tests/stand_alone/cross_serial.py

   1 # This Python file uses the following encoding: utf-8
   2 '''
   3 Created on May 19, 2011
   4
   5 @author: mjacob
   6 '''
   7 from mjacob.nltk.grammar.TreeAdjoiningGrammar import TreeAdjoiningGrammar
   8 from mjacob.nltk.parse.tag.TagChartParser import TagChartParser
   9 from mjacob.nltk.parse.tag.earley.rules import TAG_EARLEY_STRATEGY
  10
  11 grammar = TreeAdjoiningGrammar(filename="../integration/tag_chart/cross_serial_2/grammar.yaml")
  12 parser = TagChartParser(grammar)
  13
  14 for s in ["Jan Piet Marie de kinderen zag helpen leren zwemmen"]:
  15     tokens = s.split(' ')
  16     chart = parser.chart_parse(tokens)
  17 #    for goal_edge, deduction in chart.deduction_paths(grammar.start().symbol(), TAG_EARLEY_STRATEGY):
  18 #        print "GOAL: %s" % (goal_edge)
  19 #        i = 1
  20 #        for x,y,z in deduction:
  21 #            print "% 6i %s <- %s" % (i, x, y)
  22 #            for tree in z:
  23 #                print "       %s" % (z,)
  24 #            i += 1
  25
  26     print("   %s" % ("\n   ".join(tree.pprint(margin=10000) for tree in chart.parses(grammar.start().symbol(), TAG_EARLEY_STRATEGY))))
  27     #print parser.nbest_parse(tokens)
  28     #print parser.parse(tokens)