text_nlp 0.0.1 → 0.0.2

data/lib/text_nlp/string.rb

@@ -33,4 +33,8 @@ class String
     score
   end
   
+  def translate(translator)
+    translator.translate(self)
+  end
+  
 end

data/lib/text_nlp/synonyms.rb

@@ -0,0 +1,28 @@
+# encoding: UTF-8
+
+class TextNlp
+  class Synonyms
+   
+    def initialize(synonyms = [])
+      @synonyms = {}
+      @expressions = Expressions.new
+      synonyms.each do |synos|
+        name = synos.shift
+        register(name,synos)
+      end
+    end
+    
+    def register(name,synonyms)
+      normalized_name = name.normalize
+      synonyms.each do |synonym|
+        @expressions << synonym
+        @synonyms[synonym.normalize] = normalized_name
+      end
+    end
+    
+    def translate(text)
+      @expressions.expressionize(text).map { |expr| @synonyms.key?(expr) ? @synonyms[expr] : expr }.join(' ')
+    end
+  
+  end
+end

data/lib/text_nlp.rb

@@ -7,6 +7,7 @@ require "text_nlp/tokenizer.rb"
 require "text_nlp/string.rb"
 require "text_nlp/expressions.rb"
 require "text_nlp/pattern.rb"
+require "text_nlp/synonyms.rb"
 
 class TextNlp
 end

data/spec/string_spec.rb

@@ -25,6 +25,13 @@ describe String do
     text.tokenize
   end
   
+  it "should call translator" do
+    text = "TOTO"
+    translator = double()
+    translator.should_receive(:translate).with(text)
+    text.translate(translator)
+  end
+  
   it "should compute similarity" do
     "il fait chaud".similarity("il fait chaud").should eq 1.0
     "il fait chaud".similarity("putin c nul ici").should eq 0.0

data/spec/synonyms_spec.rb

@@ -0,0 +1,23 @@
+# encoding: utf-8
+require "spec_helper"
+
+describe TextNlp::Synonyms do
+  
+  it "should synonymize the text" do
+    synonyms = TextNlp::Synonyms.new
+    synonyms.register("CAEN",["smc","sm caen","stade malherbe de caen"])
+    synonyms.translate("le smc c est de la bombe").should eq "le caen c est de la bombe"
+    synonyms.translate("le truc c est de la bombe").should eq "le truc c est de la bombe"
+    synonyms.translate("le sm caen c est de la bombe").should eq "le caen c est de la bombe"
+    synonyms.translate("le stade malherbe de caen c est de la bombe").should eq "le caen c est de la bombe"
+  end
+  
+  it "should synonymize the text" do
+    synonyms = TextNlp::Synonyms.new([["CAEN","smc","sm caen","stade malherbe de caen"],["marseille","om"]])
+    synonyms.translate("le smc c est de la bombe").should eq "le caen c est de la bombe"
+    synonyms.translate("le truc c est de la bombe").should eq "le truc c est de la bombe"
+    synonyms.translate("le sm caen c est de la bombe").should eq "le caen c est de la bombe"
+    synonyms.translate("le stade malherbe de caen c est de la bombe").should eq "le caen c est de la bombe"
+  end
+  
+end

data/text_nlp.gemspec

@@ -1,6 +1,6 @@
 Gem::Specification.new do |s|
   s.name          = 'text_nlp'
-  s.version       = '0.0.1'
+  s.version       = '0.0.2'
   s.date          = '2011-07-05'
   s.summary       = "A minimalist NLP library"
   s.description   = s.summary

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: text_nlp
 version: !ruby/object:Gem::Version
-  version: 0.0.1
+  version: 0.0.2
   prerelease: 
 platform: ruby
 authors:
@@ -30,12 +30,14 @@ files:
 - lib/text_nlp/normalizer.rb
 - lib/text_nlp/pattern.rb
 - lib/text_nlp/string.rb
+- lib/text_nlp/synonyms.rb
 - lib/text_nlp/tokenizer.rb
 - spec/expressions_spec.rb
 - spec/normalizer_spec.rb
 - spec/pattern_spec.rb
 - spec/spec_helper.rb
 - spec/string_spec.rb
+- spec/synonyms_spec.rb
 - spec/tokenizer_spec.rb
 - text_nlp.gemspec
 homepage: http://github.com/fonzo14/text_nlp