text 1.0.1 → 1.0.2

data/lib/text/version.rb

@@ -2,7 +2,7 @@ module Text
   module VERSION #:nodoc:
     MAJOR = 1
     MINOR = 0
-    TINY  = 1
+    TINY  = 2
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

data/lib/text/white_similarity.rb

@@ -34,23 +34,24 @@ module Text
     end
 
     def similarity(str1, str2)
-      pairs1 = word_letter_pairs(str1)
-      pairs2 = word_letter_pairs(str2)
+      pairs1, length1 = word_letter_pairs(str1)
+      pairs2, length2 = word_letter_pairs(str2)
 
       intersection = pairs1.inject(0) { |acc, pair|
         pairs2.include?(pair) ? acc + 1 : acc
       }
-      union = pairs1.length + pairs2.length
+      union = length1 + length2
 
       (2.0 * intersection) / union
     end
 
   private
     def word_letter_pairs(str)
-      @word_letter_pairs[str] ||= Set.new(
-        str.upcase.split(/\s+/).map{ |word|
-          (0 ... (word.length - 1)).map { |i| str[i, 2] }
+      @word_letter_pairs[str] ||= (
+        pairs = str.upcase.split(/\s+/).map{ |word|
+          (0 ... (word.length - 1)).map { |i| word[i, 2] }
         }.flatten
+        [Set.new(pairs), pairs.length]
       )
     end
   end

data/test/test_white_similarity.rb

@@ -26,4 +26,13 @@ class WhiteSimilarityTest < Test::Unit::TestCase
     assert_in_delta 0.25, white.similarity(word, "Help"),    0.01
     assert_in_delta 0.0,  white.similarity(word, "Sold"),    0.01
   end
+
+  def test_similarity_with_examples_from_article
+    assert_in_delta 0.4,  Text::WhiteSimilarity.similarity("GGGGG", "GG"),                           0.01
+    assert_in_delta 0.56, Text::WhiteSimilarity.similarity("REPUBLIC OF FRANCE", "FRANCE"),          0.01
+    assert_in_delta 0.0,  Text::WhiteSimilarity.similarity("FRANCE", "QUEBEC"),                      0.01
+    assert_in_delta 0.72, Text::WhiteSimilarity.similarity("FRENCH REPUBLIC", "REPUBLIC OF FRANCE"), 0.01
+    assert_in_delta 0.61, Text::WhiteSimilarity.similarity("FRENCH REPUBLIC", "REPUBLIC OF CUBA"),   0.01
+  end
+
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: text
 version: !ruby/object:Gem::Version
-  version: 1.0.1
+  version: 1.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -11,7 +11,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-11-22 00:00:00.000000000Z
+date: 2011-12-13 00:00:00.000000000 Z
 dependencies: []
 description: ! 'A collection of text algorithms: Levenshtein, Soundex, Metaphone,
   Double Metaphone, Figlet, Porter Stemming'