# -*- coding: utf-8 -*-

  from BaseProcessor import baseProcessor
  import nltk

  import re

  from LiaTools import *
  class Orkis(baseProcessor):
      """ Processor for Orkis """
      def __init__(self,dirtyString):

  	self.lem=u""

          self.tagger=Tagger()

          self.dico ={}
          self.string=dirtyString

          print(dirtyString)

      def isReady(self):

          self.tagger.isReady()

      def __unicode__(self):
  	string = u""

          for word in self.dico:

              string += word+u";"

              for lemWord in self.dico[word][0]:

                  string += lemWord + u" "

              string+=u"
  "

          return string

      def clean(self):
          stopword=StopWord()
          self.string=stopword.RemoveStopList(self.string)
      def insertLem(self):

  	self.lem=u""
          self.cleanString=self.tagger.clean(self.string).rstrip()

          taggedString=self.tagger.tagg(self.cleanString).rstrip().decode("utf8")
          self.tableLem = taggedString.split(u"
  ")
          for line in taggedString.rstrip().split(u"
  "):
              table = line.rstrip().split(u" ")
  	    #print(u"table2" + table[2])

  	    if not table[2].isspace():
                  if not table[0] in self.dico :
                      self.dico[table[0]]=[set(),set()]
                  self.dico[table[0]][0].add(table[2])

  	        self.lem = self.lem +u"
  "+ table[2]

      def getDico(self):
          self.clean()
          self.insertLem()

          self.table=[]

          for i,v in self.dico.iteritems():    

              if not re.match(r".s>",i):

                  list=[]          
                  list.append(i)

                  for indice in v[0]:

                      list.append(indice) 

                  ligne= " ".join(list)     

                  self.table.append(ligne) 

          return u"
  ".join(self.table)