carpodaster-unicase 0.2.0

data/lib/case_data_creator.rb

@@ -0,0 +1,19 @@
+#!/usr/bin/env ruby
+require 'open-uri'
+
+#downloads the latest UnicodeData.txt file, and then spits out
+#codepoint, lowercase, uppercase
+#for codepoiunts that have lowercase or uppercase versions
+out = File.open("case_data.dat", "w")
+open('http://unicode.org/Public/UNIDATA/UnicodeData.txt') do |f|
+  f.each_line do |line|
+    parts = line.chomp!.split(';')
+    cp = parts[0]
+    lc = parts[13]
+    uc = parts[12]
+    if lc || uc
+      out.puts [cp, lc || "", uc || ""].join("\t")
+    end
+  end
+end
+out.close

data/lib/unicase.rb

@@ -0,0 +1,49 @@
+# encoding: utf-8
+module Unicase
+  Uppercase = Hash.new { |h,k| k}
+  Lowercase = Hash.new { |h,k| k}
+  def self.to_u(x)
+    [x.to_i(16)].pack("U").freeze 
+  end
+  
+  File.open(File.join(File.dirname(__FILE__), "case_data.dat")) do |f|
+    f.each do |line|
+      parts = line.chomp!.split("\t")
+      cp = parts[0]
+      lc = parts[1] || ""
+      uc = parts[2] || "" 
+      char = Unicase::to_u(cp)
+      Lowercase[char] = Unicase::to_u(lc) if lc != ""
+      Uppercase[char] = Unicase::to_u(uc) if uc != ""
+    end
+  end
+end
+
+class String
+  def first_unicode_capitalize
+    if self =~ /\A(.)/um
+      Unicase::Uppercase[$1] + $'
+    else
+      self.dup
+    end
+  end
+  
+  def first_unicode_lowercase
+    if self =~ /\A(.)/um
+      Unicase::Lowercase[$1] + $'
+    else
+      self.dup
+    end
+  end
+
+  def unicode_lowercase
+    res = ""
+    each_char { |x| res << Unicase::Lowercase[x] }
+    res
+  end
+  def unicode_uppercase
+    res = ""
+    each_char { |x| res << Unicase::Uppercase[x] }
+    res
+  end
+end

metadata

@@ -0,0 +1,76 @@
+--- !ruby/object:Gem::Specification 
+name: carpodaster-unicase
+version: !ruby/object:Gem::Version 
+  hash: 23
+  prerelease: false
+  segments: 
+  - 0
+  - 2
+  - 0
+  version: 0.2.0
+platform: ruby
+authors: 
+- bkkgrad
+- Carsten Zimmermann
+autorequire: 
+bindir: bin
+cert_chain: []
+
+date: 2010-11-11 00:00:00 +01:00
+default_executable: 
+dependencies: []
+
+description: Upcase and downcase methods for non-ASCII string; originally taken from bkkbrad/unicase
+email: carp@hacksocke.de
+executables: []
+
+extensions: []
+
+extra_rdoc_files: 
+- LICENSE
+- README
+files: 
+- CHANGELOG.rdoc
+- LICENSE
+- README
+- Rakefile
+- VERSION
+- lib/case_data.dat
+- lib/case_data_creator.rb
+- lib/unicase.rb
+has_rdoc: true
+homepage: http://github.com/carpodaster/unicase
+licenses: []
+
+post_install_message: 
+rdoc_options: 
+- --charset=UTF-8
+require_paths: 
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+required_rubygems_version: !ruby/object:Gem::Requirement 
+  none: false
+  requirements: 
+  - - ">="
+    - !ruby/object:Gem::Version 
+      hash: 3
+      segments: 
+      - 0
+      version: "0"
+requirements: []
+
+rubyforge_project: 
+rubygems_version: 1.3.7
+signing_key: 
+specification_version: 3
+summary: Unicase-aware upcase and downcase
+test_files: []
+