def prepare(self, word_type):
"""
Generates trie structure containing all words from clp
:param word_type: part of speech of which kind read words from clp data
:return: None
"""
old = ""
index = 0
keys = []
values = []
for i in range(16777231, 18663982):
if Util.is_word_unneeded(i):
continue
if Util.is_word_appropriate_type(self.plp.label(i)[0], word_type):
continue
form = self.plp.bform(i)
if old != form:
for s in self.plp.forms(i):
if len(s) > 0:
a = Util.substring(s, form)
to_remove = s[len(a): len(s)]
to_add = form[len(a): len(form)]
keys.append(Util.reverse(s))
a = unicode(to_remove).encode('utf-8')
b = unicode(to_add).encode('utf-8')
values.append((a, b))
index += 1
old = self.plp.bform(i)
return zip(keys, values)
