unichars 0.3 → 0.4

data/ext/glib/glib.c

@@ -35,6 +35,7 @@ static VALUE utf8_upcase(VALUE self, VALUE string)
   Check_Type(string, T_STRING);
   temp = g_utf8_strup(StringValuePtr(string), RSTRING(string)->len);
   result = rb_str_new2(temp);
+  free(temp);
 
   return result;
 }
@@ -55,6 +56,7 @@ static VALUE utf8_downcase(VALUE self, VALUE string)
   Check_Type(string, T_STRING);
   temp = g_utf8_strdown(StringValuePtr(string), RSTRING(string)->len);
   result = rb_str_new2(temp);
+  free(temp);
 
   return result;
 }
@@ -75,6 +77,7 @@ static VALUE utf8_reverse(VALUE self, VALUE string)
   Check_Type(string, T_STRING);
   temp = g_utf8_strreverse(StringValuePtr(string), RSTRING(string)->len);
   result = rb_str_new2(temp);
+  free(temp);
 
   return result;
 }
@@ -116,6 +119,7 @@ static VALUE utf8_normalize(VALUE self, VALUE string, VALUE form)
 
   temp = g_utf8_normalize(StringValuePtr(string), RSTRING(string)->len, mode);
   result = rb_str_new2(temp);
+  free(temp);
 
   return result;
 }

data/lib/chars.rb

@@ -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

data/lib/unichars.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 its superclass if it's loaded. Otherwise it will use its 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

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification 
 name: unichars
 version: !ruby/object:Gem::Version 
-  version: "0.3"
+  version: "0.4"
 platform: ruby
 authors: 
 - Manfred Stienstra
@@ -9,7 +9,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2008-12-22 00:00:00 +01:00
+date: 2009-01-06 00:00:00 +01:00
 default_executable: 
 dependencies: []
 
@@ -53,7 +53,7 @@ required_rubygems_version: !ruby/object:Gem::Requirement
 requirements: []
 
 rubyforge_project: 
-rubygems_version: 1.2.0
+rubygems_version: 1.3.1
 signing_key: 
 specification_version: 2
 summary: Unichars is a wrapper around Glib2 UTF8 functions. It was written to speed up ActiveSupport::Multibyte, but I'm sure people can find other uses for it.