@@ -1,5 +1,9 @@ # encoding: utf-8 +# Unichars is a proxy class for String. It's used by Unichars as a trimmed down version of ActiveSupport::Multibyte::Chars +# when ActiveSupport isn't loaded. +# +# Chars.new('João') #=> #<Chars:0x34c240 @wrapped_string="João"> class Chars attr_reader :wrapped_string alias to_s wrapped_string lib/chars.rb @@ -9,29 +9,65 @@ rescue NameError class Unichars < Chars; end end +# Unichars is a proxy class for String. It wraps a String and implements UTF-8 safe versions of various String operations. +# Unimplemented methods are forwarded to the wrapped string. +# +# Unichars uses ActiveSupport::Multibyte::Chars as a superclass if it's loaded. Otherwise it will use it's own Chars class +# which is basically a trimmed down version of ActiveSupport's Chars class. +# +# require 'unichars' +# Unichars.superclass #=> Chars +# +# require 'active_support' +# require 'unichars' +# Unichars.superclass #=> ActiveSupport::Multibyte::Chars +# +# Note that all the operations on strings are implemented using Glib2, so the outcome of the methods will be influenced by the +# Glib2 version installed on your system. class Unichars + # Valid normalization forms NORMALIZATION_FORMS = [:c, :kc, :d, :kd] class << self attr_accessor :default_normalization_form end + # Returns the length of the string expressed in codepoints. + # + # Unichars.new('A ehm…, word.').size #=> 13 def size Glib.utf8_size(@wrapped_string) end + # Returns a Unichars instance with the string in capitals if they are are available for the supplied string. + # + # Unichars.new('Sluß').upcase.to_s #=> SLUSS def upcase self.class.new(Glib.utf8_upcase(@wrapped_string)) end + # Returns a Unichars instance with the string in lowercase characters if they are are available for the supplied string. + # + # Unichars.new('ORGANISÉE').downcase.to_s #=> organisée def downcase self.class.new(Glib.utf8_downcase(@wrapped_string)) end + # Returns a Unichars instance with the string in reverse order. + # + # Unichars.new('Comment ça va?').reverse.to_s #=> av aç tnemmoC def reverse self.class.new(Glib.utf8_reverse(@wrapped_string)) end + # Returns a Unichars instance with the string in normalize form. See http://www.unicode.org/reports/tr15/tr15-29.html + # for more information about normalization. + # + # <i>form</i> can be one of the following: <tt>:c</tt>, <tt>:kc</tt>, <tt>:d</tt>, or <tt>:kd</tt>. + # + # decomposed = [101, 769].pack('U*') + # composed = Unichars.new(decomposed).normalize(:kc) + # composed.to_s.unpack('U*') #=> [233] def normalize(form=Unichars.default_normalization_form) self.class.new(Glib.utf8_normalize(@wrapped_string, form)) end lib/unichars.rb @@ -2,7 +2,7 @@ require 'rubygems' Gem::Specification.new do |spec| spec.name = 'unichars' - spec.version = '0.3' + spec.version = '0.4' spec.author = "Manfred Stienstra" spec.email = "manfred@fngtps.com" unichars.gemspec 235823df9ee7d36ce158298d2f72688865595bea Manfred Stienstra manfred@fngtps.com http://github.com/Manfred/unichars/commit/ca2ac762e6664cc426bf2a07686b0db6b3de90c8 ca2ac762e6664cc426bf2a07686b0db6b3de90c8 2008-12-31T06:38:39-08:00 2008-12-31T06:38:39-08:00 Add documentation to the Unichars class and update the Chars documentation. Bump version to 0.4. b02bdb797de9182d864679b4a7d7aa8adb8f80a5 Manfred Stienstra manfred@fngtps.com