From 408ce64883cb450ab0acbf9694e90e75d3952b46 Mon Sep 17 00:00:00 2001
From: Erickson Silva <erickson.silva@lavid.ufpb.br>
Date: Fri, 29 Jan 2016 15:48:11 -0300
Subject: [PATCH] Remove caractere especial 'marcador' ao classificar sentença.

---
 src/ClassificaSentencas.py | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/src/ClassificaSentencas.py b/src/ClassificaSentencas.py
index e4ad7d9..70d4129 100644
--- a/src/ClassificaSentencas.py
+++ b/src/ClassificaSentencas.py
@@ -50,6 +50,7 @@ class ClassificaSentencas(object):
 			decodificada = regex2.sub('',regex.sub('"',s.replace("–", "-").replace("—", "-"))).decode("utf-8")
 		except:
 			decodificada = s.decode("utf-8")
+
 		return Toqueniza.TOK_PORT.tokenize(decodificada)
 
 	def obter_classificacao_morfologica(self):
@@ -63,7 +64,7 @@ class ClassificaSentencas(object):
 		while (anotada[0][1] is None):
 			time.sleep(random.choice(sleep_times))
 			anotada = AnotaCorpus.anota_sentencas([s],etiquetador,"hunpos")[0]
-		regex = re.compile('[%s]' % re.escape('!"#&\'()*+,-./:;<=>?@[\\]^_`{|}~'))
+		regex = re.compile('[%s]' % re.escape(u'\u2022''!"#&\'()*+,-./:;<=>?@[\\]^_`{|}~'))
 		tag_punctuation = [".",",","QT","("]
 		anotada_corrigida = []
 		for x in anotada:
--
libgit2 0.21.2