From 673c81bbb4c832c645768302415108327f9c878a Mon Sep 17 00:00:00 2001
From: svlandeg <sofie.vanlandeghem@gmail.com>
Date: Tue, 2 Apr 2019 13:52:07 +0200
Subject: [PATCH] unicode string for python 2.7

---
 spacy/lang/en/stop_words.py              | 2 +-
 spacy/tests/regression/test_issue3521.py | 8 ++++----
 2 files changed, 5 insertions(+), 5 deletions(-)

diff --git a/spacy/lang/en/stop_words.py b/spacy/lang/en/stop_words.py
index bdc36bdd7..07d4ff34c 100644
--- a/spacy/lang/en/stop_words.py
+++ b/spacy/lang/en/stop_words.py
@@ -70,6 +70,6 @@ yet you your yours yourself yourselves
 )
 
 for hyphen in ["'", "`", "‘", "´", "’"]:
-    for stopword in "n't 'd 'll 'm 're 's 've".split():
+    for stopword in u"n't 'd 'll 'm 're 's 've".split():
         STOP_WORDS.add(stopword.replace("'", hyphen))
 
diff --git a/spacy/tests/regression/test_issue3521.py b/spacy/tests/regression/test_issue3521.py
index 2a4568a17..1fe5e00d7 100644
--- a/spacy/tests/regression/test_issue3521.py
+++ b/spacy/tests/regression/test_issue3521.py
@@ -4,10 +4,10 @@ import pytest
 @pytest.mark.parametrize(
     "word",
     [
-        "don't",
-        "don’t",
-        "I'd",
-        "I’d",
+        u"don't",
+        u"don’t",
+        u"I'd",
+        u"I’d",
     ],
 )
 def test_issue3521(en_tokenizer, word):