tests/stand_alone/test_tag_chart_parser.py

   1 # This Python file uses the following encoding: utf-8
   2 '''
   3 Created on May 30, 2011
   4
   5 @author: mjacob
   6 '''
   7 import nltk
   8 from mjacob.nltk.grammar.TreeAdjoiningGrammar import TreeAdjoiningGrammar
   9 from mjacob.nltk.parse.tag.earley.rules import TAG_EARLEY_STRATEGY
  10 from mjacob.nltk.parse.tag.TagChartParser import TagChartParser
  11
  12 VAR = 1
  13
  14 raw_data = nltk.data.load('grammars/large_grammars/atis_sentences.txt', format='raw')
  15 sentences = [x[0] for x in nltk.parse.util.extract_test_sentences(raw_data)][:1]
  16
  17 if VAR == 0:
  18     grammar = TreeAdjoiningGrammar(cfg=nltk.data.load('grammars/large_grammars/atis.cfg'))
  19 elif VAR == 1:
  20     grammar = TreeAdjoiningGrammar("../integration/tag_chart/degenerate_1/grammar.yaml")
  21 elif VAR == 2:
  22     grammar = TreeAdjoiningGrammar("../integration/tag_chart/cross_serial_2/grammar.yaml")
  23
  24 parser = TagChartParser(grammar)
  25
  26 if VAR == 0:
  27     chart = parser.chart_parse(sentences[0])
  28 elif VAR == 1:
  29     chart = parser.chart_parse(['John'])
  30 elif VAR == 2:
  31     chart = parser.chart_parse("Jan Piet Marie de kinderen zag helpen leren zwemmen".split(' '))
  32
  33 print(chart)
  34 print("\n\n\n")
  35 print(chart.parses(grammar.start().symbol(), TAG_EARLEY_STRATEGY))