Merge pull request #252 from behrtam/add-unicode-test-word-count

word-count: Add test case for unicode support

Author: Dog

word-count/example.py

@@ -1,7 +1,15 @@
 from collections import Counter
 
 
+# to be backwards compatible with the old Python 2.X
+def decode_if_needed(string):
+    try:
+        return string.decode('utf-8')
+    except AttributeError:
+        return string
+
+
 def word_count(text):
     replace_nonalpha = lambda c: c.lower() if c.isalnum() else ' '
-    text = ''.join(replace_nonalpha(c) for c in text)
+    text = ''.join(replace_nonalpha(c) for c in decode_if_needed(text))
     return Counter(text.split())

word-count/word_count_test.py

@@ -1,8 +1,17 @@
+# -*- coding: utf-8 -*-
 import unittest
 
 from wordcount import word_count
 
 
+# to be backwards compatible with the old Python 2.X
+def decode_if_needed(string):
+    try:
+        return string.decode('utf-8')
+    except AttributeError:
+        return string
+
+
 class WordCountTests(unittest.TestCase):
 
     def test_count_one_word(self):
@@ -69,5 +78,11 @@ def test_non_alphanumeric(self):
             word_count('hey,my_spacebar_is_broken.')
         )
 
+    def test_unicode(self):
+        self.assertEqual(
+            {decode_if_needed('до'): 1, decode_if_needed('свидания'): 1},
+            word_count('до🖖свидания!')
+        )
+
 if __name__ == '__main__':
     unittest.main()