treat 1.2.0 → 2.0.0rc1

data/lib/treat/entities/collection.rb

@@ -1,40 +0,0 @@
-module Treat::Entities
-  # Represents a collection of texts.
-  class Collection < Entity
-
-    # Initialize the collection with a folder
-    # containing the texts of the collection.
-    def initialize(folder = nil, id = nil)
-      super('', id)
-      if folder
-        if !FileTest.directory?(folder)
-          FileUtils.mkdir(folder) 
-        end
-        set :folder, folder if folder
-        i = folder + '/.index'
-        if FileTest.directory?(i)
-          set :index, i
-        end
-      end
-    end
-
-    # Works like the default <<, but if the
-    # file being added is a collection or a
-    # document, then copy that collection or
-    # document into this collection's folder.
-    def <<(entities, copy = true)
-      unless entities.is_a?(Array)
-        entities = [entities]
-      end
-      entities.each do |entity|
-        if [:document, :collection].
-          include?(entity.type) && copy &&
-          @features[:folder] != nil
-          entity = entity.copy_into(self)
-        end
-      end
-      super(entities)
-    end
-    
-  end
-end

data/lib/treat/entities/document.rb

@@ -1,10 +0,0 @@
-module Treat::Entities
-  # Represents a document.
-  class Document < Entity
-    # Initialize a document with a file name.
-    def initialize(file = nil, id = nil)
-      super('', id)
-      set :file, file
-    end
-  end
-end

data/lib/treat/entities/group.rb

@@ -1,18 +0,0 @@
-module Treat::Entities
-  
-  # Represents a group of tokens.
-  class Group < Entity; end
-
-  # Represents a group of words 
-  # with a sentence ender (.!?)
-  class Sentence < Group; end
-
-  # Represents a group of words,
-  # with no sentence ender.
-  class Phrase < Group; end
-
-  # Represents a non-linguistic
-  # fragment (e.g. stray symbols).
-  class Fragment < Group; end
-
-end

data/lib/treat/entities/section.rb

@@ -1,13 +0,0 @@
-module Treat::Entities
-  # Represents a section.
-  class Section < Entity; end
-
-  # Represents a page of text.
-  class Page < Section; end
-
-  # Represents a block of text 
-  class Block < Section; end
-
-  # Represents a list.
-  class List < Section; end
-end

data/lib/treat/entities/token.rb

@@ -1,47 +0,0 @@
-module Treat::Entities
-  
-  # Represents a terminal element 
-  # (leaf) in the text structure.
-  class Token < Entity; end
-
-  # Represents a word.  Strictly,
-  # this is /^[[:alpha:]\-']+$/.
-  class Word < Token; end
-
-  # Represents an enclitic.
-  # Strictly, this is any of 
-  # 'll 'm 're 's 't or 've.
-  class Enclitic < Token; end
-
-  # Represents a number. Strictly,
-  # this is /^#?([0-9]+)(\.[0-9]+)?$/.
-  class Number < Token
-    def to_i; to_s.to_i; end
-    def to_f; to_s.to_f; end
-  end
-
-  # Represents a punctuation sign.
-  # Strictly, this is /^[[:punct:]\$]+$/.
-  class Punctuation < Token; end
-
-  # Represents a character that is neither
-  # a word, an enclitic, a number or a
-  # punctuation character (e.g. @#$%&*).
-  class Symbol < Token; end
-
-  # Represents a url. This is (imperfectly)
-  # defined as /^(http|https):\/\/[a-z0-9] 
-  # +([\-\.]{1}[a-z0-9]+)*\.[a-z]{2,5}
-  # (([0-9]{1,5})?\/.*)?$/ix
-  class Url < Token; end
-
-  # Represents a valid RFC822 address.
-  # This is (imperfectly) defined as 
-  # /.+\@.+\..+/ (fixme maybe?)
-  class Email < Token; end
-  
-  # Represents a token whose type
-  # cannot be identified.
-  class Unknown; end
-  
-end

data/lib/treat/entities/zone.rb

@@ -1,12 +0,0 @@
-module Treat::Entities
-  # Represents a zone of text.
-  class Zone < Entity; end
-
-  # Represents a title, subtitle, 
-  # logical header of a text.
-  class Title < Zone; end
-
-  # Represents a paragraph (group 
-  # of sentences and/or phrases).
-  class Paragraph < Zone; end
-end

data/lib/treat/entities.rb

@@ -1,6 +0,0 @@
-# Contains the textual model used by Treat.
-module Treat::Entities
-  require 'treat/entities/entity'
-  p = Treat.paths.lib + 'treat/entities/*.rb'
-  Dir.glob(p).each { |f| require f }
-end

data/lib/treat/helpers/didyoumean.rb

@@ -1,57 +0,0 @@
-# Search the list to see if there are 
-# words similar to #name in the #list 
-# If yes, return a string saying 
-# "Did you mean ... ?" with the names.
-def did_you_mean?(list, name)
-  return '' # Fix
-  list = list.map { |e| e.to_s }
-  name = name.to_s
-  sugg = []
-  list.each do |element|
-    l = levenshtein(element,name)
-    if  l > 0 && l < 2
-      sugg << element
-    end
-  end
-  unless sugg.size == 0
-    if sugg.size == 1
-      msg += " Perhaps you meant '#{sugg[0]}' ?"
-    else
-      sugg_quote = sugg[0..-2].map do
-        |x| '\'' + x + '\''
-      end
-      msg += " Perhaps you meant " +
-      "#{sugg_quote.join(', ')}," +
-      " or '#{sugg[-1]}' ?"
-    end
-  end
-  msg
-end
-
-alias :dym? :did_you_mean?
-
-# Return the levensthein distance between 
-# two strings taking into account the costs 
-# of insertion, deletion, and substitution. 
-# Used by did_you_mean? to detect typos.
-def levenshtein(first, other, ins=1, del=1, sub=1)
-  return nil if first.nil? || other.nil?
-  dm = []
-  dm[0] = (0..first.length).collect { |i| i * ins}
-  fill = [0] * (first.length - 1).abs
-  for i in 1..other.length
-    dm[i] = [i * del, fill.flatten]
-  end
-  for i in 1..other.length
-    for j in 1..first.length
-      dm[i][j] = [
-        dm[i-1][j-1] +
-        (first[i-1] ==
-        other[i-1] ? 0 : sub),
-        dm[i][j-1] + ins,
-        dm[i-1][j] + del
-      ].min
-    end
-  end
-  dm[other.length][first.length]
-end

data/lib/treat/helpers/escaping.rb

@@ -1,15 +0,0 @@
-# This is ugly, we should change it.
-EscapeChar = '^^'
-EscapedEscapeChar = '\^\^'
- 
-def escape_floats!(s)
-  s.gsub!(/([0-9]+)\.([0-9]+)/) do
-    $1 + EscapeChar + $2
-  end
-end
-
-def unescape_floats!(s)
-  s.gsub!(/([0-9]+)#{EscapedEscapeChar}([0-9]+)/) do
-    $1 + '.' + $2
-  end
-end

data/lib/treat/helpers/formatting.rb

@@ -1,41 +0,0 @@
-# A cache to optimize camel casing.
-@@cc_cache = {}
-
-# A cache to optimize un camel casing.
-@@ucc_cache = {}
-
-# Convert un_camel_case to CamelCase.
-def camel_case(o_phrase)
-  phrase = o_phrase.to_s.dup
-  return @@cc_cache[o_phrase] if @@cc_cache[o_phrase]
-
-  if Treat.core.acronyms.include?(phrase)
-    phrase = phrase.upcase
-  else
-    phrase.gsub!(/^[a-z]|_[a-z]/) { |a| a.upcase }
-    phrase.gsub!('_', '')
-  end
-  @@cc_cache[o_phrase] = phrase
-end
-
-alias :cc :camel_case
-
-# Convert CamelCase to un_camel_case.
-def un_camel_case(o_phrase)
-  phrase = o_phrase.to_s.dup
-  return @@ucc_cache[o_phrase] if @@ucc_cache[o_phrase]
-  if Treat.core.acronyms.include?(phrase.downcase)
-    phrase = phrase.downcase
-  else
-    phrase.gsub!(/[A-Z]/) { |p| '_' + p.downcase  }
-    phrase = phrase[1..-1] if phrase[0] == '_'
-  end
-  @@ucc_cache[o_phrase] = phrase
-end
-
-alias :ucc :un_camel_case
-
-# Retrieve the Class from a Module::Class.
-def class_name(n); n.to_s.split('::')[-1]; end
-
-alias :cl :class_name

data/lib/treat/helpers/objtohash.rb

@@ -1,8 +0,0 @@
-def object_to_hash(obj)
-  hash = {}
-  obj.instance_variables.each do |var|
-    val = obj.instance_variable_get(var)
-    hash[var.to_s.delete("@")] = val
-  end
-  hash
-end

data/lib/treat/helpers/platform.rb

@@ -1,15 +0,0 @@
-# Detect the platform we're running on.
-def detect_platform
-  p = RUBY_PLATFORM.downcase
-  return :mac if p.include?("darwin")
-  return :windows if p.include?("mswin")
-  return :linux if p.include?("linux")
-  return :unknown
-end
-
-# Set up the right NULL device.
-if detect_platform == :windows
-  NULL_DEVICE = 'NUL'
-else
-  NULL_DEVICE = '/dev/null'
-end

data/lib/treat/helpers/reflection.rb

@@ -1,17 +0,0 @@
-# Return the name of the method that 
-# called the method that calls this method.
-def caller_method(n = 3)
-  at = caller(n).first
-  /^(.+?):(\d+)(?::in `(.*)')?/ =~ at
-  Regexp.last_match[3].
-  gsub('block in ', '').intern
-end
-
-Object.module_eval do
-  # Unset a constant publicly.
-  def self.const_unset(const)
-    Object.instance_eval do 
-      remove_const(const)
-    end
-  end
-end

data/lib/treat/helpers/temporary.rb

@@ -1,27 +0,0 @@
-# Require file utilities for creating and
-# deleting temporary files.
-require 'fileutils'
-
-# Create a temporary file which is deleted
-# after execution of the block.
-def create_temp_file(ext, value = nil, &block)
-  fname = Treat.paths.tmp + 
-  "#{Random.rand(10000000).to_s}.#{ext}"
-  File.open(fname, 'w') do |f|
-    f.write(value) if value
-    block.call(f.path)
-  end
-ensure
-  File.delete(fname)
-end
-
-# Create a temporary directory, which is
-# deleted after execution of the block.
-def create_temp_dir(&block)
-  dname = Treat.paths.tmp +
-  "#{Random.rand(10000000).to_s}"
-  Dir.mkdir(dname)
-  block.call(dname)
-ensure
-  FileUtils.rm_rf(dname)
-end

data/lib/treat/helpers/verbosity.rb

@@ -1,19 +0,0 @@
-# Runs a block of code without warnings.
-def silence_warnings(&block)
-  warn_level = $VERBOSE
-  $VERBOSE = nil
-  result = block.call
-  $VERBOSE = warn_level
-  result
-end
-
-# Runs a block of code while blocking stdout.
-def silence_stdout(log = '/dev/null')
-  unless Treat.core.verbosity.silence
-    yield; return
-  end
-  old = $stdout.dup
-  $stdout.reopen(File.new(log, 'w'))
-  yield
-  $stdout = old
-end

data/lib/treat/helpers.rb

@@ -1,5 +0,0 @@
-# Contains utility functions used by Treat.
-module Treat::Helpers
-  p = Treat.paths.lib + 'treat/helpers/*.rb'
-  Dir.glob(p).each { |f| require f }
-end

data/lib/treat/loaders.rb

@@ -1,10 +0,0 @@
-# Contains classes to load external libraries.
-module Treat::Loaders
-  # Autoload all of the loaders.
-  def self.const_missing(const)
-    name = const.to_s.downcase
-    require Treat.paths.lib + 
-    "treat/loaders/#{name}.rb"
-    self.const_get(const)
-  end
-end

data/lib/treat/proxies.rb

@@ -1,106 +0,0 @@
-# Proxies install builders on core Ruby objects;
-# when a method defined by Treat is called on these
-# objects, the Ruby object is cast to a Treat entity
-# and the method is called on the resultant type.
-module Treat::Core::Proxies
-
-  # Provides a base functionality for proxies.
-  module Proxy
-
-    # Build the entity corresponding to the proxied
-    # object and send the method call to the entity.
-    def method_missing(sym, *args, &block)
-      if sym == :do || Treat::Workers.lookup(sym)
-        to_entity.send(sym, *args)
-      else
-        super(sym, *args, &block)
-      end
-    end
-
-    # Create an unknown type of entity by default.
-    def to_entity(builder = nil)
-      Treat::Entities::Unknown(self.to_s)
-    end
-
-  end
-
-  # Install Treat functions on String objects.
-  module String
-
-    # Include base proxy functionality.
-    include Treat::Core::Proxies::Proxy
-
-    # Return the entity corresponding to the string.
-    def to_entity
-      Treat::Entities::Entity.from_string(self)
-    end
-
-  end
-
-  # Install Treat functions on Numeric objects.
-  module Numeric
-
-    # Include base proxy functionality.
-    include Treat::Core::Proxies::Proxy
-
-    # Return the entity corresponding to the number.
-    def to_entity(builder = nil)
-      Treat::Entities::Number.from_numeric(self)
-    end
-
-  end
-  
-  # Include Treat methods on strings.
-  ::String.class_eval do
-    include Treat::Core::Proxies::String
-  end
-  
-  # Include Treat methods on numerics.
-  ::Numeric.class_eval do
-    include Treat::Core::Proxies::Numeric
-  end
-  
-  # This is kind of ugly; need to find a
-  # better solution eventually (?)
-  Treat::Entities::Entity.class_eval do
-    
-    # Rename the true language detection
-    # method to :language_proxied, and
-    # only call it if language detection
-    # is turned on in the configuration.
-    alias :language_proxied :language
-    
-    # Proxy the #language method, defined on
-    # all textual entities, in order to catch
-    # the method call if language detection is
-    # turned off and return the default language
-    # in that case.
-    def language(extractor = nil, options = {})
-  
-      return Treat.core.language.default if
-      !Treat.core.language.detect
-
-      if is_a?(Treat::Entities::Symbol) ||
-        is_a?(Treat::Entities::Number)
-        return Treat.core.language.default
-      end
-
-      dlvl = Treat.core.language.detect_at
-      dklass = Treat::Entities.const_get(cc(dlvl))
-      
-      if self.class.compare_with(
-        dklass) < 1 && has_parent?
-        anc = ancestor_with_type(dlvl)
-        return anc.language if anc
-      end
-
-      extractor ||= Treat.workers.
-      extractors.language.default
-
-      language_proxied(extractor, options)
-
-    end
-
-  end
-
-end

data/lib/treat/workers/formatters/unserializers/autoselect.rb

@@ -1,17 +0,0 @@
-class Treat::Workers::Formatters::Unserializers::Autoselect
-
-  def self.unserialize(document, options = {})
-    file = document.file
-    if file.index('yml') || file.index('yaml')
-      document.unserialize(:yaml, options)
-    elsif file.index('xml')
-      document.unserialize(:xml, options)
-    elsif file.index('mongo')
-      document.unserialize(:mongo, options)
-    else
-      raise Treat::Exception,
-      "Unreadable serialized format for file #{file}."
-    end
-  end
-
-end

data/lib/treat/workers/inflectors/declensors/active_support.rb

@@ -1,31 +0,0 @@
-# This class is a wrapper for the ActiveSupport
-# declension tools.
-class Treat::Workers::Inflectors::Declensors::English
-
-  require 'active_support/inflector/inflections'
-
-  # Declense a word using ActiveSupport::Inflector::Inflections
-  def self.declense(entity, options)
-    
-    cat = entity.check_has(:category)
-    unless ['noun', 'adjective', 'determiner'].
-      include?(cat)
-        return
-    end
-    
-    unless options[:count]
-      raise Treat::Exception,
-      "Must supply option count (:singular or :plural)."
-    end
-    
-    string = entity.to_s
-
-    if options[:count] == :plural
-      ActiveSupport::Inflector::Inflections.pluralize(string)
-    elsif options[:count] == :singular
-      ActiveSupport::Inflector::Inflections.singularize(string)
-    end
-    
-  end
-  
-end

data/lib/treat/workers/processors/tokenizers/tactful.rb

@@ -1,68 +0,0 @@
-# A tokenizer class lifted from the 'tactful-tokenizer' gem.
-#
-# Copyright © 2010 Matthew Bunday. All rights reserved.
-# Released under the GNU GPL v3. Modified by Louis Mullie.
-#
-# Project website: https://github.com/SlyShy/Tactful_Tokenizer
-class Treat::Workers::Processors::Tokenizers::Tactful
-
-  ReTokenize = [
-    # Uniform Quotes
-    [/''|``/, '"'],
-    # Separate punctuation from words.
-    [/(^|\s)(')/, '\1\2'],
-    [/(?=[\("`{\[:;&#*@\.])(.)/, '\1 '],
-    [/(.)(?=[?!\)";}\]*:@\.'])|(?=[\)}\]])(.)|(.)(?=[({\[])|((^|\s)-)(?=[^-])/, '\1 '],
-    # Treat double-hyphen as a single token.
-    [/([^-])(--+)([^-])/, '\1 \2 \3'],
-    [/(\s|^)(,)(?=(\S))/, '\1\2 '],
-    # Only separate a comma if a space follows.
-    [/(.)(,)(\s|$)/, '\1 \2\3'],
-    # Combine dots separated by whitespace to be a single token.
-    [/\.\s\.\s\./, '...'],
-    # Separate "No.6"
-    [/([\W]\.)(\d+)/, '\1 \2'],
-    # Separate words from ellipses
-    [/([^\.]|^)(\.{2,})(.?)/, '\1 \2 \3'],
-    [/(^|\s)(\.{2,})([^\.\s])/, '\1\2 \3'],
-    [/(^|\s)(\.{2,})([^\.\s])/, '\1 \2\3'],
-    ##### Some additional fixes.
-    # Fix %, $, &
-    [/(\d)%/, '\1 %'],
-    [/\$(\.?\d)/, '$ \1'],
-    [/(\W)& (\W)/, '\1&\2'],
-    [/(\W\W+)&(\W\W+)/, '\1 & \2'],
-    # Fix (n 't) -> ( n't)
-    [/n 't( |$)/, " n't\\1"],
-    [/N 'T( |$)/, " N'T\\1"],
-    # Treebank tokenizer special words
-    [/([Cc])annot/, '\1an not']
-
-  ]
-
-
-  # Tokenize the entity using a rule-based algorithm
-  # that has been lifted from the 'tactful-tokenizer'
-  # gem.
-  def self.tokenize(entity, options = {})
-    
-    entity.check_hasnt_children
-    
-    s = entity.to_s
-    
-    escape_floats!(s)
-    
-    ReTokenize.each do |rules|
-      s.gsub!(rules[0], rules[1])
-    end
-    
-    s.split(' ').each do |token|
-      
-      unescape_floats!(token)
-      entity << Treat::Entities::Token.
-      from_string(token)
-    end
-    
-  end
-
-end