bpo-32498: urllib.parse.unquote also accepts bytes (GH-7768)

2 files changed, 25 insertions, 2 deletions

diff --git a/Lib/test/test_urllib.py b/Lib/test/test_urllib.py
index 9a6b5f6..3f59c66 100644
--- a/Lib/test/test_urllib.py
+++ b/Lib/test/test_urllib.py
@@ -1049,8 +1049,6 @@ class UnquotingTests(unittest.TestCase):
                          "%s" % result)
         self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, None)
         self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, ())
-        with support.check_warnings(('', BytesWarning), quiet=True):
-            self.assertRaises((TypeError, AttributeError), urllib.parse.unquote, b'')
 
     def test_unquoting_badpercent(self):
         # Test unquoting on bad percent-escapes
@@ -1210,6 +1208,29 @@ class UnquotingTests(unittest.TestCase):
         self.assertEqual(expect, result,
                          "using unquote(): %r != %r" % (expect, result))
 
+    def test_unquoting_with_bytes_input(self):
+        # ASCII characters decoded to a string
+        given = b'blueberryjam'
+        expect = 'blueberryjam'
+        result = urllib.parse.unquote(given)
+        self.assertEqual(expect, result,
+                         "using unquote(): %r != %r" % (expect, result))
+
+        # A mix of non-ASCII hex-encoded characters and ASCII characters
+        given = b'bl\xc3\xa5b\xc3\xa6rsyltet\xc3\xb8y'
+        expect = 'bl\u00e5b\u00e6rsyltet\u00f8y'
+        result = urllib.parse.unquote(given)
+        self.assertEqual(expect, result,
+                         "using unquote(): %r != %r" % (expect, result))
+
+        # A mix of non-ASCII percent-encoded characters and ASCII characters
+        given = b'bl%c3%a5b%c3%a6rsyltet%c3%b8j'
+        expect = 'bl\u00e5b\u00e6rsyltet\u00f8j'
+        result = urllib.parse.unquote(given)
+        self.assertEqual(expect, result,
+                         "using unquote(): %r != %r" % (expect, result))
+
+
 class urlencode_Tests(unittest.TestCase):
     """Tests for urlencode()"""
 
diff --git a/Lib/urllib/parse.py b/Lib/urllib/parse.py
index b660878..3a38dc1 100644
--- a/Lib/urllib/parse.py
+++ b/Lib/urllib/parse.py
@@ -631,6 +631,8 @@ def unquote(string, encoding='utf-8', errors='replace'):
 
     unquote('abc%20def') -> 'abc def'.
     """
+    if isinstance(string, bytes):
+        return unquote_to_bytes(string).decode(encoding, errors)
     if '%' not in string:
         string.split
         return string