nlp 0.2.1 → 0.2.2

data/lib/analyzer.rb

@@ -1,12 +1,10 @@
 require 'dictionary'
-require 'morfeusz'
+#require 'morfeusz'
 require 'token'
 require 'word'
 require 'emoticon'
 require 'sentence'
 require "token_scanner.rb"
-require "inflectable"
-require "meaningable"
 
 $KODE = "UTF8"
 
@@ -15,7 +13,6 @@ module NLP
   class Analyzer
 
     include REXML
-   #Lexeme = Apohllo::Morfeusz::Lexeme 
     
     def initialize( category_file, restore = true )
         state_file = File.expand_path(Dictionary::CACHE_DIR)
@@ -46,7 +43,6 @@ module NLP
         unless categories.nil?
             categories.each do |category|
                 
-                puts "#{word} : #{category.name}"
                 results[:scores][category] = results[:scores][category] + 1
             end
             

data/lib/dictionary.rb

@@ -59,7 +59,6 @@ module NLP
               @tree.insert( word, category )
             end
           rescue
-            puts "Error for line: #{line}"
             raise
           end
         end

data/lib/emoticon.rb

@@ -1,5 +1,6 @@
+require 'meaningable'
 module NLP
-class Emoticon < Token
+    class Emoticon < Token
 	include Meaningable
 
 	def initialize(tokens,tags)
@@ -8,6 +9,6 @@ class Emoticon < Token
 	end
 
 
-end
+    end
 end
 

data/lib/inflectable.rb

@@ -2,8 +2,8 @@ module Inflectable
 	
 	GRAM_CAT = { 
 		#rzeczownik
-		[:subst, :depr] => 'rzeczownik',
 		:adj => 'przymiotnik',
+                [:subst,:depr] => 'rzeczownik',
 		:adv => 'przyslowek',
 		:num => 'liczebnik',
 		[:pron,:siebie] => 'zaimek',
@@ -22,15 +22,15 @@ module Inflectable
     		:voc => 'wolacz',
 		
 		#Rodzaje
-    		:m1 => 'męski_osobowy',
-    		:m2 => 'męski_zwierzęcy',
-    		:m3 => 'męski_rzeczowy',
-    		:f => 'żeński',
-    		:n1 => 'nijaki zbiorowy',
-		:n2 => 'nijaki zwykły',
-    		:p1 => 'przymnogi osobowy',
-		:p2 => 'przymnogi zwykły',
-		:p3 => 'przymnogi opisowy',
+    		:m1 => 'meski_osobowy',
+    		:m2 => 'meski_zwierzecy',
+    		:m3 => 'meski_rzeczowy',
+    		:f => 'zenski',
+    		:n1 => 'nijaki_zbiorowy',
+		:n2 => 'nijaki zwykly',
+    		:p1 => 'przymnogi_osobowy',
+		:p2 => 'przymnogi_zwykly',
+		:p3 => 'przymnogi_opisowy',
 		
 		#Osoby	
 		:pri => "pierwsza_osoba",
@@ -44,16 +44,18 @@ module Inflectable
 	}
 
 	      GRAM_CAT.each do |key,value|
-		if key.kind_of? Array
-			key = key.first
-		else
-			define_method(value+"?"){ 
-				inflection.split(":").any?{|e| e.include? key.to_s[1..-1]}
-			}
-		end
+			
+                  define_method(value+"?"){ 
+			    inflection.split(":").any?{|e| 
+                                if key.is_a? Array
+                                    key.any?{|k| e.include? k.to_s}
+                                else 
+                                    e.include? key.to_s 
+                                end
+                            }
+		    }
 	      end
 
 
-	
 	       	
 end

data/lib/sentence.rb

@@ -1,14 +1,12 @@
 module NLP
-class Sentence
-    attr_reader :tokens
-    def initialize()
-        @tokens = []
-    end
+    class Sentence
+        attr_reader :tokens
+        def initialize()
+            @tokens = []
+        end
 
-    def << tokens
-        @tokens.concat tokens
+        def << tokens
+            @tokens.concat tokens
+        end
     end
-
-
-end
 end

data/lib/token.rb

@@ -1,35 +1,34 @@
-require 'inflectable'
+
 module NLP
-class Token
-   attr_reader :orth
-   attr_reader :tags
-
-    
-    def initialize(orth,tags)
-        @orth = orth
-        @tags = tags
-    end
+    class Token
+       attr_reader :orth
+       attr_reader :tags
 
-    def interp?
-        @tags.eql? "interp"
-    end
+        
+        def initialize(orth,tags)
+            @orth = orth
+            @tags = tags
+        end
 
-    def word?
-        not interp? and not number?
-    end
+        def interp?
+            @tags.eql? "interp"
+        end
 
-    def number?
-        @tags.include?("tnum")
-    end
+        def word?
+            not interp? and not number?
+        end
 
-    def integer?
-        @tags.include?("tnum:integer")
-    end
+        def number?
+            @tags.include?("tnum")
+        end
 
-    def float?
-        @tags.include?("tnum:frac")
-    end
+        def integer?
+            @tags.include?("tnum:integer")
+        end
 
+        def float?
+            @tags.include?("tnum:frac")
+        end
 
-end
+    end
 end

data/lib/token_scanner.rb

@@ -1,137 +1,136 @@
 
 require 'rexml/document'
 require 'soap/rpc/driver'
-module NLP
-class TokenScanner
-include REXML
-  attr_reader :text, :tokens
 
-    def initialize(text, method)
-        @pos = 0
+module NLP
+    class TokenScanner
+        include REXML
         
-        if method === :file
-            puts "laduje tekst"
-            @text = load_lemated_text(text)
-        elsif method === :text
-            @text = lematize_text(text)
-        else
-            @text = text
+        attr_reader :text, :tokens
+
+        def initialize(text, method)
+            
+            if method === :takipi
+                @text = load_lemated_text(text)
+            elsif method === :morfeusz
+                @text = lematize_text(text)
+            else
+                @text = text
+            end
+            
+            @pos = 0
+            @tokens = flatten_text(@text)
         end
 
-        @tokens = flatten_text(@text)
-    end
+        def next(type)
+            @pos+=1
 
-    def next(type)
-        @pos+=1
-        case type
-        when :word
-            while @pos < @tokens.size and !@tokens[@pos].word?
-                @pos+= 1
-            end
+            case type
+            when :word
+                while @pos < @tokens.size and !@tokens[@pos].word?
+                    @pos+= 1
+                end
 
-        when :interp
-            while @pos < @tokens.size and !@tokens[@pos].interp?
-                @pos+= 1
-            end
-           
-         when :number
-            while @pos < @tokens.size and !@tokens[@pos].number?
-                @pos+= 1
+            when :interp
+                while @pos < @tokens.size and !@tokens[@pos].interp?
+                    @pos+= 1
+                end
+               
+             when :number
+                while @pos < @tokens.size and !@tokens[@pos].number?
+                    @pos+= 1
+                end
             end
-           
         end
-    end
 
-    def current
-        
-        if @pos == @tokens.size
+        def current
+            
+            if @pos == @tokens.size
                 nil
-        else
+            else
                 @tokens[@pos]
-        end
-
-    end
-
-    def index
-        @pos
-    end
-
-    def end?
-        @pos == tokens.size
-    end
-	   
-
-    private 
-
-    def flatten_text(text)
-        flattened = []
-        text.each { |s| s.tokens.each {|t| flattened.push t } }
-        flattened
-    end
-    
-   def load_lemated_text(text_file)
-
-        	t1 = Thread.new do 
-          	`takipi -i #{text_file} -o output.xml -it TXT`
-        	end
-        	t1.join
+            end
 
+        end
 
+        def index
+            @pos
+        end
 
-		
+        def end?
+            @pos == tokens.size
+        end
+               
 
-       text = []
-       File.open("output.xml") do |f|
-           doc = Document.new(f)
+        private 
 
-           doc.elements.each("*/chunkList/chunk") do |chunk| 
-                sentence = Sentence.new
-                tokens = []
+        def flatten_text(text)
+            flattened = []
+            text.each { |s| s.tokens.each {|t| flattened.push t } }
+            flattened
+        end
 
-                chunk.elements.each("tok") do |tok|
-                   word = tok.elements[1].text
-                   lemat, inflect = ""
+        #Tok
+        
+       def load_lemated_text(text)
 
-                   tok.elements.each("lex") do |lex|
-                        if lex.has_attributes?
-                            lemat = lex.elements[1].text
-                            inflect = lex.elements[2].text
-                        end
+            t1 = Thread.new do 
+                `echo #{text} | takipi -i  -o output.xml -it TXT`
+            end
+            t1.join
+
+           text = []
+           File.open("output.xml") do |f|
+               doc = Document.new(f)
+
+               doc.elements.each("*/chunkList/chunk") do |chunk| 
+                    sentence = Sentence.new
+                    tokens = []
+
+                    chunk.elements.each("tok") do |tok|
+                       word = tok.elements[1].text
+                       lemat, inflect = ""
+
+                       tok.elements.each("lex") do |lex|
+                            if lex.has_attributes?
+                                lemat = lex.elements[1].text
+                                inflect = lex.elements[2].text
+                            end
+                       end
+                      
+                       tokens << Word.new(word,lemat,inflect)
                    end
-                  
-                   tokens << Word.new(word,lemat,inflect)
-               end
 
-                sentence << tokens
-                text << sentence
+                    sentence << tokens
+                    text << sentence
+            end
         end
-    end
-    text
-    end 
-
-   def lematize_text(text)
-        temp_text = []
-       text.split(/\.|!|\?/).each do |s|
-            sentence = Sentence.new
-            sentence << s.split(" ").collect{ |t|
-                if word = Morfeusz::Lexeme.find(t)
-                   if word[0]
-                        Word.new(t,word[0].base_form,"") 
-                   else
+        text
+        end 
+
+       def lematize_text(text)
+            temp_text = []
+           text.split(/\.|!|\?/).each do |s|
+                sentence = Sentence.new
+                sentence << s.split(" ").collect{ |t|
+                    if word = Morfeusz::Lexeme.find(t)
+                       if word[0]
+                            Word.new(t,word[0].base_form,"") 
+                       else
+                            Word.new(t,"","")
+                       end
+                    else
                         Word.new(t,"","")
-                   end
-                else
-                    Word.new(t,"","")
-                end
-            }
-            temp_text.push  sentence
+                    end
+                }
+                temp_text.push  sentence
+           end
+           temp_text
        end
-       temp_text
-   end
 
 
 
 
-end
+    end
 
 end

data/lib/word.rb

@@ -2,20 +2,20 @@ require 'inflectable'
 require 'meaningable'
 
 module NLP
-class Word < Token
-	include Inflectable
-	include Meaningable
+    class Word < Token
+        include Inflectable
+        include Meaningable
 
         attr_reader :lemat, :orth
-	
-	def initialize(word, lemat, tags)
+        
+        def initialize(word, lemat, tags)
             super(word,tags)
             @lemat = lemat
-	end
+        end
 
         def inflection
             @tags
         end
 
-end
+    end
 end

data/test/word_test.rb

@@ -0,0 +1,42 @@
+require 'helper'
+require '../lib/word.rb'
+class WordTest < Test::Unit::TestCase
+    def setup 
+        @word_kota = NLP::Word.new('kota','kot','subst:sg:gen.acc:m2')
+        @word_siebie = NLP::Word.new('siebie','się','siebie:gen.acc')
+    end
+
+      def test_word_lematization
+            assert_equal 'kot', @word_kota.lemat 
+            assert_equal 'się', @word_siebie.lemat
+      end
+
+      def test_word_orth
+          assert_equal 'kota', @word_kota.orth
+          assert_equal 'siebie', @word_siebie.orth
+      end
+
+      def test_recognizing_part_of_speech
+          assert @word_kota.rzeczownik?
+          assert @word_siebie.zaimek?
+      end
+
+     def test_recognizing_inflection
+        assert @word_kota.liczba_pojedyncza?
+        assert @word_kota.dopelniacz?
+        assert @word_kota.biernik?
+        assert @word_kota.meski_zwierzecy?
+
+        assert_equal false,  @word_kota.liczba_mnoga?
+        assert_equal false, @word_kota.mianownik?
+
+        assert @word_siebie.biernik?
+        assert @word_siebie.dopelniacz?
+     end
+
+     def test_inflection_string
+         assert_equal  @word_kota.inflection, 'subst:sg:gen.acc:m2'
+     end
+
+
+end

metadata

@@ -1,13 +1,13 @@
 --- !ruby/object:Gem::Specification 
 name: nlp
 version: !ruby/object:Gem::Version 
-  hash: 21
+  hash: 19
   prerelease: false
   segments: 
   - 0
   - 2
-  - 1
-  version: 0.2.1
+  - 2
+  version: 0.2.2
 platform: ruby
 authors: 
 - knife
@@ -15,7 +15,7 @@ autorequire:
 bindir: bin
 cert_chain: []
 
-date: 2010-09-04 00:00:00 +02:00
+date: 2010-09-06 00:00:00 +02:00
 default_executable: 
 dependencies: []
 
@@ -51,6 +51,7 @@ files:
 - README.rdoc
 - test/helper.rb
 - test/test_nlp.rb
+- test/word_test.rb
 has_rdoc: true
 homepage: http://github.com/knife/nlp
 licenses: []
@@ -88,3 +89,4 @@ summary: Linguistics tools for processing polish language.
 test_files: 
 - test/helper.rb
 - test/test_nlp.rb
+- test/word_test.rb