zhongwen_tools 0.9.0 → 0.11.0

data/lib/zhongwen_tools/romanization/detect.rb

@@ -1,8 +1,9 @@
 # encoding: utf-8
-require File.expand_path("../../regex", __FILE__)
+require 'zhongwen_tools/regex'
 
 module ZhongwenTools
   module Romanization
+    extend self
     # Deprecated: a Regex for accurate pinyin. Use ZhongwenTools::Regex.py instead
     PY_REGEX = ZhongwenTools::Regex.py
 
@@ -10,6 +11,7 @@ module ZhongwenTools
     PINYIN_REGEX = ZhongwenTools::Regex.pyn
 
     # Public: checks if a string is pinyin.
+    #         http://en.wikipedia.org/wiki/Pinyin
     #
     # Examples
     #   py?('nǐ hǎo')
@@ -19,7 +21,8 @@ module ZhongwenTools
     def py?(str = nil)
       str ||= self
 
-      str.gsub(ZhongwenTools::Regex.py, '').strip == ''
+      # NOTE: py regex does not include capitals with tones.
+      String.downcase(str).gsub(Regex.punc,'').gsub(Regex.py, '').gsub(/[\s\-]/,'').strip == ''
     end
 
     # Public: checks if a string is pinyin.
@@ -32,51 +35,33 @@ module ZhongwenTools
     def pyn?(str = nil)
       str ||= self
 
-      str.gsub(ZhongwenTools::Regex.pyn, '').strip == ''
-    end
-
-    # Public: Checks if a string is wade-giles.
-    #
-    # Examples
-    #   wg?('pin1-yin1')
-    #   # => false
-    #
-    # Returns a Boolean.
-    def wg?(str = nil, type = :pyn)
-      # NOTE: There are some situations where wg == pyn, but there's no way to differentiate the two.
-      # FIXME: it shouldn't be pyn, but it should be able to conver to pyn
-      #        Actually, wade-giles does sometimes overlap with pyn. So this
-      #        method creates false negatives. A future :romanization method
-      #        would default to pyn, but this method shouldn't.
-      #        Add tests where str.pyn? and str.wg?
-
-      str ||= self
-      wg = ZhongwenTools::Romanization.to_wade_giles(str, type)
-      # TODO: need to convert string to pyn.
-      pyn = str
-      wg != pyn && wg.gsub(/[1-5]/,'')
+      str.gsub(Regex.punc,'').gsub(Regex.pyn, '').gsub(/[\s\-]/,'').strip == ''
     end
 
     # Public: Checks if a String is Zhuyin Fuhao (a.k.a. bopomofo).
+    #         http://en.wikipedia.org/wiki/Bopomofo
+    #         http://pinyin.info/romanization/bopomofo/index.html
     #
     # str - a String. Optional if the object calling the method is a String.
     #
     # Examples
     #
-    #   zyfh?('ㄊㄥ')
+    #   bpmf?('ㄊㄥ')
     #   # => true
     #
     # Returns a boolean.
-    def zyfh?(str = nil)
+    def bpmf?(str = nil)
       str ||= self
 
-      bopomofo = str.gsub(/[1-5\s]/,'')
-      bopomofo.scan(ZhongwenTools::Regex.bopomofo).join == bopomofo
+      bopomofo = str.gsub(/[1-5\s]/,'').gsub(Regex.punc,'')
+      bopomofo.scan(Regex.bopomofo).join == bopomofo
     end
 
-    # Public: Checks if a String is Tongyong Pinyin.
+    # Public: Checks if a String is a romanization:
+    #         Tongyong Pinyin, Wade Giles, MSP2 or Yale.
     #         http://en.wikipedia.org/wiki/Tongyong_Pinyin
     #         http://pinyin.info/romanization/tongyong/
+    #         http://en.wikipedia.org/wiki/Wade%E2%80%93Giles
     #
     # str - a String. Optional if the object calling the method is a String.
     #
@@ -84,20 +69,64 @@ module ZhongwenTools
     #
     #   typy?('chuei niou')
     #   # => true
+    #   wg?('Mao2 Tse2 Tung1')
     #
     # Returns a boolean.
-    def typy?(str = nil)
+    %w(typy wg yale mps2).each do |type|
+      define_method("#{type}?") do |str = nil|
+        str ||= self
+        # TODO: ignore tonal marks from other systems wade giles, tongyong etc.
+        s = str.downcase.gsub(Regex.punc,'').gsub(/[1-5\s\-']/,'')
+
+        s.scan(detect_regex(type.to_sym)).join == s
+      end
+    end
+
+    # Public: Checks the srings romanizaiton. It always assumes the first correct result is the correct result.
+    #         This can sometimes provide sub-optimal results
+    #         e.g.
+    #           'chuei niou'.romanization? #=> :pyn
+    #           'chuei niou'.pyn? == true # this is correct because ['chu', 'ei', 'ni', 'ou'] are all valid pinyin
+    #                                     # but the best fit for 'chuei niou' should be :typy.
+    #         But this is not considered a major issue because most of the time pyn / py will be used. It could be
+    #         extended to try and figure out the best option, maybe by comparing the syllable length of each
+    #         valid romanization.
+    #
+    # str - a String. Optional if the object calling the method is a String.
+    #
+    # Examples
+    #
+    #
+    #   'hao3'.romanization? #=> :pyn
+    #
+    # Returns a Symbol for the romanization type.
+    def romanization?(str = nil)
       str ||= self
 
-      typy = str.gsub(/[1-5\s\-']/,'')
-      # Sorting by String length means it will match the longest possible part.
-      # FIXME: it is probably possible for this to have false negatives. 
-      #        A more comprehensive regex like Regex.pyn would be needed
-      #        to accurately detect typy.
-      regex_str = ROMANIZATIONS_TABLE.map{ |r| r[:typy] || r[:py] }.sort{|x,y| x.size <=> y.size}.reverse.join('|')
-      typy.scan(/#{regex_str}/).join == typy
+      [:pyn, :py, :zyfh, :wg, :typy, :yale, :msp2].find do |type|
+        self.send("#{type}?", str)
+      end
     end
 
-    # TODO: msp2? yale? wgyrm? romanization?
+    # TODO: romanizations? method that returns all possible romanizations.
+
+    # Deprecated: ZhongwenTools::Romanizaiton.zyfh? is deprecated. Use ZhongwenTools::Romanizaiton.bpmf? instead
+    alias_method :zyfh?, :bpmf?
+
+    private
+
+    # Internal: Produces a Regexp for a romanization type.
+    #
+    # type - a Symbol for the romanization type.
+    #
+    # Examples:
+    #
+    #
+    #   detect_regex(:typy) #=> <Regexp>
+    #
+    # Returns a Regexp.
+    def detect_regex(type)
+      /#{ROMANIZATIONS_TABLE.map{ |r| r[type] || r[:pyn] }.sort{|x,y| x.size <=> y.size}.reverse.join('|')}/
+    end
   end
 end

data/lib/zhongwen_tools/romanization/pyn_to_py.rb

@@ -160,7 +160,8 @@ module ZhongwenTools
       'ng2' => "éng",
       'ng3' => "ěng",
       'ng4' => "èng",
-      'ng5' => 'eng'
+      'ng5' => 'eng',
+      'r5' => 'r'
     }
   end
 end

data/lib/zhongwen_tools/string.rb

@@ -1,9 +1,9 @@
 # encoding: utf-8
 #$:.unshift File.join(File.dirname(__FILE__),'..','lib','zhongwen_tools', 'string')
 require 'uri'
-require File.expand_path("../regex", __FILE__)
-require File.expand_path("../string/fullwidth", __FILE__)
-require File.expand_path("../string/caps", __FILE__)
+require 'zhongwen_tools/regex'
+require 'zhongwen_tools/string/fullwidth'
+require 'zhongwen_tools/string/caps'
 
 class String
   alias_method :_downcase, :downcase

data/lib/zhongwen_tools/version.rb

@@ -1,3 +1,3 @@
 module ZhongwenTools
-  VERSION = "0.9.0"
+  VERSION = '0.11.0'
 end

data/test/test_romanization.rb

@@ -15,17 +15,32 @@ class TestRomanization < Minitest::Test
     assert_equal 'Zhōngwén', 'Zhong1-wen2'.to_pinyin
     assert_equal "Tiān'ānmén",'Tian1an1men2'.to_pinyin
     assert_equal @alabo[:py], @alabo[:pyn].to_pinyin
-
+    assert_equal 'r', 'r5'.to_pinyin
     #wg -> py not yet implemented
     #mzd = "Mao Tse-tung"
     #assert_equal "Mao Zedong", mzd.to_pinyin(:wg)
+    assert @alabo[:py].py?
+    assert 'Ā-lā-bó'.py?
+    assert 'Zhong1 wen2'.to_pinyin.py?
   end
 
   def test_pyn
     assert_equal 'ni3 hao3', @py.to_pyn(:py)
     assert_equal 'tian1an1men2', 'tian1an1men2'.to_py.to_pyn(:py)
 
-    #assert_equal 'Wūlúhānuòfū'.to_pyn, 'Wu1-lu2-ha1-nuo4-fu1'
+    assert_equal 'yi4', 'yì'.to_pyn(:py)
+
+    assert_equal 'ni3 hao3', 'ㄋㄧ3 ㄏㄠ3'.to_pyn(:bpmf)
+    assert_equal 'ni3 hao3', 'ㄋㄧ3 ㄏㄠ3'.to_pyn
+    assert_equal 'zhong1 guo2', 'chung1 kuo2'.to_pyn(:wg)
+    assert_equal 'zhong1 guo2', 'chung1 kuo2'.to_pyn
+    assert_equal 'chui1 niu3', 'chuei1 niou3'.to_pyn(:typy)
+    assert_equal 'cao3 di4', 'tsau3 di4'.to_pyn(:mspy2)
+    assert_equal 'cao3 di4', 'tsau3 di4'.to_pyn
+    assert_equal 'cao3 di4', 'tsau3 di4'.to_pyn(:yale)
+    assert_equal 'cao3 di4', 'tsau3 di4'.to_pyn
+
+    #assert_equal 'Wu1-lu2-ha1-nuo4-fu1', 'Wūlúhānuòfū'.to_pyn(:py)
     #"007：Dàpò Liàngzǐ Wēijī", "007: Da4po4 Liang4zi3 Wei1ji1"
   end
 
@@ -37,6 +52,8 @@ class TestRomanization < Minitest::Test
      assert_equal 'ㄑㄧㄥ3 ㄏㄨㄟ2ㄉㄚ2 ㄨㄛ3 ㄉㄜ5 ㄨㄣ4ㄊㄧ2 .', @sent.to_zhuyin
      assert_equal 'ㄇㄠ2 ㄗㄜ2ㄉㄨㄥ1', @mzd2.to_zhuyin_fuhao
      assert 'ㄋㄧ3 ㄏㄠ3'.zyfh?
+
+     assert_equal 'ㄋㄧ3 ㄏㄠ3', 'ni3 hau3'.to_bpmf(:yale)
   end
 
   def test_wade_giles
@@ -45,6 +62,8 @@ class TestRomanization < Minitest::Test
     assert_equal 'Mao2 Tse2 tung1', @mzd.to_wg
     assert_equal 'Mao2 Tse2-tung1', @mzd2.to_wade_giles
     assert_equal 'Mao2 Tse2-tung1 te5 mao2', 'Mao2 Ze2-dong1 de5 mao2'.to_wade_giles
+
+    assert_equal 'ni3 hao3', 'ni3 hau3'.to_wg(:yale)
   end
 
   #def test_mspy2
@@ -65,11 +84,15 @@ class TestRomanization < Minitest::Test
 
   def test_yale
     assert_equal 'ni3 hau3', @str.to_yale
+
+    assert_equal 'chwei1 nyou3', 'chuei1 niou3'.to_yale(:typy)
   end
 
-  #def test_romanization?
-    #skip
-  #end
+  def test_romanization?
+    assert_equal :pyn, @alabo[:pyn].romanization?
+    assert_equal :py, @alabo[:py].romanization?
+    assert_equal :wg, @mzd.to_wg(:pyn).romanization?
+  end
 
   def test_detect
     assert @str.pyn?
@@ -78,9 +101,11 @@ class TestRomanization < Minitest::Test
 
     assert 'chung1 kuo2'.wg?
 
-    # Travis CI is having trouble with this using Ruby 1.8.7, but it works locally.
-    # I'll probably end up dropping full 1.8.7 support.
     assert @py.py?, "#{@py} should be pinyin. (#{@py.py?})" unless RUBY_VERSION < '1.9'
+    assert 'chuei1 niou3'.typy?
+    assert 'ㄋㄧ3 ㄏㄠ3'.bpmf?
+    assert 'ni3 hau3'.yale?
+    assert 'tsuen'.mps2?
   end
 
   def test_split_pyn
@@ -92,6 +117,13 @@ class TestRomanization < Minitest::Test
   end
 
   def setup
+    @romanizations = [
+      # FIXME: bopomofo, tongyong pinyin, wade-giles tones are all wrong.
+      # TODO: test IPA
+      { :pyn => 'ni3 hao', :py => 'nǐ hǎo', :bopomofo => 'ㄋㄧ3 ㄏㄠ3', :yale => 'ni3 hau3', :typy => 'ni3 hao3', :wg => 'ni3 hao3'},#, :ipa => ''}
+      { :pyn => 'zhong1 guo2', :py => 'nǐ hǎo', :bopomofo => 'ㄋㄧ3 ㄏㄠ3', :yale => 'ni3 hau3', :typy => 'ni3 hao3', :wg => 'chung1 kuo2'},#, :ipa => ''}
+      { :pyn => 'chui1 niu3', :py => '', :bopomofo => '', :yale => 'chwei1 nyou3', :typy => 'chuei1 niou3', :wg => 'chung1 kuo2'},#, :ipa => ''}
+    ]
     @str = 'ni3 hao3'
     @mzd = 'Mao2 Ze2 dong1'
     @mzd2 = 'Mao2 Ze2-dong1'

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: zhongwen_tools
 version: !ruby/object:Gem::Version
-  version: 0.9.0
+  version: 0.11.0
 platform: ruby
 authors:
 - Steven Daniels
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-05-22 00:00:00.000000000 Z
+date: 2014-06-02 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rake