chars.rb in activesupport-7.0.0.alpha1

- old
+ new

@@ -1,15 +1,14 @@
 # frozen_string_literal: true
 
 require "active_support/json"
 require "active_support/core_ext/string/access"
 require "active_support/core_ext/string/behavior"
-require "active_support/core_ext/symbol/starts_ends_with"
 require "active_support/core_ext/module/delegation"
 
-module ActiveSupport #:nodoc:
-  module Multibyte #:nodoc:
+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.
@@ -101,11 +100,11 @@
 
       # Reverses all characters in the string.
       #
       #   'Café'.mb_chars.reverse.to_s # => 'éfaC'
       def reverse
-        chars(@wrapped_string.scan(/\X/).reverse.join)
+        chars(@wrapped_string.grapheme_clusters.reverse.join)
       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.
@@ -124,30 +123,30 @@
       end
       alias_method :titlecase, :titleize
 
       # Performs canonical decomposition on all the characters.
       #
-      #   'é'.length                         # => 2
-      #   'é'.mb_chars.decompose.to_s.length # => 3
+      #   'é'.length                         # => 1
+      #   'é'.mb_chars.decompose.to_s.length # => 2
       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
+      #   'é'.length                       # => 1
+      #   'é'.mb_chars.compose.to_s.length # => 1
       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
+      #   'क्षि'.mb_chars.grapheme_length # => 2
       def grapheme_length
-        @wrapped_string.scan(/\X/).length
+        @wrapped_string.grapheme_clusters.length
       end
 
       # Replaces all ISO-8859-1 or CP1252 characters by their UTF-8 equivalent
       # resulting in a valid UTF-8 string.
       #
@@ -155,10 +154,10 @@
       # 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:
+      def as_json(options = nil) # :nodoc:
         to_s.as_json(options)
       end
 
       %w(reverse tidy_bytes).each do |method|
         define_method("#{method}!") do |*args|