vn_tagger 0.1.0 → 0.2.0

checksums.yaml

@@ -1,7 +1,7 @@
 ---
 SHA1:
-  metadata.gz: eb1eff57d9490c011b4d340c0d167a20245b8c5c
-  data.tar.gz: b411874dc1e8f7b574cc8995224f6f091bbb5206
+  metadata.gz: 637f23fe16d647e9041131360de3ffd39319c504
+  data.tar.gz: ca68ec84d04a2b0689d4f3dff85a9dd8a0a8f874
 SHA512:
-  metadata.gz: ddce5c31d971c282ba08599a5b49cd38d46ec5e0c178b5b455b47324c941470ef8b4f15fa0f5c8162460105fcaa39344809100a6e2b6f2a856e5df7e6412716b
-  data.tar.gz: 7c96f95660818f66a11607d5056546e188e18cdb7a504babe57bb27eb3fcd4102e244e9768982a442b624e3d4a1929f223c35970dd6349d843a4a2f62f67dc3d
+  metadata.gz: 2430d567ed7dc55a0cea23ffbd0781dfd147afb7576a3593bc71cbef49779c74b737c5d73f88e67101294a42ad5f629f34f9c8a0974e1d34ce7399faded2a21d
+  data.tar.gz: 2caaec9e11213317445b23651cfc64272d35208f6bd256a62a053b01f24750536564b043a9e695672fcf97c2825944e49f3ec83a5b7ddd385e9c5a3c65e30658

data/README.md

@@ -20,11 +20,10 @@ Or install it yourself as:
 text = 'Mourinho là huấn luyện viên của Chelsea'
 document = VnTagger::Tagger.tag(text)
 
-document.to_xml # => "<?xml version=\"1.0\"?>\n<doc>\n\t<s>\n\t\t<w pos=\"Np\">Mourinho</w>\n\t\t<w pos=\"V\">l&#xE0;</w>\n\t\t<w pos=\"N\">hu&#x1EA5;n luy&#x1EC7;n vi&#xEA;n</w>\n\t\t<w pos=\"E\">c&#x1EE7;a</w>\n\t\t<w pos=\"Np\">Chelsea</w>\n\t\t<w pos=\".\">.</w>\n\t</s>\n</doc>\n"
-document.class # => Nokogiri::Xml::Document
-keys = document.xpath("//w")
-keys.first.attr('pos') # => 'Np'
-keys.first.child.text #=> 'Mourinho'
+document.words.map { |word|
+  [word.text, word.position]
+} # => [["Mourinho", "Np"], ["là", "V"], ["huấn luyện viên", "N"], ["của", "E"],
+["Chelsea", "Np"]]
 ```
 
 ## Contributing

data/lib/vn_tagger/document.rb

@@ -0,0 +1,29 @@
+require 'word'
+
+module VnTagger
+  class Document
+    attr_reader :xml_document
+
+    def initialize(xml_document)
+      @xml_document = xml_document
+    end
+
+    def words
+      @words ||= uncached_words
+    end
+
+    def uncached_words
+      xml_document.xpath('//w').map do |element|
+        Word.new(element.attr('pos'), element.child.text)
+      end
+    end
+
+    def filter_by_tag(tag = 'Proper noun')
+      words.select { |word| word.is_tag?(tag) }
+    end
+
+    def proper_noun_words
+      filter_by_tag('Proper noun').map(&:text)
+    end
+  end
+end

data/lib/vn_tagger/tagger.rb

@@ -1,4 +1,5 @@
 require 'nokogiri'
+require 'document'
 
 module VnTagger
   class Tagger
@@ -13,16 +14,28 @@ module VnTagger
 
     def tag
       write_to_file
-      wasGood = system("cd #{ROOT_PATH}; #{COMMAND} -i #{INPUT} -o #{OUTPUT}")
-      if wasGood
-        result_from_output
-      else
-        nil
-      end
+      @success = system("cd #{ROOT_PATH}; #{COMMAND} -i #{INPUT} -o #{OUTPUT}")
+    end
+
+    def xml_result
+      @xml_result ||= if @success
+                        file = File.open(OUTPUT)
+                        xml_document = Nokogiri::XML(file)
+                        file.close
+                        xml_document
+                      else
+                        Nokogiri::XML('')
+                      end
+    end
+
+    def result
+      @result ||= Document.new(xml_result)
     end
 
     def self.tag(text)
-      new(text).tag
+      tagger = new(text)
+      tagger.tag
+      tagger.result
     end
 
     private
@@ -33,13 +46,6 @@ module VnTagger
       file.close
     end
 
-    def result_from_output
-      file = File.open(OUTPUT)
-      doc = Nokogiri::XML(file)
-      file.close
-      doc
-    end
-
     def normalize(string)
       string.to_s.gsub(/(\"|\')/, '')
     end

data/lib/vn_tagger/version.rb

@@ -1,3 +1,3 @@
 module VnTagger
-  VERSION = '0.1.0'
+  VERSION = '0.2.0'
 end

data/lib/vn_tagger/word.rb

@@ -0,0 +1,35 @@
+module VnTagger
+  class Word
+    attr_reader :position, :text
+
+    WORD_POSITIONS = {
+      'Np' => 'Proper noun',
+      'Nc' => 'Classifier',
+      'Nu' => 'Unit noun',
+      'N' => 'Common noun',
+      'V' => 'Verb',
+      'A' => 'Adjective',
+      'P' => 'Pronoun',
+      'R' => 'Adverb',
+      'L' => 'Determiner',
+      'M' => 'Numeral',
+      'E' => 'Preposition',
+      'C' => 'Subordinating conjunction',
+      'CC' => 'Coordinating conjunction',
+      'I' => 'Interjection',
+      'T' => 'Auxiliary, modal words',
+      'Y' => 'Abbreviation',
+      'Z' => 'Bound morphemes',
+      'X' => 'Unknown'
+    }
+
+    def initialize(position, text)
+      @position = position
+      @text = text
+    end
+
+    def is_tag?(tag = 'Proper noun')
+      WORD_POSITIONS[position] == tag
+    end
+  end
+end

data/spec/vn_tagger/document_spec.rb

@@ -0,0 +1,20 @@
+require 'spec_helper'
+
+describe VnTagger::Document do
+  let(:xml_text) { '<doc><s><w pos="Np">HLV</w></s></doc>' }
+  let(:xml_document) { Nokogiri::XML(xml_text) }
+  let(:document) { described_class.new(xml_document) }
+
+  describe '#words' do
+    it 'returns data extract from xml document' do
+      word = document.words.first
+      expect(word.text).to eq 'HLV'
+      expect(word.position).to eq 'Np'
+    end
+  end
+
+  describe '#proper_noun_words' do
+    subject { document.proper_noun_words }
+    it { is_expected.to eq ['HLV'] }
+  end
+end

data/spec/vn_tagger/tagger_spec.rb

@@ -4,13 +4,14 @@ describe VnTagger::Tagger do
   describe '#tag' do
     let(:text) { 'HLV cùa Chelsea không hối tiếc vì hành động bỏ về sớm trong trận gặp Aston Villa.' }
     let(:tagger) { described_class.new(text) }
-    let(:result) { tagger.tag }
-    let(:key) { result.xpath("//w").first }
+    let(:result) { tagger.result }
+    let(:word) { result.words.first }
 
     it 'returns xml tagged text' do
-      expect(result).to be_a(Nokogiri::XML::Document)
-      expect(key.attr('pos')).to eq 'Np'
-      expect(key.child.text).to eq 'HLV'
+      tagger.tag
+      expect(result).to be_a(VnTagger::Document)
+      expect(word.position).to eq 'Np'
+      expect(word.text).to eq 'HLV'
     end
   end
 end

data/spec/vn_tagger/word_spec.rb

@@ -0,0 +1,18 @@
+require 'spec_helper'
+
+describe VnTagger::Word do
+  describe '#is_tag?' do
+    let(:word) { described_class.new('N', 'text') }
+    subject { word.is_tag?(tag) }
+
+    context 'when tag is the same with word position' do
+      let(:tag) { 'Common noun' }
+      it { is_expected.to eq true }
+    end
+
+    context 'when tag is not the same with word position' do
+      let(:tag) { 'Verb' }
+      it { is_expected.to eq false }
+    end
+  end
+end

metadata

@@ -1,14 +1,14 @@
 --- !ruby/object:Gem::Specification
 name: vn_tagger
 version: !ruby/object:Gem::Version
-  version: 0.1.0
+  version: 0.2.0
 platform: ruby
 authors:
 - Hieu Nguyen
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2014-12-22 00:00:00.000000000 Z
+date: 2014-12-30 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: bundler
@@ -86,8 +86,10 @@ files:
 - lib/vn.hus.nlp.tokenizer-4.1.1.jar
 - lib/vn.hus.nlp.utils-1.0.0.jar
 - lib/vn_tagger.rb
+- lib/vn_tagger/document.rb
 - lib/vn_tagger/tagger.rb
 - lib/vn_tagger/version.rb
+- lib/vn_tagger/word.rb
 - resources/automata/dfaLexicon.xml
 - resources/automata/externalLexicon.xml
 - resources/bigram/bigram.xml
@@ -98,7 +100,9 @@ files:
 - resources/normalization/rules.txt
 - resources/prefix/namedEntityPrefix.xml
 - spec/spec_helper.rb
+- spec/vn_tagger/document_spec.rb
 - spec/vn_tagger/tagger_spec.rb
+- spec/vn_tagger/word_spec.rb
 - spec/vn_tagger_spec.rb
 - vn.hus.nlp.tagger-4.2.0.jar
 - vnTagger.sh
@@ -130,5 +134,7 @@ specification_version: 4
 summary: This is a wrapper for vn_tagger library, a A POS tagger for Vietnamese texts.
 test_files:
 - spec/spec_helper.rb
+- spec/vn_tagger/document_spec.rb
 - spec/vn_tagger/tagger_spec.rb
+- spec/vn_tagger/word_spec.rb
 - spec/vn_tagger_spec.rb