @wrapped_string.byteslice(0...byte_offset).unpack("U*").length
+          rescue ArgumentError
+            byte_offset -= 1
+            retry
+          end
+        end
+
+        def chars(string)
+          self.class.new(string)
+        end
+    end
+  end
+end

data/.gem/ruby/2.3.0/gems/activesupport-5.1.4/lib/active_support/multibyte/unicode.rb

@@ -0,0 +1,392 @@
+module ActiveSupport
+  module Multibyte
+    module Unicode
+      extend self
+
+      # 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 = "9.0.0"
+
+      # The default normalization used for operations that require
+      # normalization. It can be set to any of the normalizations
+      # in NORMALIZATION_FORMS.
+      #
+      #   ActiveSupport::Multibyte::Unicode.default_normalization_form = :c
+      attr_accessor :default_normalization_form
+      @default_normalization_form = :kc
+
+      # 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
+
+      # 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| database.boundary[c] === codepoint } ? true : false
+      end
+
+      # Unpack the string at grapheme boundaries. Returns a list of character
+      # lists.
+      #
+      #   Unicode.unpack_graphemes('क्षि') # => [[2325, 2381], [2359], [2367]]
+      #   Unicode.unpack_graphemes('Café') # => [[67], [97], [102], [233]]
+      def unpack_graphemes(string)
+        codepoints = string.codepoints.to_a
+        unpacked = []
+        pos = 0
+        marker = 0
+        eoc = codepoints.length
+        while (pos < eoc)
+          pos += 1
+          previous = codepoints[pos - 1]
+          current = codepoints[pos]
+
+          # See http://unicode.org/reports/tr29/#Grapheme_Cluster_Boundary_Rules
+          should_break =
+            if pos == eoc
+              true
+            # GB3. CR X LF
+            elsif previous == database.boundary[:cr] && current == database.boundary[:lf]
+              false
+            # GB4. (Control|CR|LF) ÷
+            elsif previous && in_char_class?(previous, [:control, :cr, :lf])
+              true
+            # GB5. ÷ (Control|CR|LF)
+            elsif in_char_class?(current, [:control, :cr, :lf])
+              true
+            # GB6. L X (L|V|LV|LVT)
+            elsif database.boundary[:l] === previous && in_char_class?(current, [:l, :v, :lv, :lvt])
+              false
+            # GB7. (LV|V) X (V|T)
+            elsif in_char_class?(previous, [:lv, :v]) && in_char_class?(current, [:v, :t])
+              false
+            # GB8. (LVT|T) X (T)
+            elsif in_char_class?(previous, [:lvt, :t]) && database.boundary[:t] === current
+              false
+            # GB9. X (Extend | ZWJ)
+            elsif in_char_class?(current, [:extend, :zwj])
+              false
+            # GB9a. X SpacingMark
+            elsif database.boundary[:spacingmark] === current
+              false
+            # GB9b. Prepend X
+            elsif database.boundary[:prepend] === previous
+              false
+            # GB10. (E_Base | EBG) Extend* X E_Modifier
+            elsif (marker...pos).any? { |i| in_char_class?(codepoints[i], [:e_base, :e_base_gaz]) && codepoints[i + 1...pos].all? { |c| database.boundary[:extend] === c } } && database.boundary[:e_modifier] === current
+              false
+            # GB11. ZWJ X (Glue_After_Zwj | EBG)
+            elsif database.boundary[:zwj] === previous && in_char_class?(current, [:glue_after_zwj, :e_base_gaz])
+              false
+            # GB12. ^ (RI RI)* RI X RI
+            # GB13. [^RI] (RI RI)* RI X RI
+            elsif codepoints[marker..pos].all? { |c| database.boundary[:regional_indicator] === c } && codepoints[marker..pos].count { |c| database.boundary[:regional_indicator] === c }.even?
+              false
+            # GB999. Any ÷ Any
+            else
+              true
+            end
+
+          if should_break
+            unpacked << codepoints[marker..pos - 1]
+            marker = pos
+          end
+        end
+        unpacked
+      end
+
+      # Reverse operation of unpack_graphemes.
+      #
+      #   Unicode.pack_graphemes(Unicode.unpack_graphemes('क्षि')) # => 'क्षि'
+      def pack_graphemes(unpacked)
+        unpacked.flatten.pack("U*")
+      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 = database.codepoints[codepoints[pos]], database.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(type, codepoints)
+        codepoints.inject([]) do |decomposed, cp|
+          # if it's a hangul syllable starter character
+          if HANGUL_SBASE <= cp && 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 = database.codepoints[cp].decomp_mapping) && (!database.codepoints[cp].decomp_type || type == :compatibility)
+            decomposed.concat decompose(type, ncp.dup)
+          else
+            decomposed << cp
+          end
+        end
+      end
+
+      # Compose decomposed characters to the composed form.
+      def compose(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 && lindex < HANGUL_LCOUNT
+            vindex = codepoints[starter_pos + 1] - HANGUL_VBASE rescue vindex = -1
+            if 0 <= vindex && vindex < HANGUL_VCOUNT
+              tindex = codepoints[starter_pos + 2] - HANGUL_TBASE rescue tindex = -1
+              if 0 <= tindex && 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 = database.codepoints[current_char]
+            if current.combining_class > previous_combining_class
+              if ref = database.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
+
+      # Rubinius' String#scrub, however, doesn't support ASCII-incompatible chars.
+      if !defined?(Rubinius)
+        # 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(string, force = false)
+          return string if string.empty?
+          return recode_windows1252_chars(string) if force
+          string.scrub { |bad| recode_windows1252_chars(bad) }
+        end
+      else
+        def tidy_bytes(string, force = false)
+          return string if string.empty?
+          return recode_windows1252_chars(string) if force
+
+          # We can't transcode to the same format, so we choose a nearly-identical encoding.
+          # We're going to 'transcode' bytes from UTF-8 when possible, then fall back to
+          # CP1252 when we get errors. The final string will be 'converted' back to UTF-8
+          # before returning.
+          reader = Encoding::Converter.new(Encoding::UTF_8, Encoding::UTF_16LE)
+
+          source = string.dup
+          out = "".force_encoding(Encoding::UTF_16LE)
+
+          loop do
+            reader.primitive_convert(source, out)
+            _, _, _, error_bytes, _ = reader.primitive_errinfo
+            break if error_bytes.nil?
+            out << error_bytes.encode(Encoding::UTF_16LE, Encoding::Windows_1252, invalid: :replace, undef: :replace)
+          end
+
+          reader.finish
+
+          out.encode!(Encoding::UTF_8)
+        end
+      end
+
+      # Returns the KC normalization of the string by default. NFKC is
+      # considered the best normalization form for passing strings to databases
+      # and validations.
+      #
+      # * <tt>string</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(string, form = nil)
+        form ||= @default_normalization_form
+        # See http://www.unicode.org/reports/tr15, Table 1
+        codepoints = string.codepoints.to_a
+        case form
+        when :d
+          reorder_characters(decompose(:canonical, codepoints))
+        when :c
+          compose(reorder_characters(decompose(:canonical, codepoints)))
+        when :kd
+          reorder_characters(decompose(:compatibility, codepoints))
+        when :kc
+          compose(reorder_characters(decompose(:compatibility, codepoints)))
+          else
+          raise ArgumentError, "#{form} is not a valid normalization variant", caller
+        end.pack("U*".freeze)
+      end
+
+      def downcase(string)
+        apply_mapping string, :lowercase_mapping
+      end
+
+      def upcase(string)
+        apply_mapping string, :uppercase_mapping
+      end
+
+      def swapcase(string)
+        apply_mapping string, :swapcase_mapping
+      end
+
+      # Holds data about a codepoint in the Unicode database.
+      class Codepoint
+        attr_accessor :code, :combining_class, :decomp_type, :decomp_mapping, :uppercase_mapping, :lowercase_mapping
+
+        # Initializing Codepoint object with default values
+        def initialize
+          @combining_class = 0
+          @uppercase_mapping = 0
+          @lowercase_mapping = 0
+        end
+
+        def swapcase_mapping
+          uppercase_mapping > 0 ? uppercase_mapping : lowercase_mapping
+        end
+      end
+
+      # Holds static data from the Unicode database.
+      class UnicodeDatabase
+        ATTRIBUTES = :codepoints, :composition_exclusion, :composition_map, :boundary, :cp1252
+
+        attr_writer(*ATTRIBUTES)
+
+        def initialize
+          @codepoints = Hash.new(Codepoint.new)
+          @composition_exclusion = []
+          @composition_map = {}
+          @boundary = {}
+          @cp1252 = {}
+        end
+
+        # Lazy load the Unicode database so it's only loaded when it's actually used
+        ATTRIBUTES.each do |attr_name|
+          class_eval(<<-EOS, __FILE__, __LINE__ + 1)
+            def #{attr_name}     # def codepoints
+              load               #   load
+              @#{attr_name}      #   @codepoints
+            end                  # end
+          EOS
+        end
+
+        # Loads the Unicode database and returns all the internal objects of
+        # UnicodeDatabase.
+        def load
+          begin
+            @codepoints, @composition_exclusion, @composition_map, @boundary, @cp1252 = File.open(self.class.filename, "rb") { |f| Marshal.load f.read }
+          rescue => e
+            raise IOError.new("Couldn't load the Unicode tables for UTF8Handler (#{e.message}), ActiveSupport::Multibyte is unusable")
+          end
+
+          # Redefine the === method so we can write shorter rules for grapheme cluster breaks
+          @boundary.each_key do |k|
+            @boundary[k].instance_eval do
+              def ===(other)
+                detect { |i| i === other } ? true : false
+              end
+            end if @boundary[k].kind_of?(Array)
+          end
+
+          # define attr_reader methods for the instance variables
+          class << self
+            attr_reader(*ATTRIBUTES)
+          end
+        end
+
+        # Returns the directory in which the data files are stored.
+        def self.dirname
+          File.dirname(__FILE__) + "/../values/"
+        end
+
+        # Returns the filename for the data file for this version.
+        def self.filename
+          File.expand_path File.join(dirname, "unicode_tables.dat")
+        end
+      end
+
+      private
+
+        def apply_mapping(string, mapping)
+          database.codepoints
+          string.each_codepoint.map do |codepoint|
+            cp = database.codepoints[codepoint]
+            if cp && (ncp = cp.send(mapping)) && ncp > 0
+              ncp
+            else
+              codepoint
+            end
+          end.pack("U*")
+        end
+
+        def recode_windows1252_chars(string)
+          string.encode(Encoding::UTF_8, Encoding::Windows_1252, invalid: :replace, undef: :replace)
+        end
+
+        def database
+          @database ||= UnicodeDatabase.new
+        end
+    end
+  end
+end