Fix JSON encoding issue on load

spacy/lemmatizer.py

@@ -28,7 +28,7 @@ class Lemmatizer(object):
                     exc[pos] = read_exc(file_)
             else:
                 exc[pos] = {}
-        with (path / 'vocab' / 'lemma_rules.json').open('rb') as file_:
+        with (path / 'vocab' / 'lemma_rules.json').open('r', encoding='utf8') as file_:
             rules = json.load(file_)
         return cls(index, exc, rules)
 

spacy/matcher.pyx

@@ -175,7 +175,7 @@ cdef class Matcher:
     @classmethod
     def load(cls, path, vocab):
         if (path / 'gazetteer.json').exists():
-            with (path / 'gazetteer.json').open('rb') as file_:
+            with (path / 'gazetteer.json').open('r', encoding='utf8') as file_:
                 patterns = json.load(file_)
         else:
             patterns = {}