text 1.0.2 → 1.0.3

data/lib/text/version.rb

@@ -2,7 +2,7 @@ module Text
   module VERSION #:nodoc:
     MAJOR = 1
     MINOR = 0
-    TINY  = 2
+    TINY  = 3
 
     STRING = [MAJOR, MINOR, TINY].join('.')
   end

data/lib/text/white_similarity.rb

@@ -34,25 +34,28 @@ module Text
     end
 
     def similarity(str1, str2)
-      pairs1, length1 = word_letter_pairs(str1)
-      pairs2, length2 = word_letter_pairs(str2)
+      pairs1 = word_letter_pairs(str1)
+      pairs2 = word_letter_pairs(str2).dup
 
-      intersection = pairs1.inject(0) { |acc, pair|
-        pairs2.include?(pair) ? acc + 1 : acc
-      }
-      union = length1 + length2
+      union = pairs1.count + pairs2.count
+
+      intersection = 0
+      pairs1.each_with_index do |pair1|
+        if index = pairs2.index(pair1)
+          intersection += 1
+          pairs2.delete_at(index)
+        end
+      end
 
       (2.0 * intersection) / union
     end
 
   private
     def word_letter_pairs(str)
-      @word_letter_pairs[str] ||= (
-        pairs = str.upcase.split(/\s+/).map{ |word|
+      @word_letter_pairs[str] ||=
+        str.upcase.split(/\s+/).map{ |word|
           (0 ... (word.length - 1)).map { |i| word[i, 2] }
-        }.flatten
-        [Set.new(pairs), pairs.length]
-      )
+        }.flatten.freeze
     end
   end
 end

data/test/test_white_similarity.rb

@@ -27,6 +27,12 @@ class WhiteSimilarityTest < Test::Unit::TestCase
     assert_in_delta 0.0,  white.similarity(word, "Sold"),    0.01
   end
 
+  def test_should_not_clobber_cached_values
+    white = Text::WhiteSimilarity.new
+    word = "Healed"
+    assert_equal white.similarity(word, word), white.similarity(word, word)
+  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
@@ -35,4 +41,9 @@ class WhiteSimilarityTest < Test::Unit::TestCase
     assert_in_delta 0.61, Text::WhiteSimilarity.similarity("FRENCH REPUBLIC", "REPUBLIC OF CUBA"),   0.01
   end
 
+  def test_similarity_with_equal_strings
+    assert_equal 1.0, Text::WhiteSimilarity.similarity("aaaaa", "aaaaa")
+    assert_equal 1.0, Text::WhiteSimilarity.similarity("REPUBLIC OF CUBA", "REPUBLIC OF CUBA")
+  end
+
 end

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: text
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.0.3
   prerelease: 
 platform: ruby
 authors:
@@ -11,7 +11,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2011-12-13 00:00:00.000000000 Z
+date: 2012-01-09 00:00:00.000000000 Z
 dependencies: []
 description: ! 'A collection of text algorithms: Levenshtein, Soundex, Metaphone,
   Double Metaphone, Figlet, Porter Stemming'