From 4f4d22dabd279f30f3afefdacf83acf95b6dec92 Mon Sep 17 00:00:00 2001
From: Tammo Behrends <behr.tam@gmail.com>
Date: Sat, 21 Nov 2015 00:15:51 +0100
Subject: [PATCH] word-count: Add test case for unicode support

See #177
---
 word-count/example.py         | 10 +++++++++-
 word-count/word_count_test.py | 15 +++++++++++++++
 2 files changed, 24 insertions(+), 1 deletion(-)

diff --git a/word-count/example.py b/word-count/example.py
index 1b5475c1927..1207bc6822d 100644
--- a/word-count/example.py
+++ b/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())
diff --git a/word-count/word_count_test.py b/word-count/word_count_test.py
index 13f48204aa2..1c3373c42e3 100644
--- a/word-count/word_count_test.py
+++ b/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()