ridic 0.5.2 → 0.6.1

data/README.md

@@ -20,18 +20,21 @@ Or install it yourself as:
 
     $ gem install ridic
 
-## Usage
+## Documentation
+
+API Methods :
 
-Search Methods :
   * RiDic.word_match("word")
+    * Returns categories if word is an exact match
   * RiDic.stem_match("word")
+    * Returns categories if word is a match for a provided stem
+  * RiDic.all_categories_in_document("document text")
+    * Returns a list of all categories matching all words in a given body of text
+  * RiDic.category_in_document("document text", category_number)
+    * Returns a list of a specific category hierarchy contained in a given body of text
+  * RiDic.category_distribution("document text", category_number)
+    * Returns the occurrence of each specified category (within a hierarchy) contained in a body of text
 
-word_match will match words that are *not* stemmed  
-stem_match will match words against stems  
-
-The logic is separated for performance reasons -  
-stem_match will iterate over the dictionary and return once a match has been found  
-Both methods return nil if no match is found
 
 ## Contributing
 

data/lib/ridic/version.rb

@@ -1,3 +1,3 @@
 module RiDic
-  VERSION = "0.5.2"
+  VERSION = "0.6.1"
 end

data/lib/ridic.rb

@@ -3,12 +3,42 @@ require "ridic/dictionary"
 
 module RiDic
   def self.word_match(text_word)
-    RiDic::Dictionary.words[text_word.upcase]
+    dictionary_1 = RiDic::Dictionary.words[text_word.upcase]
+    dictionary_1 == nil ? RiDic::Dictionary.word_stems[text_word.upcase] : dictionary_1
   end
 
   def self.stem_match(text_word)
     text_word.upcase!
-    RiDic::Dictionary.word_stems.each {|key, value| (return [key,value]) if text_word.match("^#{key}")}
+    RiDic::Dictionary.word_stems.each {|key, value| (return value) if text_word.match("^#{key}")}
     nil
   end  
+
+  def self.all_categories_in_document(document_text, result = [])
+    sanitize(document_text).split(' ').each do |elem| 
+      word_match(elem) == nil ? result << stem_match(elem) : result << word_match(elem)
+    end
+    result
+  end
+
+  def self.category_in_document(document_text, category_number, result = [])
+    all_categories_in_document(sanitize(document_text)).delete_if {|i| i == nil}.each {|elem| result << [elem[category_number - 1]]}
+    result.delete_if {|i| i == [""]}
+  end
+
+  def self.category_distribution(document_text, category_number, result = Hash.new(0))
+    first_categories = category_in_document(sanitize(document_text), category_number).delete_if {|i| i == nil}
+    first_categories.each {|elem| result[elem.first] += 1}
+    sort_distribution(result)
+  end
+
+  private
+
+  def self.sort_distribution(distribution_set, result = Hash.new)
+    distribution_set.sort_by {|key, value| value}.reverse.each {|i| result[i[0]] = i[1]}
+    result
+  end
+
+  def self.sanitize(document_text)
+    document_text.split(' ').each {|word| word.gsub!(/\W/, '')}.join(' ')
+  end
 end

data/spec/ridic_spec.rb

@@ -13,12 +13,129 @@ describe 'RiDic' do
 
   describe '.stem_match' do 
     it 'retrives word if stem found in dictionary' do
-      RiDic.stem_match('poetically').should eql(["POET", ["EXPRESSIVE_BEH", "", "EMOTIONS"]])
+      RiDic.stem_match('poetically').should eql(["EXPRESSIVE_BEH", "", "EMOTIONS"])
     end
 
     it 'returns il if stem does not exist in dictionary' do
       RiDic.stem_match('mxyzptlk').should eql(nil)
     end
   end
+
+  describe '.all_categories_in_document' do
+    let(:document_text) {"splendid and sublime ale critiqued and bit the harlot of an apple"}
+    let(:categories_result){RiDic.all_categories_in_document(document_text)}
+    
+    it 'returns an element for every word in the document' do
+      categories_result.length.should eql(document_text.split.length)
+    end
+
+    it 'correctly evaluates the first dictionary item' do
+      categories_result.first.should eql(["GLORY", "", "EMOTIONS"])
+    end
+  end
+
+  describe '.category_in_document' do
+    context 'category 1' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit"}
+      let(:result) {RiDic.category_in_document(document_text, 1)}
+
+      it 'returns an element for each dictionary match' do
+        result.length.should eql(6)
+      end
+
+      it 'returns a collection of elements that each also contain a single element' do      
+        result.first.length.should eql(1)      
+      end
+
+      it 'correctly evaluates the first items category to be GLORY' do
+        result.first.should eql(['GLORY'])    
+      end
+
+      it 'correctly evaluates the last items category to be SOCIAL_BEHAVIOR' do
+        result[-1].should eql(['SOCIAL_BEHAVIOR'])
+      end
+    end
+    context 'category 2' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit"}
+      let(:result) {RiDic.category_in_document(document_text, 2)}
+
+      it 'correctly evaluates the first items category to be GLORY' do
+        result.first.should eql(['NEED'])    
+      end
+    end
+    context 'category 3' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit"}
+      let(:result) {RiDic.category_in_document(document_text, 3)}
+
+      it 'correctly evaluates the first items category to be GLORY' do
+        result.first.should eql(['EMOTIONS'])    
+      end
+    end
+
+  end
+
+  describe '.category_distribution' do
+    context 'provided with a list including dictionary words - category 1' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit advisors"}    
+      let(:result) {RiDic.category_distribution(document_text, 1)}
+
+      it 'returns an element for each category match' do
+        result.length.should eql(4)
+      end
+
+      it 'evaluates the correct number of words associated with AFFECTION' do
+        result["AFFECTION"].should eql(2)
+      end
+
+      it 'evaluates the correct number of words associated with AFFECTION' do
+        result["SOCIAL_BEHAVIOR"].should eql(2)
+      end
+    end
+    
+    context 'provided a list including no dictionary words - category 1' do
+      let(:document_text) {"humpty dumpty's funky"}    
+      let(:result) {RiDic.category_distribution(document_text, 1)}
+
+      it 'returns no elements' do
+        result.length.should eql(0)
+      end
+    end
+
+    context 'category 2' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit advisors"}    
+      let(:result) {RiDic.category_distribution(document_text, 2)}      
+      
+      it 'evaluates the correct number of words associated with NEED' do
+        result["NEED"].should eql(1)
+      end      
+    end
+
+    context 'category 3' do
+      let(:document_text) {"splendid endearing and cordial sublime ale whatcha callit advisors"}    
+      let(:result) {RiDic.category_distribution(document_text, 3)}      
+      
+      it 'evaluates the correct number of words associated with EMOTIONS' do
+        result["EMOTIONS"].should eql(4)
+      end      
+    end
+  end
+
+  describe '.sanitize' do
+    let(:document_text) {"test, this is a string."}
+    it 'removes non alphabetic chars from a string' do
+      RiDic.sanitize(document_text).should eql("test this is a string")
+    end
+  end
+
+  describe '.sort_distribution' do
+    let(:document_text) {"splendid amorous endearing and cordial sublime ale whatcha callit advisors"}    
+    let(:distribution_set) {RiDic.category_distribution(document_text,1)}
+    let(:result) {RiDic.sort_distribution(distribution_set)}
+    
+    it 'sorts a distribution set based on the value (number of instances)' do
+      result.first.should eql(["AFFECTION", 3])
+    end
+  end
+
 end
 

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: ridic
 version: !ruby/object:Gem::Version
-  version: 0.5.2
+  version: 0.6.1
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-12-17 00:00:00.000000000 Z
+date: 2012-12-19 00:00:00.000000000 Z
 dependencies: []
 description: Ruby gem wrapper for the Regressive Imagery Dictionary
 email: