treat 1.0.6 → 1.1.0

data/lib/treat/languages/russian.rb

@@ -1,16 +0,0 @@
-class Treat::Languages::Russian
-  
-  RequiredDependencies = []
-  OptionalDependencies = []
-
-  Extractors = {}
-  Inflectors = {}
-  Lexicalizers = {}
-  Processors = {
-    :chunkers => [:txt],
-    :segmenters => [:punkt],
-    :tokenizers => [:tactful]
-  }
-  Retrievers = {}
-  
-end

data/lib/treat/languages/spanish.rb

@@ -1,16 +0,0 @@
-class Treat::Languages::Spanish
-  
-  RequiredDependencies = []
-  OptionalDependencies = []
-  
-  Extractors = {}
-  Inflectors = {}
-  Lexicalizers = {}
-  Processors = {
-    :chunkers => [:txt],
-    :segmenters => [:tactful],
-    :tokenizers => [:tactful]
-  }
-  Retrievers = {}
-  
-end

data/lib/treat/languages/swedish.rb

@@ -1,16 +0,0 @@
-class Treat::Languages::Swedish
-  
-  RequiredDependencies = []
-  OptionalDependencies = []
-  
-  Extractors = {}
-  Inflectors = {}
-  Lexicalizers = {}
-  Processors = {
-    :chunkers => [:txt],
-    :segmenters => [:punkt],
-    :tokenizers => [:tactful]
-  }
-  Retrievers = {}
-  
-end

data/lib/treat/languages.rb

@@ -1,132 +0,0 @@
-# This module provides linguistic resources
-# for the Treat library, including information
-# about language codes, the functions available
-# for each language, and the different tags used
-# to markup that language.
-module Treat::Languages
-
-  def self.const_missing(const)
-    lang = const.to_s.downcase
-    f = File.join(File.dirname(__FILE__), "languages", lang)
-    unless File.readable?(f + '.rb')
-      raise Treat::Exception,
-      "Language #{lang} is not supported."
-    end
-    require f
-    const_get(const)
-  end
-
-  # Yield a lowercase symbol for each
-  # defined language.
-  def self.each
-    constants.each do |constant|
-      yield constant.to_s.downcase.intern
-    end
-  end
-
-  # Identifier constants for language codes.
-  ISO639_1 = 1
-  ISO639_2 = 2
-
-  # Describe a language code (ISO-639-1 or ISO-639-2)
-  # or its full text description in full French or English.
-  def self.describe(lang, desc_lang = :en)
-    raise "Must provide a non-nil language "+
-    "identifier to describe." if lang.nil?
-    lang = code(lang).to_s
-    if [:en, :eng, :english, :anglais].
-      include?(desc_lang)
-      l = @@english_full.key(lang)
-    elsif [:fr, :fra, :french, :french].
-      include?(desc_lang)
-      l = @@french_full.key(lang)
-    else
-      raise Treat::Exception,
-      "Unknown language to describe: #{desc_lang}."
-    end
-    not_found(lang) if l.nil?
-    l.intern
-  end
-
-  # Raise an error message when a language code
-  # or description is not found and suggest
-  # possible misspellings.
-  def self.not_found(lang)
-    msg = "Language '#{lang}' does not exist."
-    all = @@iso639_2.keys + @@iso639_1.keys +
-    @@english_full.keys + @@french_full.keys
-    msg += did_you_mean?(all, lang)
-    raise Treat::Exception, msg
-  end
-
-  # Return the class representing a language.
-  def self.get(lang)
-    lang = Treat::Languages.describe(lang).to_s
-    begin
-      const_get(lang.capitalize)
-    rescue
-      not_found(lang)
-    end
-  end
-
-  # Find a language by ISO-639-1 or ISO-639-2 code
-  # or full name (in English or French) and return
-  # the ISO-639-1 or ISO-639-2 language code as a
-  # lowercase identifier.
-  def self.code(lang, rc = ISO639_2)
-    raise "Must provide a non-nil language "+
-    "identifier to describe." if lang.nil?
-    get_languages
-    lang = lang.to_s.downcase
-    if @@iso639_1.has_key?(lang)
-      return lang.intern if rc == ISO639_2
-      return @@iso639_1[lang].intern if rc == ISO639_1
-    elsif @@iso639_2.has_key?(lang)
-      return lang.intern if rc == ISO639_2
-      return @@iso639_2[lang].intern if rc == ISO639_1
-    elsif @@english_full.has_key?(lang)
-      return @@english_full[lang].intern if rc == ISO639_2
-      return @@iso639_2[@@english_full[lang]].intern if rc == ISO639_1
-    elsif @@french_full.has_key?(lang)
-      return @@french_full[lang].intern if rc == ISO639_2
-      return @@iso639_2[@@french_full[lang]].intern if rc == ISO639_1
-    else
-      not_found(lang)
-    end
-
-  end
-
-  # Whether the language list has been loaded or not.
-  @@loaded = false
-
-  # Get the languages from the dictionary.
-  def self.get_languages
-    return if @@loaded
-    @@iso639_1 = {}; @@iso639_2 = {};
-    @@english_full = {}; @@french_full = {}
-    languages = IO.readlines(File.join(
-    File.dirname(__FILE__), "languages", "list.txt"))
-    languages.each do |language|
-      iso639_2, iso639_1, english_desc, french_desc =
-      language.split(',')
-      @@iso639_1[iso639_1] = iso639_2
-      @@iso639_2[iso639_2] = iso639_1
-      unless english_desc.nil?
-        english_desc.strip.downcase.split('|').each do |l|
-          @@english_full[l.downcase.strip] = iso639_2
-        end
-      end
-      unless french_desc.nil?
-        french_desc.strip.downcase.split('|').each do |l|
-          @@french_full[l.downcase.strip] = iso639_2
-        end
-      end
-    end
-    @@loaded = true
-  end
-
-  # Get the language list.
-  get_languages
-
-
-end

data/lib/treat/lexicalizers.rb

@@ -1,37 +0,0 @@
-# Lexicalizers allow to retrieve lexical information
-# (part of speech tag, general word category, synsets,
-# synonyms, antonyms, hyponyms, hypernyms, lexical
-# relations, grammatical links).
-# of an entity.
-module Treat::Lexicalizers
-
-  # Taggers return the part of speech tag of a word.
-  module Taggers
-    extend Treat::Groupable
-    self.type = :annotator
-    self.targets = [:sentence, :phrase, :token]
-  end
-
-  # Return the general category of a word.
-  module Categorizers
-    extend Treat::Groupable
-    self.type = :annotator
-    self.targets = [:sentence, :phrase, :token]
-    self.recursive = true
-    self.default = :from_tag
-  end
-
-  # Find the synsets of a word in a lexicon.
-  module Sensers
-    extend Treat::Groupable
-    self.type = :annotator
-    self.targets = [:word]
-    self.preset_option = :nym
-    self.presets = [:synonyms, :antonyms, 
-                    :hyponyms, :hypernyms]
-  end
-  
-  # Make Lexicalizers categorizable.
-  extend Treat::Categorizable
-
-end

data/lib/treat/object.rb

@@ -1,7 +0,0 @@
-# Make undefining constants publicly available on any object.
-Object.module_eval do
-  # Unset a constant without private access.
-  def self.const_unset(const)
-    Object.instance_eval { remove_const(const) }
-  end
-end

data/lib/treat/processors/chunkers/autoselect.rb

@@ -1,16 +0,0 @@
-class Treat::Processors::Chunkers::Autoselect
-
-  def self.chunk(entity, options = {})
-    entity.check_has(:format)
-    begin
-      k = Treat::Processors::
-      Chunkers.const_get(cc(entity.format))
-      k.chunk(entity, options)
-    rescue Treat::Exception
-      Treat::Processors::
-      Chunkers::TXT.chunk(entity, options)
-    end
-    
-  end
-
-end

data/lib/treat/processors/chunkers/txt.rb

@@ -1,21 +0,0 @@
-class Treat::Processors::Chunkers::TXT
-  
-  # Separates a string into
-  # zones on the basis of newlines.
-  #
-  # Options: none.
-  def self.chunk(entity, options = {})
-    
-    entity.check_hasnt_children
-    zones = entity.to_s.split("\n")
-    
-    zones.each do |zone|
-      zone.strip!
-      next if zone == ''
-      entity << Treat::Entities::
-      Zone.from_string(zone)
-    end
-    
-  end
-  
-end

data/lib/treat/processors.rb

@@ -1,38 +0,0 @@
-# Processors build trees representing textual entities.
-module Treat::Processors
-  
-  # Chunkers split a document into sections and zones.
-  module Chunkers
-    extend Treat::Groupable
-    self.type = :transformer
-    self.targets = [:document]
-    self.default = :autoselect
-  end
-  
-  # Segmenters split a document or zone into sentences.
-  module Segmenters
-    extend Treat::Groupable
-    self.type = :transformer
-    self.targets = [:zone]
-  end
-  
-  # Tokenizers splits a sentence into Token objects.
-  module Tokenizers
-    extend Treat::Groupable
-    self.type = :transformer
-    self.targets = [:phrase]
-  end
-  
-  # Parsers split a sentence into phrase objects
-  # representing its syntactic structure, with the
-  # Token objects as children of the phrases.
-  module Parsers
-    extend Treat::Groupable
-    self.type = :transformer
-    self.targets = [:phrase]
-  end
-  
-  # Make Processors categorizable.
-  extend Treat::Categorizable
-  
-end

data/lib/treat/retrievers.rb

@@ -1,27 +0,0 @@
-# Retrievers find documents in collections.
-module Treat::Retrievers
-  
-  # Indexers create an index of words used
-  # in the documents within a collection.
-  module Indexers
-    extend Treat::Groupable
-    self.type = :annotator
-    self.targets = [:collection]
-    self.default = :ferret
-  end
-  
-  # Searchers perform full-text search
-  # on indexed collections in order
-  # to retrieve documents matching
-  # a query.
-  module Searchers
-    extend Treat::Groupable
-    self.type = :computer
-    self.targets = [:collection]
-    self.default = :ferret
-  end
-  
-  # Make Retrievers categorizable.
-  extend Treat::Categorizable
-  
-end

data/lib/treat/server.rb

@@ -1,26 +0,0 @@
-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/universalisation/encodings.rb

@@ -1,12 +0,0 @@
-module Treat::Universalisation
-  
-  Encodings = {
-    :arabic => 'UTF-8',
-    :chinese => 'GB18030',
-    :english => 'UTF-8',
-    :french => 'ISO_8859-1',
-    :german => 'ISO_8859-1',
-    :hebrew => 'UTF-8'
-  }
-  
-end