script_detector 0.1.2

data/lib/script_detector.rb

@@ -0,0 +1,48 @@
+# encoding: UTF-8
+
+require File.dirname(__FILE__) + '/chinese_detector.rb'
+
+class String
+  include ChineseDetector
+
+  # Returns true if the string contains Chinese characters _and_ no Japanese or Korean characters
+  def chinese?
+    look_for_chars_in(/\p{Han}/) and !self.japanese? and !self.korean?
+  end
+
+  # Return true if the string contains traditional Chinese characters (繁體字)
+  def traditional_chinese?
+    look_for_chars_in traditional_chinese_regex
+  end
+
+  # Return true if the string contains simplified Chinese characters (简体字)
+  def simplified_chinese?
+    look_for_chars_in simplified_chinese_regex
+  end
+
+  # Returns true if the string contains specifically Japanese (hiragana or katakana) characters
+  def japanese?
+    look_for_chars_in /(\p{Katakana}|\p{Hiragana})/
+  end
+
+  # Returns true if the string contains specifically Korean (hangul) characters
+  def korean?
+    look_for_chars_in /\p{Hangul}/
+  end
+
+  # Try to detect script and return one of "Japanese", "Korean", "Traditional Chinese", "Simplified Chinese", "Ambiguous Chinese" or "Unknown"
+  def identify_script
+    return "Japanese" if self.japanese?
+    return "Korean" if self.korean?
+    return "Traditional Chinese" if self.traditional_chinese?
+    return "Simplified Chinese" if self.simplified_chinese?
+    return "Ambiguous Chinese" if self.chinese?
+    "Unknown"
+  end
+
+  private
+
+  def look_for_chars_in regex
+    !! (self =~ regex)
+  end
+end

data/spec/helper.rb

@@ -0,0 +1,21 @@
+require 'rubygems'
+require 'bundler'
+begin
+  Bundler.setup(:default, :development)
+rescue Bundler::BundlerError => e
+  $stderr.puts e.message
+  $stderr.puts "Run `bundle install` to install missing gems"
+  exit e.status_code
+end
+
+require 'simplecov'
+SimpleCov.start do
+  add_filter "config"
+  add_filter "spec"
+  add_filter "vendor"
+  coverage_dir "target/reports/coverage"
+end
+
+$LOAD_PATH.unshift(File.join(File.dirname(__FILE__), '..', 'lib'))
+$LOAD_PATH.unshift(File.dirname(__FILE__))
+require 'script_detector'

data/spec/script_detector_spec.rb

@@ -0,0 +1,132 @@
+# encoding: UTF-8
+
+require 'helper'
+
+describe 'Script detection for' do
+  describe 'Simplified Chinese text' do
+    before do
+      @chinese_simpl = ' 我的气垫船充满了鳝鱼.'
+    end
+
+    it 'is Chinese' do
+      @chinese_simpl.should be_chinese
+    end
+
+    it 'is simplified Chinese' do
+      @chinese_simpl.should be_simplified_chinese
+    end
+
+    it 'is not traditional Chinese' do
+      @chinese_simpl.should_not be_traditional_chinese
+    end
+
+    it 'is not Japanese' do
+      @chinese_simpl.should_not be_japanese
+    end
+
+    it 'is identified as Simplified Chinese' do
+      @chinese_simpl.identify_script.should == "Simplified Chinese"
+    end
+  end
+
+  describe 'Traditional Chinese text' do
+    before do
+      @chinese_trad = ' 我的氣墊船充滿了鱔魚.'
+    end
+
+    it 'is Chinese' do
+      @chinese_trad.should be_chinese
+    end
+
+    it 'is traditional Chinese' do
+      @chinese_trad.should be_traditional_chinese
+    end
+
+    it 'is not simplified Chinese' do
+      @chinese_trad.should_not be_simplified_chinese
+    end
+
+    it 'is not Japanese' do
+      @chinese_trad.should_not be_japanese
+    end
+
+    it 'is identified as traditional Chinese' do
+      @chinese_trad.identify_script.should == "Traditional Chinese"
+    end
+  end
+
+  describe 'Ambiguous Chinese text' do
+    before do
+      @chinese_amb = '你好.'
+    end
+
+    it 'is Chinese' do
+      @chinese_amb.should be_chinese
+    end
+
+    it 'is neither traditional nor simplified Chinese' do
+      @chinese_amb.should_not be_simplified_chinese
+      @chinese_amb.should_not be_traditional_chinese
+    end
+
+    it 'is identified as ambiguous Chinese' do
+      @chinese_amb.identify_script.should == "Ambiguous Chinese"
+    end
+  end
+
+  describe 'Japanese text' do
+    before do
+      @japanese = ' 私のホバークラフトは鰻でいっぱいです.'
+    end
+
+    it 'is Japanese' do
+      @japanese.should be_japanese
+    end
+
+    it 'is not Chinese or Korean' do
+      @japanese.should_not be_chinese
+      @japanese.should_not be_korean
+    end
+
+    it 'is identified as Japanese' do
+      @japanese.identify_script.should == "Japanese"
+    end
+  end
+
+  describe 'Korean text' do
+    before do
+      @korean = ' 내 호버크라프트는 장어로 가득 차 있어요.'
+    end
+
+    it 'is Korean' do
+      @korean.should be_korean
+    end
+
+    it 'is not Chinese or Japanese' do
+      @korean.should_not be_chinese
+      @korean.should_not be_japanese
+    end
+
+    it 'is identified as Korean' do
+      @korean.identify_script.should == "Korean"
+    end
+  end
+
+  describe 'Non-CJK text' do
+    before do
+      @unknown = 'Hello world.'
+    end
+
+    it 'is none of the above' do
+      @unknown.should_not be_chinese
+      @unknown.should_not be_simplified_chinese
+      @unknown.should_not be_traditional_chinese
+      @unknown.should_not be_japanese
+      @unknown.should_not be_korean
+    end
+
+    it 'is identified as unknown' do
+      @unknown.identify_script.should == "Unknown"
+    end
+  end
+end

data/spec/spec.opts

@@ -0,0 +1 @@
+--colour --format progress

metadata

@@ -0,0 +1,176 @@
+--- !ruby/object:Gem::Specification
+name: script_detector
+version: !ruby/object:Gem::Version
+  version: 0.1.2
+  prerelease: 
+platform: ruby
+authors:
+- Jani Patokallio
+autorequire: 
+bindir: bin
+cert_chain: []
+date: 2012-05-30 00:00:00.000000000 Z
+dependencies:
+- !ruby/object:Gem::Dependency
+  name: shoulda
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: rdoc
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '3.12'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: '3.12'
+- !ruby/object:Gem::Dependency
+  name: rspec-rails
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: simplecov
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: nokogiri
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ! '>='
+      - !ruby/object:Gem::Version
+        version: '0'
+- !ruby/object:Gem::Dependency
+  name: bundler
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.1.0
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.1.0
+- !ruby/object:Gem::Dependency
+  name: jeweler
+  requirement: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.8.3
+  type: :development
+  prerelease: false
+  version_requirements: !ruby/object:Gem::Requirement
+    none: false
+    requirements:
+    - - ~>
+      - !ruby/object:Gem::Version
+        version: 1.8.3
+description: Utility library for determining if string is traditional Chinese, simplified
+  Chinese, Japanese or Korean
+email: jpatokal@iki.fi
+executables: []
+extensions: []
+extra_rdoc_files:
+- LICENSE.txt
+- README.rdoc
+files:
+- .document
+- Gemfile
+- Gemfile.lock
+- LICENSE.txt
+- README.rdoc
+- Rakefile
+- VERSION
+- lib/chinese_detector.rb
+- lib/script_detector.rb
+- spec/helper.rb
+- spec/script_detector_spec.rb
+- spec/spec.opts
+homepage: http://github.com/jpatokal/script_detector
+licenses:
+- MIT
+post_install_message: 
+rdoc_options: []
+require_paths:
+- lib
+required_ruby_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+      segments:
+      - 0
+      hash: -1438461098142701375
+required_rubygems_version: !ruby/object:Gem::Requirement
+  none: false
+  requirements:
+  - - ! '>='
+    - !ruby/object:Gem::Version
+      version: '0'
+requirements: []
+rubyforge_project: 
+rubygems_version: 1.8.21
+signing_key: 
+specification_version: 3
+summary: Utility library for determining if string is traditional Chinese, simplified
+  Chinese, Japanese or Korean
+test_files: []