Fixing encoding issue #3

This commit is contained in:
maxirmx 2015-10-21 15:52:34 +03:00
parent fe9d2e2c4e
commit fcbfff043f

View File

@ -19,7 +19,10 @@ class Lemmatizer(object):
index[pos] = read_index(path.join(data_dir, 'wordnet', 'index.%s' % pos))
exc[pos] = read_exc(path.join(data_dir, 'wordnet', '%s.exc' % pos))
if path.exists(path.join(data_dir, 'vocab', 'lemma_rules.json')):
rules = json.load(open(path.join(data_dir, 'vocab', 'lemma_rules.json'), encoding="UTF-8"))
if sys.version_info[0] == 3:
rules = json.load(open(path.join(data_dir, 'vocab', 'lemma_rules.json'), encoding="utf_8"))
else:
rules = json.load(open(path.join(data_dir, 'vocab', 'lemma_rules.json')))
else:
rules = {}
return cls(index, exc, rules)