nlp_toolz 1.0.5 → 1.1.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: af17d3c95ae954aa92e06936825db1e051e6a3b5
-  data.tar.gz: 1248afa56f823c06d39c8121dcacb1759ebfff75
+  metadata.gz: ce5f4cad49039b0d8cb6d626facc67a4efa32ae4
+  data.tar.gz: 0565742385f0a34aabe4e456cde014ba2673a589
 SHA512:
-  metadata.gz: fb31bdd348f1a32e8f979b84787b1aec0ef1170f3a9c5e150968f1cb88f9ad91ac3b1ef7893bdfdd05a9b31e90e3c79764ed98d0a607a49f7a8766d0c7859596
-  data.tar.gz: 7a95371db1dc274e08478f9cf27f13badea7b97fd04e5963ac2209a66f07d227e0b6b8d9d12646e266a7d3c53bb3eac48db4e1a0f8c4737e799d1db82c095e67
+  metadata.gz: 1ec11ec4b9b07437fb16f9ab0c181c9cee40a0cc900f90d02d2a6e4fc3bac7efaae890e8eda16bf7dcf8e3595bcb4010cf9d3893bee2a7a937b0fd527c40356f
+  data.tar.gz: 06d53b1bfe11004d0abeba1db130a13f664a054e8ed56f5edb260ec3f8bf189b0f9cb64687a471d2241ffd0d612ae632b50853737282a8d1901ea0645be4426a

data/Guardfile

@@ -6,7 +6,7 @@ guard :bundler do
   watch(/^.+\.gemspec/)
 end
 
-guard :rspec do
+guard :rspec, cmd: 'bundle exec rspec' do
   watch(%r{^spec/.+_spec\.rb$})
   watch(%r{^lib/(.+)\.rb$})     { |m| "spec/lib/#{m[1]}_spec.rb" }
   watch('spec/spec_helper.rb')  { "spec" }

data/lib/nlp_toolz/language.rb

@@ -0,0 +1,52 @@
+# coding:  utf-8
+# author LeFnord
+# email  pscholz.le@gmail.com
+# date   2014-10-12
+
+module NlpToolz
+  
+  class Language
+    
+    # load java classes
+    # Enumeration = Rjb::import("java.util.Enumeration")
+    HashSet     = Rjb::import("java.util.HashSet")
+    # Hashtable   = Rjb::import("java.util.Hashtable")
+    # Set         = Rjb::import("java.util.Set")
+    
+    DataSourceException = Rjb::import("de.uni_leipzig.asv.toolbox.jLanI.kernel.DataSourceException")
+    LanIKernel = Rjb::import("de.uni_leipzig.asv.toolbox.jLanI.kernel.LanIKernel")
+    Request = Rjb::import("de.uni_leipzig.asv.toolbox.jLanI.kernel.Request")
+    RequestException = Rjb::import("de.uni_leipzig.asv.toolbox.jLanI.kernel.RequestException")
+    Response = Rjb::import("de.uni_leipzig.asv.toolbox.jLanI.kernel.Response")
+    
+    def self.get_language(text = nil)
+      return -1 if text.nil? || text.empty?
+      lang_probability = identify text
+      lang_probability.first
+    end
+    
+    # set language and probability of sentence
+    def self.identify(text)
+      languages = HashSet.new
+      modus = 0
+      reduce = true
+      
+      req = Request.new(text, languages, modus, reduce)
+      
+      LanIKernel.propertyFile = File.join(MODELS, 'language', 'lanikernel')
+      kernel = LanIKernel.getInstance()
+      res = kernel.evaluate(req)
+      
+      @lang,@probability = get_most_probability_lang(res.getResult.toString)
+    end
+    
+    private
+    def self.get_most_probability_lang(result)
+      res = []
+      foo = result.sub!("{","").sub!("}","").split(', ').collect{ |x| x.split('=') }
+      foo.each{ |x| res << [x.first,x.last.to_f] }
+      res.max{|a,b| a.last <=> b.last}
+    end
+  end
+
+end

data/lib/nlp_toolz/load_jars.rb

@@ -2,11 +2,13 @@ module NlpToolz
   MODELS = File.join(File.dirname(__FILE__), '..', '..', "models")
   JARS = File.join(File.dirname(__FILE__), '..', '..', "jars")
   
-  CLASS_PATH = [
-    File.join(JARS, "jwnl-1.3.3.jar"),
-    File.join(JARS, "opennlp-tools-1.5.3.jar"),
-    File.join(JARS, "opennlp-maxent-3.0.3.jar")
-  ].join(":")
+  # CLASS_PATH = [
+  #   File.join(JARS, "jwnl-1.3.3.jar"),
+  #   File.join(JARS, "opennlp-tools-1.5.3.jar"),
+  #   File.join(JARS, "opennlp-maxent-3.0.3.jar")
+  # ].join(":")
+  
+  CLASS_PATH = Dir.glob(File.join(JARS,'*.jar')).join(':')
   
   Rjb::load(CLASS_PATH,['-X+C','-Xmx4096m','-Djava.awt.headless=true'])
   # Rjb::load(CLASS_PATH,['-Xmx4096m','-XX:+UseParallelGC','-XX:+UseParallelOldGC','-Djava.awt.headless=true'])

data/lib/nlp_toolz/parser.rb

@@ -4,12 +4,11 @@
 # date:   2012-12-10
 
 module NlpToolz
-
+  
   class Parser
-
-    include Lang
+    
     include TmpFile
-
+    
     # load java classes
     FileInputStream = Rjb::import('java.io.FileInputStream')
     
@@ -18,7 +17,7 @@ module NlpToolz
     
     def initialize(input, lang = nil)
       @input = input
-      @lang = lang || get_language
+      @lang = lang || NlpToolz::Language.get_language(input)
       @model_name = "#{@lang}-sm5.gr"
       get_model
     end

data/lib/nlp_toolz/pos_tags.rb

@@ -8,8 +8,6 @@ module NlpToolz
 
   class PosTags
     
-    include Lang
-    
     # load java classes
     FileInputStream = Rjb::import('java.io.FileInputStream')
     POSModel        = Rjb::import('opennlp.tools.postag.POSModel')
@@ -19,7 +17,7 @@ module NlpToolz
     
     def initialize(input, lang = nil)
       @input = input
-      @lang = lang || get_language
+      @lang = lang || NlpToolz::Language.get_language(input)
       @model_name = "#{@lang}-pos-maxent.bin"
       get_model
     end

data/lib/nlp_toolz/sentences.rb

@@ -8,8 +8,6 @@ module NlpToolz
 
   class Sentences
     
-    include Lang
-    
     # load java classes
     FileInputStream    = Rjb::import('java.io.FileInputStream')
     SentenceDetectorME = Rjb::import('opennlp.tools.sentdetect.SentenceDetectorME')
@@ -19,7 +17,7 @@ module NlpToolz
     
     def initialize(input,lang = nil)
       @input = input
-      @lang = lang || get_language
+      @lang = lang || NlpToolz::Language.get_language(input)
       @model_name = "#{@lang}-sent.bin"
       get_model
     end

data/lib/nlp_toolz/tokens.rb

@@ -7,8 +7,6 @@ module NlpToolz
 
   class Tokens
     
-    include Lang
-    
     # load java classes
     FileInputStream = Rjb::import('java.io.FileInputStream')
     TokenizerModel  = Rjb::import('opennlp.tools.tokenize.TokenizerModel')
@@ -18,7 +16,7 @@ module NlpToolz
     
     def initialize(input, lang = nil)
       @input = input
-      @lang = lang || get_language
+      @lang = lang || NlpToolz::Language.get_language(input)
       @model_name = "#{@lang}-token.bin"
       get_model
     end

data/lib/nlp_toolz/version.rb

@@ -4,5 +4,5 @@
 # date:   2012-10-23
 
 module NlpToolz
-  VERSION = "1.0.5"
+  VERSION = "1.1.0"
 end

data/lib/nlp_toolz.rb

@@ -13,24 +13,22 @@ require "multi_json"
 # internal requirements
 require "nlp_toolz/version"
 require "nlp_toolz/helpers/url_handler"
-require "nlp_toolz/helpers/lang"
 require "nlp_toolz/helpers/string_extended"
 require "nlp_toolz/helpers/tmp_file"
 
 # NLP Tools
 require "nlp_toolz/load_jars"
+require "nlp_toolz/language"
 require "nlp_toolz/sentences"
 require "nlp_toolz/pos_tags"
 require "nlp_toolz/tokens"
 require "nlp_toolz/parser"
 
 module NlpToolz
-  extend Lang
-  
   module_function
   
   def get_lang(input)
-    NlpToolz.get_language(input)
+    NlpToolz::Language.get_language(input)
   end
   
   def get_sentences(input,lang = nil)

data/spec/helpers/string_extended_spec.rb

@@ -7,10 +7,10 @@ describe String do
   end
   
   it "should delete quotations marks" do
-    @a.join("").clean_up.should be_empty
+    expect(@a.join("").clean_up).to be_empty
     chars =  (@a.length - 1) * 3
-    @a.join(" ap").clean_up.length.should == chars
-    @a.join("ap ").clean_up.length.should == chars                      
+    expect(@a.join(" ap").clean_up.length).to be == chars
+    expect(@a.join("ap ").clean_up.length).to be == chars
   end
   
   

data/spec/lib/nlp_toolz/language_spec.rb

@@ -0,0 +1,23 @@
+require "spec_helper"
+
+describe 'Language' do
+  
+  before(:all) do
+    @en_text = "Military historian Basil Liddell Hart famously declared that Sherman was the first modern general."
+    @de_text = "Die erste Ausgabe der von Arwidsson herausgegebenen, kurzlebigen Zeitschrift Abo Morgonblad vom 5. Januar 1821."
+  end
+  
+  it 'do nothings if text empty or nil' do
+    res = NlpToolz::Language.get_language
+    expect(res).to be == -1
+    res = NlpToolz::Language.get_language('')
+    expect(res).to be == -1
+  end
+  
+  it 'gets language' do
+    res = NlpToolz::Language.get_language @en_text
+    expect(res).to be == 'en'
+    res = NlpToolz::Language.get_language @de_text
+    expect(res).to be == 'de'
+  end
+end

data/spec/lib/nlp_toolz/parser_spec.rb

@@ -7,44 +7,44 @@ describe NlpToolz do
       @text = "Military historian Basil Liddell Hart famously declared that Sherman was the first modern general."
       @g_text = "μακεδονικού εκκεντροφόρου πολιτισμός του. την ανάφλεξης πολιτισμική. πολιτισμού του να. τόπος επειδή σε. καθορίσουν χρόνια Στα από."
     end
-    
+
     describe "attributes" do
       it "should respond to #attribute" do
         text = NlpToolz::Parser.new(@text)
-        text.should respond_to(:input)
-        text.should respond_to(:lang)
-        text.should respond_to(:model_name)
-        text.should respond_to(:model)
-        text.should respond_to(:parse_hash)
+        expect(text).to respond_to(:input)
+        expect(text).to respond_to(:lang)
+        expect(text).to respond_to(:model_name)
+        expect(text).to respond_to(:model)
+        expect(text).to respond_to(:parse_hash)
       end
     end
-    
+
     describe "model" do
       it "should have a model, if lang 'en'" do
         sent = NlpToolz::Parser.new(@text,'en')
-        sent.model_name.should == 'en-sm5.gr'
-        sent.has_model?.should be_true
+        expect(sent.model_name).to be == 'en-sm5.gr'
+        expect(sent.has_model?).to be_truthy
       end
-      
+
       it "should not have a model, if lang not known" do
         sent = NlpToolz::Parser.new(@g_text)
-        sent.has_model?.should be_false
+        expect(sent.has_model?).to be_falsey
       end
     end
-    
+
     describe "object" do
       it "should create a valid object" do
         expect{ text = NlpToolz::Parser.new(@text,"en") }.to_not raise_error
       end
-      
+
       it "should set the language of input" do
         text = NlpToolz::Parser.new(@text)
-        text.lang.should == "en"
+        expect(text.lang).to be == "en"
       end
-      
+
       it "should build the right model name" do
         text = NlpToolz::Parser.new(@text)
-        text.model_name.should == "en-sm5.gr"
+        expect(text.model_name).to be == "en-sm5.gr"
       end
     end
 
@@ -52,15 +52,15 @@ describe NlpToolz do
       it "should store tree in a hash" do
         text = NlpToolz::Parser.new(@text)
         text.parse_text
-        text.parse_hash.should be_a(Hash)
+        expect(text.parse_hash).to be_a(Hash)
       end
 
       it "should have a token hash after parsing" do
         text = NlpToolz::Parser.new(@text)
         text.parse_text
-        text.layer.should be_a Hash
-        text.layer.should include(:tags)
-        text.layer.should include(:tokens)
+        expect(text.layer).to be_a Hash
+        expect(text.layer).to include(:tags)
+        expect(text.layer).to include(:tokens)
       end
     end
   end # Parser

data/spec/lib/nlp_toolz/pos_tags_spec.rb

@@ -7,60 +7,58 @@ describe NlpToolz do
       @text = "Military historian Basil Liddell Hart famously declared that Sherman was the first modern general."
       @g_text = "μακεδονικού εκκεντροφόρου πολιτισμός του. την ανάφλεξης πολιτισμική. πολιτισμού του να. τόπος επειδή σε. καθορίσουν χρόνια Στα από."
     end
-    
+
     describe "attributes" do
       it "should respond to #attribute" do
         text = NlpToolz::PosTags.new(@text)
-        text.should respond_to(:input)
-        text.should respond_to(:lang)
-        text.should respond_to(:model_name)
-        text.should respond_to(:model)
-        text.should respond_to(:tokenized)
+        expect(text).to respond_to(:input)
+        expect(text).to respond_to(:lang)
+        expect(text).to respond_to(:model_name)
+        expect(text).to respond_to(:model)
+        expect(text).to respond_to(:tokenized)
       end
     end
-    
+
     describe "model" do
       it "should have a model, if lang 'en'" do
         sent = NlpToolz::PosTags.new(@text,'en')
-        sent.model_name.should == 'en-pos-maxent.bin'
-        sent.has_model?.should be_true
+        expect(sent.model_name).to be == 'en-pos-maxent.bin'
+        expect(sent.has_model?).to be_truthy
       end
-      
+
       it "should not have a model, if lang not known" do
         sent = NlpToolz::PosTags.new(@g_text)
-        sent.has_model?.should be_false
+        expect(sent.has_model?).to be_falsey
       end
     end
-    
+
     describe "object" do
       it "should create a valid object" do
         expect{ text = NlpToolz::PosTags.new(@text,"en") }.to_not raise_error
       end
-      
+
       it "should set the language of input" do
         text = NlpToolz::PosTags.new(@text)
-        text.lang.should == "en"
+        expect(text.lang).to be == "en"
       end
-      
+
       it "should build the right model name" do
         text = NlpToolz::PosTags.new(@text)
-        text.model_name.should == "en-pos-maxent.bin"
+        expect(text.model_name).to be == "en-pos-maxent.bin"
       end
-      
+
       it "should be a hash after pos tagging" do
         text = NlpToolz::PosTags.new(@text,"en")
         text.get_pos_tags
-        text.tokenized.should include(:tokens)
-        text.tokenized.should include(:tags)
-        text.tokenized.should be_a Hash
+        expect(text.tokenized).to include(:tokens)
+        expect(text.tokenized).to include(:tags)
+        expect(text.tokenized).to be_a Hash
       end
-      
+
       it "should get pos text of given text" do
         text = NlpToolz::PosTags.new(@text,"en")
         text.get_pos_tags
-        text.tokenized[:tokens].should have(15).items
-        text.tokenized[:tags].should have(15).items
-        text.tokenized[:tokens].length.should == text.tokenized[:tags].length
+        expect(text.tokenized[:tokens].length).to be == text.tokenized[:tags].length
       end
     end
   end # POS Tags

data/spec/lib/nlp_toolz/sentences_spec.rb

@@ -7,53 +7,53 @@ describe NlpToolz do
       @text = "William Tecumseh Sherman (February 8, 1820 – February 14, 1891) was an American soldier, businessman, educator, and author.
                He served as a general in the United States Army during the American Civil War (1861–65), receiving both recognition for his outstanding command of military strategy, and criticism for the harshness of the scorched earth policies he implemented in conducting total war against the Confederate States of America.
                Military historian Basil Liddell Hart famously declared that Sherman was the first modern general."
-      
+
       @g_text = "μακεδονικού εκκεντροφόρου πολιτισμός του. την ανάφλεξης πολιτισμική. πολιτισμού του να. τόπος επειδή σε. καθορίσουν χρόνια Στα από."
     end
-    
+
     describe "attributes" do
       it "should respond to #attribute" do
         sent = NlpToolz::Sentences.new(@text)
-        sent.should respond_to(:input)
-        sent.should respond_to(:lang)
-        sent.should respond_to(:model_name)
-        sent.should respond_to(:model)
-        sent.should respond_to(:sentences)
+        expect(sent).to respond_to(:input)
+        expect(sent).to respond_to(:lang)
+        expect(sent).to respond_to(:model_name)
+        expect(sent).to respond_to(:model)
+        expect(sent).to respond_to(:sentences)
       end
     end
-    
+
     describe "model" do
       it "should have a model, if lang 'en'" do
         sent = NlpToolz::Sentences.new(@text,'en')
-        sent.has_model?.should be_true
-        sent.model_name.should == 'en-sent.bin'
+        expect(sent.has_model?).to be_truthy
+        expect(sent.model_name).to be == 'en-sent.bin'
       end
-      
+
       it "should not have a model, if lang not known" do
         sent = NlpToolz::Sentences.new(@g_text)
-        sent.has_model?.should be_false
+        expect(sent.has_model?).to be_falsey
       end
     end
-    
+
     describe "object" do
       it "should create a valid object" do
         expect{ sent = NlpToolz::Sentences.new(@text) }.to_not raise_error
       end
-    
+
       it "should set the language of input" do
         sent = NlpToolz::Sentences.new(@text)
-        sent.lang.should == "en"
+        expect(sent.lang).to be == "en"
       end
-    
+
       it "should build the right model name" do
         sent = NlpToolz::Sentences.new(@text)
-        sent.model_name.should == "en-sent.bin"
+        expect(sent.model_name).to be == "en-sent.bin"
       end
-    
+
       it "should split incoming text into sentences" do
         text = NlpToolz::Sentences.new(@text,"en")
         text.split_into_sentences
-        text.sentences.should have(3).items
+        expect(text.sentences.length).to be == 3
       end
     end
   end # Sentences

data/spec/lib/nlp_toolz/tokens_spec.rb

@@ -7,55 +7,55 @@ describe NlpToolz do
       @text = "Military historian Basil Liddell Hart famously declared that Sherman was the first modern general."
       @g_text = "μακεδονικού εκκεντροφόρου πολιτισμός του. την ανάφλεξης πολιτισμική. πολιτισμού του να. τόπος επειδή σε. καθορίσουν χρόνια Στα από."
     end
-    
+
     describe "attributes" do
       it "should respond to #attribute" do
         text = NlpToolz::Tokens.new(@text)
-        text.should respond_to(:input)
-        text.should respond_to(:lang)
-        text.should respond_to(:model_name)
-        text.should respond_to(:model)
-        text.should respond_to(:tokens)
+        expect(text).to respond_to(:input)
+        expect(text).to respond_to(:lang)
+        expect(text).to respond_to(:model_name)
+        expect(text).to respond_to(:model)
+        expect(text).to respond_to(:tokens)
       end
     end
-    
+
     describe "model" do
       it "should have a model, if lang 'en'" do
         sent = NlpToolz::Tokens.new(@text,'en')
-        sent.has_model?.should be_true
+        expect(sent.has_model?).to be_truthy
       end
-      
+
       it "should not have a model, if lang not known" do
         sent = NlpToolz::Tokens.new(@g_text)
-        sent.has_model?.should be_false
+        expect(sent.has_model?).to be_falsey
       end
     end
-    
+
     describe "object" do
       it "should create a valid object" do
         expect{ text = NlpToolz::Tokens.new(@text,"en") }.to_not raise_error
       end
-    
+
       it "should set the language of input" do
         text = NlpToolz::Tokens.new(@text)
-        text.lang.should == "en"
+        expect(text.lang).to be == "en"
       end
-    
+
       it "should build the right model name" do
         text = NlpToolz::Tokens.new(@text)
-        text.model_name.should == "en-token.bin"
+        expect(text.model_name).to be == "en-token.bin"
       end
-    
+
       it "should be a arrar after tokenizing" do
         text = NlpToolz::Tokens.new(@text,"en")
         text.tokenize
-        text.tokens.should be_a Array
+        expect(text.tokens).to be_a Array
       end
-    
+
       it "should tokenize given text" do
         text = NlpToolz::Tokens.new(@text,"en")
         text.tokenize
-        text.tokens.should have(15).items
+        expect(text.tokens.length).to be == 15
       end
     end
   end # Tokens

data/spec/lib/nlp_toolz_spec.rb

@@ -11,33 +11,33 @@ describe NlpToolz do
   describe "detect language" do
     it "should description" do
       lang = NlpToolz.get_lang(@text)
-      lang.should == 'en'
+      expect(lang).to be == 'en'
     end
   end
   
   describe "sentence detection" do
     it "should input text split into its sentences" do
       sentences = NlpToolz.get_sentences(@text)
-      sentences.should have(3).items
+      expect(sentences.length).to be == 3
     end
     
     it "should be 'nil', if text lang is unsupported" do
       sentences = NlpToolz.get_sentences(@g_text)
-      sentences.should be_nil
+      expect(sentences).to be_nil
     end
   end
   
   describe "tokenizing" do
     it "should tag a sentence" do
       tokens = NlpToolz.tokenize_sentence(@sentence)
-      tokens.should have(26).items
-      tokens.should be_a Array
+      expect(tokens.length).to be == 26
+      expect(tokens).to be_a Array
     end
     
     it "should tokenize a whole text" do
       token_arr = NlpToolz.tokenize_text(@text)
-      token_arr.should have(3).items
-      token_arr.first.should have(26).items
+      expect(token_arr.length).to be == 3
+      expect(token_arr.first.length).to be == 26
     end
   end
   
@@ -45,12 +45,12 @@ describe NlpToolz do
     it "should tag a sentence" do
       sentence = NlpToolz.get_sentences(@sentence).last
       tags = NlpToolz.tag_sentence(sentence)
-      tags[:tokens].length.should == tags[:tags].length
+      expect(tags[:tokens].length).to be == tags[:tags].length
     end
     
     it "should be 'nil', if sentence language not supported " do
       tags = NlpToolz.tag_sentence(@g_text)
-      tags.should be_nil
+      expect(tags).to be_nil
     end
   end
   
@@ -58,12 +58,12 @@ describe NlpToolz do
     it "should parse a sentence" do
       sentence = NlpToolz.get_sentences(@sentence).last
       parsed = NlpToolz.parse_sentence(sentence)
-      parsed.should be_a Hash
+      expect(parsed).to be_a Hash
     end
-    
+
     it "should should be 'nil', if sentence language is not supported" do
       parsed = NlpToolz.parse_sentence(@g_text)
-      parsed.should be_nil
+      expect(parsed).to be_nil
     end
   end
 end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: nlp_toolz
 version: !ruby/object:Gem::Version
-  version: 1.0.5
+  version: 1.1.0
 platform: ruby
 authors:
 - LeFnord
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-01-06 00:00:00.000000000 Z
+date: 2014-10-12 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rspec
@@ -209,10 +209,10 @@ files:
 - Rakefile
 - bin/nlp_toolz
 - lib/nlp_toolz.rb
-- lib/nlp_toolz/helpers/lang.rb
 - lib/nlp_toolz/helpers/string_extended.rb
 - lib/nlp_toolz/helpers/tmp_file.rb
 - lib/nlp_toolz/helpers/url_handler.rb
+- lib/nlp_toolz/language.rb
 - lib/nlp_toolz/load_jars.rb
 - lib/nlp_toolz/parser.rb
 - lib/nlp_toolz/pos_tags.rb
@@ -221,6 +221,7 @@ files:
 - lib/nlp_toolz/version.rb
 - nlp_toolz.gemspec
 - spec/helpers/string_extended_spec.rb
+- spec/lib/nlp_toolz/language_spec.rb
 - spec/lib/nlp_toolz/parser_spec.rb
 - spec/lib/nlp_toolz/pos_tags_spec.rb
 - spec/lib/nlp_toolz/sentences_spec.rb
@@ -247,12 +248,13 @@ required_rubygems_version: !ruby/object:Gem::Requirement
       version: '0'
 requirements: []
 rubyforge_project: 
-rubygems_version: 2.2.0
+rubygems_version: 2.2.2
 signing_key: 
 specification_version: 4
 summary: wrapper around the openNLP toolset
 test_files:
 - spec/helpers/string_extended_spec.rb
+- spec/lib/nlp_toolz/language_spec.rb
 - spec/lib/nlp_toolz/parser_spec.rb
 - spec/lib/nlp_toolz/pos_tags_spec.rb
 - spec/lib/nlp_toolz/sentences_spec.rb

data/lib/nlp_toolz/helpers/lang.rb

@@ -1,26 +0,0 @@
-module Lang
-
-  include UrlHandler
-
-  def get_language(text = nil)
-    uri = build_url("arielle.tm.informatik.uni-leipzig.de", 55700, "/langid", nil)
-
-    if @input
-      asv_response = post_data(URI.escape(@input),uri,{'Content-type'=>'text/plain;charset=utf-8'})
-    elsif text
-      asv_response = post_data(URI.escape(text),uri,{'Content-type'=>'text/plain;charset=utf-8'})
-    end
-    response = MultiJson.load(asv_response.body)
-
-    response["lang"]
-  end
-  
-  # ToDo 2013-02-26: make different lang identifier available
-  def alternative_langs lang
-    langs = {
-      en: [:eng, :english],
-      de: [:ger, :german]
-    }.each.collect{|x| x.flatten}
-  end
-
-end