treat 1.2.0 → 2.0.0rc1

data/lib/treat/config/data/tags.rb

@@ -0,0 +1,328 @@
+{
+  aligned: {
+    tag_sets: [
+      :claws_c5, :brown, :penn,
+      :stutgart, :chinese, :paris7
+    ],
+    phrase_tags: [
+      'Adjectival phrase', ['', '', 'ADJP', '', '', 'AP'],
+      'Adverbial phrase', ['', '', 'ADVP', '', '', 'AdP'],
+      'Conjunction phrase', ['', '', 'CONJP', '', '', 'Ssub'],
+      'Fragment', ['', '', 'FRAG', '', '', ''],
+      'Interjectional phrase', ['', '', 'INTJ', '', '', ''],
+      'List marker', ['', '', 'LST', '', '', ''],
+      'Not a phrase', ['', '', 'NAC', '', '', ''],
+      'Noun phrase', ['', '', 'NP', '', '', 'NP'],
+      'Verbal nucleus',  ['', '', '', '', '', 'VN'],
+      'Head of noun phrase', ['', '', 'NX', '', '', ''],
+      'Prepositional phrase', ['', '', 'PP', '', '', 'PP'],
+      'Parenthetical', ['', '', 'PRN', '', '', ''],
+      'Particle', ['', '', 'PRT', '', '', ''],
+      'Participial phrase', ['', '', '', '', '', 'VPart'],
+      'Quantifier phrase', ['', '', 'QP', '', '', ''],
+      'Relative clause', ['', '', 'RRC', '', '', 'Srel'],
+      'Coordinated phrase', ['', '', 'UCP', '', '', 'COORD'],
+      'Infinitival phrase', ['', '', '', '', '', 'VPinf'],
+      'Verb phrase', ['', '', 'VP', '', '', ''],
+      'Inverted yes/no question', ['', '', 'SQ', '', '', ''],
+      'Wh adjective phrase', ['', '', 'WHADJP', '', '', ''],
+      'Wh adverb phrase', ['', '', 'WHAVP', '', '', ''],
+      'Wh noun phrase', ['', '', 'WHNP', '', '', ''],
+      'Wh prepositional phrase', ['', '', 'WHPP', '', '', ''],
+      'Unknown', ['', '', 'X', '', '', ''],
+      'Phrase', ['', '', 'P', '', '', 'Sint'],
+      'Sentence', ['', '', 'S', '', '', 'SENT'],
+      'Phrase', ['', '', 'SBAR', '', '', ''] # Fix
+    ],
+    word_tags: [
+
+      # Aligned tags for the Claws C5, Brown and Penn tag sets.
+      'Adjective', ['AJ0', 'JJ', 'JJ', '', 'JJ', 'A'],
+      'Adjective', ['AJ0', 'JJ', 'JJ', '', 'JJ', 'ADJ'],
+      'Ajective, adverbial or predicative', ['', '', '', 'ADJD', '', 'ADJ'],
+      'Adjective, attribute', ['', '', '', 'ADJA', 'VA', 'ADJ'],
+      'Adjective, ordinal number', ['ORD', 'OD', 'JJ', '', 'OD', 'ADJ'],
+      'Adjective, comparative', ['AJC', 'JJR', 'JJR', 'KOKOM', '', 'ADJ'],
+      'Adjective, superlative', ['AJS', 'JJT', 'JJS', '', 'JJ', 'ADJ'],
+      'Adjective, superlative, semantically', ['AJ0', 'JJS', 'JJ', '', '', 'ADJ'],
+      'Adjective, cardinal number', ['CRD', 'CD', 'CD', 'CARD', 'CD', 'ADJ'],
+      'Adjective, cardinal number, one', ['PNI', 'CD', 'CD', 'CARD', 'CD', 'ADJ'],
+
+      'Adverb', ['AV0', 'RB', 'RB', 'ADV', 'AD', 'ADV'],
+      'Adverb, negative', ['XX0', '*', 'RB', 'PTKNEG', '', 'ADV'],
+      'Adverb, comparative', ['AV0', 'RBR', 'RBR', '', 'AD', 'ADV'],
+      'Adverb, superlative', ['AV0', 'RBT', 'RBS', '', 'AD', 'ADV'],
+      'Adverb, particle', ['AVP', 'RP', 'RP', '', '', 'ADV'],
+      'Adverb, question', ['AVQ', 'WRB', 'WRB', '', 'AD', 'ADV'],
+      'Adverb, degree & question', ['AVQ', 'WQL', 'WRB', '', 'ADV'],
+      'Adverb, degree', ['AV0', 'QL', 'RB', '', '', 'ADV'],
+      'Adverb, degree, postposed', ['AV0', 'QLP', 'RB', '', '', 'ADV'],
+      'Adverb, nominal', ['AV0', 'RN', 'RB', 'PROP', '', 'ADV'],
+      'Adverb, pronominal', ['', '', '', '', 'PROP', '', 'ADV'],
+
+      'Conjunction, coordination', ['CJC', 'CC', 'CC', 'KON', 'CC', 'COOD'],
+      'Conjunction, coordination, and', ['CJC', 'CC', 'CC', 'KON', 'CC', 'ET'],
+      'Conjunction, subordination', ['CJS', 'CS', 'IN', 'KOUS', 'CS', 'CONJ'],
+      'Conjunction, subordination with to and infinitive', ['', '', '', 'KOUI', '', ''],
+      'Conjunction, complementizer, that', ['CJT', 'CS', 'IN', '', '', 'C'],
+
+      'Determiner', ['DT0', 'DT', 'DT', '', 'DT', 'D'],
+      'Determiner, pronoun', ['DT0', 'DTI', 'DT', '', '', 'D'],
+      'Determiner, pronoun, plural', ['DT0', 'DTS', 'DT', '', '', 'D'],
+      'Determiner, prequalifier', ['DT0', 'ABL', 'DT', '', '', 'D'],
+      'Determiner, prequantifier', ['DT0', 'ABN', 'PDT', '', 'DT', 'D'],
+      'Determiner, pronoun or double conjunction', ['DT0', 'ABX', 'PDT', '', '', 'D'],
+      'Determiner, pronoun or double conjunction', ['DT0', 'DTX', 'DT', '', '', 'D'],
+      'Determiner, article', ['AT0', 'AT', 'DT', 'ART', '', 'D'],
+      'Determiner, postdeterminer', ['DT0', 'AP', 'DT', '', '', 'D'],
+      'Determiner, possessive', ['DPS', 'PP$', 'PRP$', '', '', 'D'],
+      'Determiner, possessive, second', ['DPS', 'PP$', 'PRPS', '', '', 'D'],
+      'Determiner, question', ['DTQ', 'WDT', 'WDT', '', 'DT', 'D'],
+      'Determiner, possessive & question', ['DTQ', 'WP$', 'WP$', '', '', 'D'],
+      'Interjection', ['', '', '', '', '', 'I'],
+      'Localizer', ['', '', '', '', 'LC'],
+
+      'Measure word', ['', '', '', '', 'M'],
+
+      'Noun, common', ['NN0', 'NN', 'NN', 'N', 'NN', 'NN'],
+      'Noun, singular', ['NN1', 'NN', 'NN', 'NN', 'NN', 'N'],
+      'Noun, plural', ['NN2', 'NNS', 'NNS', 'NN', 'NN', 'N'],
+      'Noun, proper, singular', ['NP0', 'NP', 'NNP', 'NE', 'NR', 'N'],
+      'Noun, proper, plural', ['NP0', 'NPS', 'NNPS', 'NE', 'NR', 'N'],
+      'Noun, adverbial', ['NN0', 'NR', 'NN', 'NE', '', 'N'],
+      'Noun, adverbial, plural', ['NN2', 'NRS', 'NNS', '', 'N'],
+      'Noun, temporal', ['', '', '', '', 'NT', 'N'],
+      'Noun, verbal', ['', '', '', '', 'NN', 'N'],
+
+      'Pronoun, nominal (indefinite)', ['PNI', 'PN', 'PRP', '', 'PN', 'CL'],
+      'Pronoun, personal, subject', ['PNP', 'PPSS', 'PRP', 'PPER'],
+      'Pronoun, personal, subject, 3SG', ['PNP', 'PPS', 'PRP', 'PPER'],
+      'Pronoun, personal, object', ['PNP', 'PPO', 'PRP', 'PPER'],
+      'Pronoun, reflexive', ['PNX', 'PPL', 'PRP', 'PRF'],
+      'Pronoun, reflexive, plural', ['PNX', 'PPLS', 'PRP', 'PRF'],
+      'Pronoun, question, subject', ['PNQ', 'WPS', 'WP', 'PWAV'],
+      'Pronoun, question, subject', ['PNQ', 'WPS', 'WPS', 'PWAV'],  # FIXME
+      'Pronoun, question, object', ['PNQ', 'WPO', 'WP', 'PWAV', 'PWAT'],
+      'Pronoun, existential there', ['EX0', 'EX', 'EX'],
+      'Pronoun, attributive demonstrative', ['', '', '', 'PDAT'],
+      'Prounoun, attributive indefinite without determiner', ['', '', '', 'PIAT'],
+      'Pronoun, attributive possessive', ['', '', '', 'PPOSAT', ''],
+      'Pronoun, substituting demonstrative', ['', '', '', 'PDS'],
+      'Pronoun, substituting possessive', ['', '', '', 'PPOSS', ''],
+      'Prounoun, substituting indefinite', ['', '', '', 'PIS'],
+      'Pronoun, attributive relative', ['', '', '', 'PRELAT', ''],
+      'Pronoun, substituting relative', ['', '', '', 'PRELS', ''],
+      'Pronoun, attributive interrogative', ['', '', '', 'PWAT'],
+      'Pronoun, adverbial interrogative', ['', '', '', 'PWAV'],
+
+      'Pronoun, substituting interrogative', ['', '', '', 'PWS'],
+      'Verb, main, finite', ['', '', '', 'VVFIN', '', 'V'],
+      'Verb, main, infinitive', ['', '', '', 'VVINF', '', 'V'],
+      'Verb, main, imperative', ['', '', '', 'VVIMP', '', 'V'],
+      'Verb, base present form (not infinitive)', ['VVB', 'VB', 'VBP', '', '', 'V'],
+      'Verb, infinitive', ['VVI', 'VB', 'VB', 'V', '', 'V'],
+      'Verb, past tense', ['VVD', 'VBD', 'VBD', '', '', 'V'],
+      'Verb, present participle', ['VVG', 'VBG', 'VBG', 'VAPP', '', 'V'],
+      'Verb, past/passive participle', ['VVN', 'VBN', 'VBN', 'VVPP', '', 'V'],
+      'Verb, present, 3SG, -s form', ['VVZ', 'VBZ', 'VBZ', '', '', 'V'],
+      'Verb, auxiliary', ['', '', '', 'VAFIN', '', 'V'],
+      'Verb, imperative', ['', '', '', 'VAIMP', '', 'V'],
+      'Verb, imperative infinitive', ['', '', '', 'VAINF', '', 'V'],
+      'Verb, auxiliary do, base', ['VDB', 'DO', 'VBP', '', '', 'V'],
+      'Verb, auxiliary do, infinitive', ['VDB', 'DO', 'VB', '', '', 'V'],
+      'Verb, auxiliary do, past', ['VDD', 'DOD', 'VBD', '', '', 'V'],
+      'Verb, auxiliary do, present participle', ['VDG', 'VBG', 'VBG', '', '', 'V'],
+      'Verb, auxiliary do, past participle', ['VDN', 'VBN', 'VBN', '', '', 'V'],
+      'Verb, auxiliary do, present 3SG', ['VDZ', 'DOZ', 'VBZ', '', '', 'V'],
+      'Verb, auxiliary have, base', ['VHB', 'HV', 'VBP', 'VA', '', 'V'],
+      'Verb, auxiliary have, infinitive', ['VHI', 'HV', 'VB', 'VAINF', '', 'V'],
+      'Verb, auxiliary have, past', ['VHD', 'HVD', 'VBD', 'VA', '', 'V'],
+      'Verb, auxiliary have, present participle', ['VHG', 'HVG', 'VBG', 'VA', '', 'V'],
+      'Verb, auxiliary have, past participle', ['VHN', 'HVN', 'VBN', 'VAPP', '', 'V'],
+      'Verb, auxiliary have, present 3SG', ['VHZ', 'HVZ', 'VBZ', 'VA', '', 'V'],
+      'Verb, auxiliary be, infinitive', ['VBI', 'BE', 'VB', '', '', 'V'],
+      'Verb, auxiliary be, past', ['VBD', 'BED', 'VBD', '', '', 'V'],
+      'Verb, auxiliary be, past, 3SG', ['VBD', 'BEDZ', 'VBD', '', '', 'V'],
+      'Verb, auxiliary be, present participle', ['VBG', 'BEG', 'VBG', '', '', 'V'],
+      'Verb, auxiliary be, past participle', ['VBN', 'BEN', 'VBN', '', '', 'V'],
+      'Verb, auxiliary be, present, 3SG', ['VBZ', 'BEZ', 'VBZ', '', '', 'V'],
+      'Verb, auxiliary be, present, 1SG', ['VBB', 'BEM', 'VBP', '', '', 'V'],
+      'Verb, auxiliary be, present', ['VBB', 'BER', 'VBP', '', '', 'V'],
+      'Verb, modal', ['VM0', 'MD', 'MD', 'VMFIN', 'VV', 'V'],
+      'Verb, modal', ['VM0', 'MD', 'MD', 'VMINF', 'VV', 'V'],
+      'Verb, modal, finite', ['', '', '', '', 'VMFIN', 'V'],
+      'Verb, modal, infinite', ['', '', '', '', 'VMINF', 'V'],
+      'Verb, modal, past participle', ['', '', '', '', 'VMPP', 'V'],
+
+      'Particle', ['', '', '', '', '', 'PRT'],
+      'Particle, with adverb', ['', '', '', 'PTKA', '', 'PRT'],
+      'Particle, answer', ['', '', '', 'PTKANT', '', 'PRT'],
+      'Particle, negation', ['', '', '', 'PTKNEG', '', 'PRT'],
+      'Particle, separated verb', ['', '', '', 'PTKVZ', '', 'PRT'],
+      'Particle, to as infinitive marker', ['TO0', 'TO', 'TO', 'PTKZU', '', 'PRT'],
+
+      'Preposition, comparative', ['', '', '', 'KOKOM', '', 'P'],
+      'Preposition, to', ['PRP', 'IN', 'TO', '', '', 'P'],
+      'Preposition', ['PRP', 'IN', 'IN', 'APPR', 'P', 'P'],
+      'Preposition, with aritcle', ['', '', '', 'APPART', '', 'P'],
+      'Preposition, of', ['PRF', 'IN', 'IN', '', '', 'P'],
+
+      'Possessive', ['POS', '$', 'POS'],
+
+      'Postposition', ['', '', '', 'APPO'],
+
+      'Circumposition, right', ['', '', '', 'APZR', ''],
+
+      'Interjection, onomatopoeia or other isolate', ['ITJ', 'UH', 'UH', 'ITJ', 'IJ'],
+
+      'Onomatopoeia', ['', '', '', '', 'ON'],
+
+      'Punctuation', ['', '', '', '', 'PU', 'PN'],
+      'Punctuation, sentence ender', ['PUN', '.', '.', '', '', 'PN'],
+
+      'Punctuation, semicolon', ['PUN', '.', '.', '', '', 'PN'],
+      'Puncutation, colon or ellipsis', ['PUN', ':', ':'],
+      'Punctuationm, comma', ['PUN', ',', ',', '$,'],
+      'Punctuation, dash', ['PUN', '-', '-'],
+      'Punctuation, dollar sign', ['PUN', '', '$'],
+      'Punctuation, left bracket', ['PUL', '(', '(', '$('],
+      'Punctuation, right bracket', ['PUR', ')', ')'],
+      'Punctuation, quotation mark, left', ['PUQ', '', '``'],
+      'Punctuation, quotation mark, right', ['PUQ', '', '"'],
+
+      'Punctuation, left bracket', ['PUL', '(', 'PPL'],
+      'Punctuation, right bracket', ['PUR', ')', 'PPR'],
+      'Punctuation, left square bracket', ['PUL', '(', 'LSB'],
+      'Punctuation, right square bracket', ['PUR', ')', 'RSB'],
+      'Punctuation, left curly bracket', ['PUL', '(', 'LCB'],
+      'Punctuation, right curly bracket', ['PUR', ')', 'RCB'],
+
+      'Unknown, foreign words (not in lexicon)', ['UNZ', '(FW-)', 'FW', '', 'FW'],
+
+      'Symbol', ['', '', 'SYM', 'XY'],
+      'Symbol, alphabetical', ['ZZ0', '', ''],
+      'Symbol, list item', ['', '', 'LS'],
+
+      # Not sure about these tags from the Chinese PTB.
+      'Aspect marker', ['', '', '', '', 'AS'],                         # ?
+      'Ba-construction', ['', '', '', '', 'BA'],                       # ?
+      'In relative', ['', '', '', '', 'DEC'],                          # ?
+      'Associative', ['', '', '', '', 'DER'],                          # ?
+      'In V-de or V-de-R construct', ['', '', '', '', 'DER'],          # ?
+      'For words ? ', ['', '', '', '', 'ETC'],                         # ?
+      'In long bei-construct', ['', '', '', '', 'LB'],                 # ?
+      'In short bei-construct', ['', '', '', '', 'SB'],                # ?
+      'Sentence-nal particle', ['', '', '', '', 'SB'],                 # ?
+      'Particle, other', ['', '', '', '', 'MSP'],                      # ?
+      'Before VP', ['', '', '', '', 'DEV'],                            # ?
+      'Verb, ? as main verb', ['', '', '', '', 'VE'],                  # ?
+      'Verb, ????', ['', '', '', '', 'VC']                             # ?
+    ]},
+  enju: {
+    cat_to_category: {
+      'ADJ' => 'adjective',
+      'ADV' => 'adverb',
+      'CONJ' => 'conjunction',
+      'COOD' => 'conjunction',
+      'C' => 'complementizer',
+      'D' => 'determiner',
+      'N' => 'noun',
+      'P' => 'preposition',
+      'PN' => 'punctuation',
+      'SC' => 'conjunction',
+      'V' => 'verb',
+      'PRT' => 'particle'
+    },
+    cat_to_description: [
+      ['ADJ',	'Adjective'],
+      ['ADV',	'Adverb'],
+      ['CONJ',	'Coordination conjunction'],
+      ['C',	'Complementizer'],
+      ['D',	'Determiner'],
+      ['N',	'Noun'],
+      ['P',	'Preposition'],
+      ['SC',	'Subordination conjunction'],
+      ['V',	'Verb'],
+      ['COOD',	'Part of coordination'],
+      ['PN',	'Punctuation'],
+      ['PRT',	'Particle'],
+      ['S',	'Sentence']
+    ],
+    xcat_to_description: [
+      ['COOD',	'Coordinated phrase/clause'],
+      ['IMP',	'Imperative sentence'],
+      ['INV',	'Subject-verb inversion'],
+      ['Q',	'Interrogative sentence with subject-verb inversion'],
+      ['REL',	'A relativizer included'],
+      ['FREL', 'A free relative included'],
+      ['TRACE',	'A trace included'],
+      ['WH', 'A wh-question word included']
+    ],
+    xcat_to_ptb: [
+      ['ADJP', '', 'ADJP'],
+      ['ADJP', 'REL', 'WHADJP'],
+      ['ADJP', 'FREL', 'WHADJP'],
+      ['ADJP', 'WH', 'WHADJP'],
+      ['ADVP', '', 'ADVP'],
+      ['ADVP', 'REL', 'WHADVP'],
+      ['ADVP', 'FREL', 'WHADVP'],
+      ['ADVP', 'WH', 'WHADVP'],
+      ['CONJP', '', 'CONJP'],
+      ['CP', '', 'SBAR'],
+      ['DP', '', 'NP'],
+      ['NP', '', 'NP'],
+      ['NX', 'NX', 'NAC'],
+      ['NP'	'REL'	'WHNP'],
+      ['NP'	'FREL'	'WHNP'],
+      ['NP'	'WH'	'WHNP'],
+      ['PP', '', 'PP'],
+      ['PP', 'REL', 'WHPP'],
+      ['PP', 'WH', 'WHPP'],
+      ['PRT', '', 'PRT'],
+      ['S', '', 'S'],
+      ['S', 'INV', 'SINV'],
+      ['S', 'Q', 'SQ'],
+      ['S', 'REL', 'SBAR'],
+      ['S', 'FREL', 'SBAR'],
+      ['S', 'WH', 'SBARQ'],
+      ['SCP', '', 'SBAR'],
+      ['VP', '', 'VP'],
+      ['VP', '', 'VP'],
+      ['', '', 'UK']
+    ]},
+  paris7: {
+    tag_to_category: {
+      'C' => :complementizer,
+      'PN' => :punctuation,
+      'SC' => :conjunction
+    }
+    # Paris7 Treebank functional tags
+=begin
+  SUJ (subject)
+  OBJ (direct object)
+  ATS (predicative complement of a subject)
+  ATO (predicative complement of a direct object)
+  MOD (modifier or adjunct)
+  A-OBJ (indirect complement introduced by à)
+  DE-OBJ (indirect complement introduced by de)
+  P-OBJ (indirect complement introduced by another preposition)
+=end
+  },
+  ptb: {
+    escape_characters: {
+    '(' => '-LRB-',
+    ')' => '-RRB-',
+    '[' => '-LSB-',
+    ']' => '-RSB-',
+    '{' => '-LCB-',
+    '}' => '-RCB-'
+    },
+    phrase_tag_to_description: [
+      ['S', 'Paris7 declarative clause'],
+      ['SBAR', 'Clause introduced by a (possibly empty) subordinating conjunction'],
+      ['SBARQ', 'Direct question introduced by a wh-word or a wh-phrase'],
+      ['SINV', 'Inverted declarative sentence'],
+      ['SQ', 'Inverted yes/no question']
+    ]
+  } 
+}

data/lib/treat/config/{workers → data/workers}/extractors.rb

@@ -6,7 +6,7 @@
   },
   time: {
     type: :annotator,
-    targets: [:phrase]
+    targets: [:group]
   },
   topics: {
     type: :annotator,
@@ -22,18 +22,10 @@
   },
   name_tag: {
     type: :annotator,
-    targets: [:phrase, :word]
-  },
-  coreferences: {
-    type: :annotator,
-    targets: [:zone]
+    targets: [:group]
   },
   tf_idf: {
     type: :annotator,
     targets: [:word]
-  },
-  summary: {
-    type: :annotator,
-    targets: [:document]
   }
 }

data/lib/treat/config/{workers → data/workers}/lexicalizers.rb

@@ -1,11 +1,12 @@
 {
   taggers: {
     type: :annotator,
-    targets: [:phrase, :token]
+    targets: [:group, :token],
+    recursive: true
   },
   categorizers: {
     type: :annotator,
-    targets: [:phrase, :token],
+    targets: [:group, :token],
     recursive: true
   },
   sensers: {
@@ -14,5 +15,5 @@
     preset_option: :nym,
     presets: [:synonyms, :antonyms, 
               :hyponyms, :hypernyms],
-  }        
+  }
 }

data/lib/treat/config/{workers → data/workers}/processors.rb

@@ -1,7 +1,7 @@
 {
   chunkers: {
     type: :transformer,
-    targets: [:document],
+    targets: [:document, :section],
     default: :autoselect
   },
   segmenters: {
@@ -10,10 +10,10 @@
   },
   tokenizers: {
     type: :transformer,
-    targets: [:sentence, :phrase]
+    targets: [:group]
   },
   parsers: {
     type: :transformer,
-    targets: [:sentence, :phrase] 
+    targets: [:group] 
   }
 }

data/lib/treat/config/importable.rb

@@ -0,0 +1,31 @@
+# Mixin that is extended by Treat::Config
+# in order to provide a single point of 
+# access method to trigger the import.
+module Treat::Config::Importable
+  
+  # Import relies on each configuration.
+  require_relative 'configurable'
+  
+   # Store all the configuration in self.config
+  def self.extended(base)
+    class << base; attr_accessor :config; end
+  end
+
+  # Main function; loads all configuration options.
+  def import!
+    config, c = {}, Treat::Config::Configurable
+    definition = :define_singleton_method
+    Treat::Config.constants.each do |const|
+      next if const.to_s.downcase.is_mixin?
+      klass = Treat::Config.const_get(const)
+      klass.class_eval { extend c }.configure!
+      name = const.to_s.downcase.intern
+      config[name] = klass.config
+      Treat.send(definition, name) do
+        Treat::Config.config[name]
+      end
+    end
+    self.config = config.to_struct
+  end
+  
+end

data/lib/treat/config/paths.rb

@@ -0,0 +1,23 @@
+# Generates the following path config options:
+# Treat.paths.tmp, Treat.paths.bin, Treat.paths.lib,
+# Treat.paths.models, Treat.paths.files, Treat.paths.spec.
+class Treat::Config::Paths
+  
+  # Get the path configuration based on the 
+  # directory structure loaded into Paths.
+  # Note that this doesn't call super, as
+  # there is no external config files to load.
+  def self.configure!
+    root = File.dirname(File.expand_path(         # FIXME
+    __FILE__)).split('/')[0..-4].join('/') + '/'
+    self.config = Hash[
+    # Get a list of directories in treat/
+    Dir.glob(root + '*').select do |path|
+      FileTest.directory?(path)
+    # Map to pairs of [:name, path]
+    end.map do |path|
+      [File.basename(path).intern, path + '/']
+    end]
+  end
+  
+end

data/lib/treat/config/tags.rb

@@ -0,0 +1,37 @@
+# Handles all configuration related 
+# to understanding of part of speech
+# and phrasal tags.
+class Treat::Config::Tags
+  
+  # Generate a map of word and phrase tags 
+  # to their syntactic category, keyed by 
+  # tag set.
+  def self.configure!
+    super
+    config = self.config[:aligned].dup
+    word_tags, phrase_tags, tag_sets = 
+    config[:word_tags], config[:phrase_tags]
+    tag_sets = config[:tag_sets]
+    config[:word_tags_to_category] = 
+    align_tags(word_tags, tag_sets)
+    config[:phrase_tags_to_category] =
+    align_tags(phrase_tags, tag_sets)
+    self.config[:aligned] = config
+ end
+ 
+ # Helper methods for tag set config.
+ # Align tag tags in the tag set 
+ def self.align_tags(tags, tag_sets)
+   wttc = {}
+   tags.each_slice(2) do |desc, tags|
+     category = desc.gsub(',', ' ,').
+     split(' ')[0].downcase
+     tag_sets.each_with_index do |tag_set, i|
+       next unless tags[i]
+       wttc[tags[i]] ||= {}
+       wttc[tags[i]][tag_set] = category
+     end
+   end; return wttc
+ end
+   
+end

data/lib/treat/core/dsl.rb

@@ -0,0 +1,55 @@
+module Treat::Core::DSL
+  
+  # Message for deprecation of old DSL syntax.
+  DeprecationMessage = "The DSL that used " +
+  "capitalized entity names is now deprecated. " +
+  "Use `include Treat::Core::DSL` along with " +
+  "lowercase names from now on." 
+  
+  # Map all classes in Treat::Entities to
+  # a global builder function (entity, word,
+  # phrase, punctuation, symbol, list, etc.)
+  def self.included(base)
+    self.sweeten_entities(base)
+    self.sweeten_learning(base)
+  end
+  
+  # Map all classes in Treat::Entities to
+  # a global builder function (word, etc.)
+  def self.sweeten_entities(base, on = true)
+    Treat.core.entities.list.each do |type|
+      kname = type.cc.intern
+      mname = type.intern
+      klass = Treat::Entities.const_get(kname)
+      base.class_eval do
+        define_method(mname.capitalize) do |*args|
+          raise DeprecationMessage
+        end
+        old_mm = instance_method(:method_missing)
+        define_method(:method_missing) do |sym,*args,&block|
+          return klass.build(*args) if sym == mname
+          old_mm.bind(self).call(sym,*args,&block)
+        end
+      end
+    end
+  end
+  
+  # Map all classes in the Learning module
+  # to a global builder function. Defines:
+  # dataset, export, feature, tag, problem
+  # question.
+  def self.sweeten_learning(base, on = true)
+    Treat::Learning.constants.each do |kname|
+      mname = kname.downcase
+      klass = Treat::Learning.const_get(kname)
+      base.class_eval do
+        old_mm = instance_method(:method_missing)
+        define_method(:method_missing) do |sym,*args,&block|
+          return klass.new(*args) if sym == mname
+          old_mm.bind(self).call(sym,*args,&block)
+        end
+      end
+    end
+  end
+  
+end

data/lib/treat/{installer.rb → core/installer.rb}

@@ -1,13 +1,7 @@
 # A dependency manager for Treat language plugins.
-# It can be called by using Treat.install(language).
-module Treat::Installer
-
-  # Require the Rubygem dependency installer.
-  silence_warnings do
-    require 'rubygems/dependency_installer'
-  end
+# Usage: Treat::Installer.install('language')
+module Treat::Core::Installer
   
-  require 'treat/version'  
   require 'schiphol'
 
   # Address of the server with the files.
@@ -31,6 +25,11 @@ module Treat::Installer
   # dependencies for a specific language.
   def self.install(language = 'english')
     
+    # Require the Rubygem dependency installer.
+    silence_warnings do
+      require 'rubygems/dependency_installer'
+    end
+    
     @@installer = Gem::DependencyInstaller.new
     
     if language == 'travis'
@@ -182,10 +181,9 @@ module Treat::Installer
     begin
       puts "Installing #{dependency}...\n"
       @@installer.install(dependency)
-    rescue Exception => error
-      raise
-      puts "Couldn't install gem '#{dependency}' " +
-           "(#{error.message})."
+    rescue Gem::InstallError => error
+      puts "Warning: couldn't install " +
+      "gem '#{dependency}' (#{error.message})."
     end
     
   end

data/lib/treat/core/server.rb

@@ -0,0 +1,40 @@
+class Treat::Core::Server 
+  
+  # Refer to http://rack.rubyforge.org/doc/classes/Rack/Server.html
+  # for possible options to configure.
+  def initialize(handler = 'thin', options = {})
+    require 'json'; require 'rack'
+    @handler, @options = handler.capitalize, options
+  end
+  
+  def start
+    handler = Rack::Handler.const_get(@handler)
+    handler.run(self, @options)
+  end
+  
+  def call(env)
+    headers = { 'content-type' => 'application/json' }
+    rack_input = env["rack.input"].read
+    if rack_input.strip == ''
+      return [500, headers, {
+        'error' => 'Empty JSON request.'
+      }]
+    end
+    rack_json = JSON.parse(rack_input)
+    unless rack_json['type'] && 
+    rack_json['value'] && rack_json['do']
+      return [500, headers, {
+        'error' => 'Must specify "type", "value" and "do".'
+      }]
+    end
+    if rack_json['conf']
+      # Set the configuration.
+    end
+    method = rack_json['type'].capitalize.intern
+    resp = send(method, rack_json[value]).do(rack_json['do'])
+    
+    response = [rack_input.to_json]
+    [200, headers, response]
+  end
+  
+end