treat 1.0.2 → 1.0.3

data/lib/treat/classification.rb

@@ -60,4 +60,4 @@ class Treat::Classification
     line
   end
 
-end
+end

data/lib/treat/dependencies.rb

@@ -1,12 +1,12 @@
 class Treat::Dependencies
   
   Gem = [
-    ['psych', '1.2.2', '(un)serialize annotated entities to YAML format'],
-    ['nokogiri', '>= 1.4.0', 'read and parse XML and HTML formats'],
+    ['psych', '>= 1.2.2', '(un)serialize annotated entities to YAML format'],
+    ['nokogiri', '>= 1.5.2', 'read and parse XML and HTML formats'],
     ['sdsykes-ferret', '>= 0.11.6.19', 'perform full-text search in collections'],
     ['lda-ruby', '>= 0.3.8', 'extract topic words from documents and collections'],
-    ['ruby-readability', '>= 0.5.0', 'extract the readable content from HTML pages'],
-    ['stanford-core-nlp', '>= 0.1.8', 'tokenize, segment, parse texts and perform named entity recognition'],
+    ['ruby-readability', '>= 0.5.2', 'extract the readable content from HTML pages'],
+    ['stanford-core-nlp', '>= 0.3.0', 'tokenize, segment, parse texts and perform named entity recognition'],
     ['whatlanguage', '>= 1.0.0', 'detect the language of text'],
     ['linguistics', '>= 1.0.9', 'retrieve the inflection of nouns, verbs and numbers in English'],
     ['punkt-segmenter', '>= 0.9.1', 'segment texts into sentences'],

data/lib/treat/entities/abilities/buildable.rb

@@ -13,7 +13,8 @@ module Treat::Entities::Abilities::Buildable
   PunctRegexp = /^[[:punct:]\$]+$/
   UriRegexp = /^(http|https):\/\/[a-z0-9]+([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}(([0-9]{1,5})?\/.*)?$/ix
   EmailRegexp = /.+\@.+\..+/
-
+  Enclitics = %w['ll 'm 're 's 't 've]
+  
   # Reserved folder names
   Reserved = ['.index']
 
@@ -264,8 +265,8 @@ module Treat::Entities::Abilities::Buildable
   def token_from_string(string)
 
     check_encoding(string)
-    if string == "'s" || string == "'S"
-      Treat::Entities::Clitic.new(string)
+    if Enclitics.include?(string.downcase)
+      Treat::Entities::Enclitic.new(string)
     elsif string =~ WordRegexp &&
       string.count(' ') == 0 &&
       string != '-'

data/lib/treat/entities/abilities/iterable.rb

@@ -30,7 +30,7 @@ module Treat::Entities::Abilities::Iterable
     a = []
     type = :entity unless type
     each_entity(type) do |e|
-      r = e.send(feature)
+      r = e.get(feature)
       a << e if r == value
     end
     a

data/lib/treat/entities/abilities/stringable.rb

@@ -69,7 +69,7 @@ module Treat::Entities::Abilities::Stringable
       
       if child.is_a?(Treat::Entities::Token) || child.value != ''
         if child.is_a?(Treat::Entities::Punctuation) ||
-          child.is_a?(Treat::Entities::Clitic)
+          child.is_a?(Treat::Entities::Enclitic)
           value.strip!
         end
         value += child.to_s + ' '

data/lib/treat/entities/entities.rb

@@ -76,7 +76,7 @@ module Treat::Entities
   end
 
   # Represents a clitic ('s).
-  class Clitic < Token; end
+  class Enclitic < Token; end
 
   # Represents a number.
   class Number < Token

data/lib/treat/extractors/name_tag/stanford.rb

@@ -4,7 +4,8 @@
 class Treat::Extractors::NameTag::Stanford
 
   require 'treat/loaders/stanford'
-
+  Treat::Loaders::Stanford.load
+  
   @@classifiers = {}
 
   def self.name_tag(entity, options = {})
@@ -12,9 +13,10 @@ class Treat::Extractors::NameTag::Stanford
     pp = nil
 
     lang = entity.language
-
+    
     language = Treat::Languages.describe(lang)
-
+    Treat::Loaders::Stanford.load(language)
+    
     isolated_token = entity.is_a?(Treat::Entities::Token)
     tokens = isolated_token ? [entity] : entity.tokens
 

data/lib/treat/formatters/readers/xml.rb

@@ -1,6 +1,7 @@
 class Treat::Formatters::Readers::XML
 
   require 'treat/loaders/stanford'
+  Treat::Loaders::Stanford.load
   require 'cgi'
 
   # By default, don't backup the XML

data/lib/treat/formatters/serializers/mongo.rb

@@ -32,7 +32,7 @@ class Treat::Formatters::Serializers::Mongo
       coll = @@db[type_id[0]][type_id[1]]
     end
     
-    # Store path
+    # Store path 
     
     Treat::Entities.list.each do |type|
 

data/lib/treat/lexicalizers/taggers/stanford.rb

@@ -22,6 +22,7 @@ class Treat::Lexicalizers::Taggers::Stanford
     
     # Handle options and initialize the tagger.
     lang = entity.language
+    
     options = get_options(options, lang)
     tokens, list = get_token_list(entity)
     init_tagger(lang)
@@ -55,14 +56,13 @@ class Treat::Lexicalizers::Taggers::Stanford
   
   # Initialize the tagger for a language.
   def self.init_tagger(lang)
-    
     language = Treat::Languages.describe(lang)
+    Treat::Loaders::Stanford.load(language)
     model = StanfordCoreNLP::Config::Models[:pos][language]
-    model = Treat.models + 'stanford/' +
+    model = Treat::Loaders::Stanford.model_path +
     StanfordCoreNLP::Config::ModelFolders[:pos] + model
     @@taggers[lang] ||= 
     StanfordCoreNLP::MaxentTagger.new(model)
-    
   end
   
   # Handle the options for the tagger.

data/lib/treat/loaders/linguistics.rb

@@ -1,26 +1,30 @@
-class Treat::Loaders
+module Treat
 
-  # A helper class to load a language class
-  # registered with the Linguistics gem.
-  class Linguistics
+  module Loaders
 
-    silence_warnings { require 'linguistics' }
-    @@languages = {}
+    # A helper class to load a language class
+    # registered with the Linguistics gem.
+    class Linguistics
 
-    def self.load(language)
-      if @@languages[language]
-        return @@languages[language]
-      end
-      begin
-        l = language.to_s.upcase
-        silence_warnings do
-          @@languages[language] =
-          ::Linguistics.const_get(l)
+      silence_warnings { require 'linguistics' }
+      @@languages = {}
+
+      def self.load(language)
+        if @@languages[language]
+          return @@languages[language]
         end
-      rescue RuntimeError
-        raise "Ruby Linguistics does " +
-        "not have a module installed " +
-        "for the #{language} language."
+        begin
+          l = language.to_s.upcase
+          silence_warnings do
+            @@languages[language] =
+            ::Linguistics.const_get(l)
+          end
+        rescue RuntimeError
+          raise "Ruby Linguistics does " +
+          "not have a module installed " +
+          "for the #{language} language."
+        end
+
       end
 
     end

data/lib/treat/loaders/stanford.rb

@@ -1,27 +1,45 @@
-class Treat::Loaders
+module Treat
+  
+  module Loaders
 
-  # A helper class to load a language class
-  # registered with the Linguistics gem.
-  class Stanford
+    class Stanford
 
-    require 'stanford-core-nlp'
-    
-    StanfordCoreNLP.jar_path = 
-    Treat.bin + 'stanford/'
-    
-    StanfordCoreNLP.model_path = 
-    Treat.models + 'stanford/'
-    
-    StanfordCoreNLP.use(
-    Treat::Languages.describe(
-    Treat.default_language))
+      require 'stanford-core-nlp'
 
-    StanfordCoreNLP.log_file = 
-    NULL_DEVICE if Treat.silence
-    
-    StanfordCoreNLP.bind
-    @@loaded = true
+      class << self
+        attr_accessor :jar_path
+        attr_accessor :model_path
+        attr_accessor :loaded
+      end
+
+      self.jar_path = Treat.bin + 'stanford/'
+      self.model_path = Treat.models + 'stanford/'
+      self.loaded = false
+
+      def self.load(language = nil)
+
+        return if self.loaded
+
+        language ||=
+        Treat::Languages.describe(
+        Treat.default_language)
+
+        StanfordCoreNLP.jar_path = self.jar_path
+        StanfordCoreNLP.model_path = self.model_path
+
+        StanfordCoreNLP.use(language)
+
+        StanfordCoreNLP.log_file =
+        NULL_DEVICE if Treat.silence
+
+        StanfordCoreNLP.bind
+
+        self.loaded = true
+
+      end
+
+    end
     
   end
-
-end
+  
+end

data/lib/treat/processors/parsers/stanford.rb

@@ -26,7 +26,7 @@ class Treat::Processors::Parsers::Stanford
     val = entity.to_s
     lang = entity.language
     init(lang, options)
-
+    
     text = ::StanfordCoreNLP::Text.new(val)
     @@parsers[lang].annotate(text)
 
@@ -52,6 +52,10 @@ class Treat::Processors::Parsers::Stanford
 
   def self.init(lang, options)
     return if @@parsers[lang]
+    
+    language = Treat::Languages.describe(lang)
+    Treat::Loaders::Stanford.load(language)
+    
     options = DefaultOptions.merge(options)
     StanfordCoreNLP.use(lang)
     if options[:tagger_model]

data/lib/treat/processors/segmenters/stanford.rb

@@ -3,7 +3,8 @@
 class Treat::Processors::Segmenters::Stanford
 
   require 'treat/loaders/stanford'
-
+  Treat::Loaders::Stanford.load
+  
   DefaultOptions = {
     :also_tokenize => false
   }

data/lib/treat/processors/tokenizers/stanford.rb

@@ -3,7 +3,8 @@
 class Treat::Processors::Tokenizers::Stanford
 
   require 'treat/loaders/stanford'
-
+  Treat::Loaders::Stanford.load
+  
   @@tokenizer = nil
 
   # Tokenize the entity using a Penn-Treebank

data/lib/treat/server.rb

@@ -0,0 +1,26 @@
+class Treat::Server
+  
+  require 'thin'
+
+  def self.start
+    app = proc do |env|
+      #!/usr/bin/env ruby -w
+      # simple_service.rb
+      # A simple DRb service
+
+      # load DRb
+      require 'drb'
+
+      # start up the DRb service
+      DRb.start_service nil, []
+
+      # We need the uri of the service to connect a client
+      puts DRb.uri
+
+      # wait for the DRb service to finish before exiting
+      DRb.thread.join
+    end
+    run app
+  end
+  
+end

data/lib/treat/tree.rb

@@ -175,6 +175,8 @@ module Treat::Tree
 
     # Return a feature.
     def get(feature)
+      return @value if feature == :value
+      return @id if feature == :id
       @features[feature]
     end
 

data/lib/treat.rb

@@ -10,7 +10,7 @@ module Treat
   end
 
   # The current version of Treat.
-  VERSION = "1.0.2"
+  VERSION = "1.0.3"
 
   # Add methods to handle syntactic sugar,
   # language configuration options, and paths.
@@ -49,7 +49,7 @@ module Treat
   require 'treat/categories'
   require 'treat/data_set'
   require 'treat/proxies'
-
+  
   # Install packages for a given language.
   def self.install(language = :english)
     require 'treat/installer'

data/spec/sandbox.rb

@@ -1,24 +0,0 @@
-require_relative '../lib/treat'
-
-s = Sentence "Barack Obama was killed last night."
-s.tokenize
-
-puts s.word_with_position(2).inspect
-
-s.word_with_position(2).set :highlighted, 1
-
-cl = Treat::Classification.new(
-  :word,
-  [[:position, 0]],
-  :highlighted,
-  0
-)
-
-data_set = s.export(cl)
-
-s2 = Sentence ''
-w = Word 'Hello'
-s2 << w
-w.set :position, 2
-
-puts w.classify(:mlp, :training => data_set).inspect

metadata

@@ -1,7 +1,7 @@
 --- !ruby/object:Gem::Specification
 name: treat
 version: !ruby/object:Gem::Version
-  version: 1.0.2
+  version: 1.0.3
   prerelease: 
 platform: ruby
 authors:
@@ -9,7 +9,7 @@ authors:
 autorequire: 
 bindir: bin
 cert_chain: []
-date: 2012-04-14 00:00:00.000000000 Z
+date: 2012-04-20 00:00:00.000000000 Z
 dependencies:
 - !ruby/object:Gem::Dependency
   name: rubyzip
@@ -200,6 +200,7 @@ files:
 - lib/treat/retrievers/indexers/ferret.rb
 - lib/treat/retrievers/searchers/ferret.rb
 - lib/treat/retrievers.rb
+- lib/treat/server.rb
 - lib/treat/tree.rb
 - lib/treat.rb
 - spec/collection.rb
@@ -224,7 +225,9 @@ files:
 - LICENSE
 homepage: https://github.com/louismullie/treat
 licenses: []
-post_install_message: 
+post_install_message: ! "********************************************************************************\n\nThank
+  you for installing Treat!\n\nComplete the installation by running:\n\n    require
+  'treat'\n    Treat.install\n\ninside IRB or a Ruby script.\n\n********************************************************************************\n\n"
 rdoc_options: []
 require_paths:
 - lib