Issue #4426: The UTF-7 decoder was too strict and didn't accept some legal sequences.

Patch by Nick Barnes and Victor Stinner.

Include/unicodeobject.h

@@ -740,10 +740,8 @@ PyAPI_FUNC(PyObject*) PyUnicode_DecodeUTF7Stateful(
 PyAPI_FUNC(PyObject*) PyUnicode_EncodeUTF7(
     const Py_UNICODE *data, 	/* Unicode char buffer */
     Py_ssize_t length,	 		/* number of Py_UNICODE chars to encode */
-    int encodeSetO,             /* force the encoder to encode characters in
-                                   Set O, as described in RFC2152 */
-    int encodeWhiteSpace,       /* force the encoder to encode space, tab,
-                                   carriage return and linefeed characters */
+    int base64SetO,		/* Encode RFC2152 Set O characters in base64 */
+    int base64WhiteSpace,	/* Encode whitespace (sp, ht, nl, cr) in base64 */
     const char *errors		/* error handling */
     );
 

Lib/test/test_unicode.py

@@ -521,19 +521,28 @@ class UnicodeTest(
             (u'+?', '+-?'),
             (ur'\\?', '+AFwAXA?'),
             (ur'\\\?', '+AFwAXABc?'),
-            (ur'++--', '+-+---')
+            (ur'++--', '+-+---'),
+            (u'\U000abcde', '+2m/c3g-'),                  # surrogate pairs
+            (u'/', '/'),
         ]
 
         for (x, y) in utfTests:
             self.assertEqual(x.encode('utf-7'), y)
 
-        # surrogates not supported
+        # Unpaired surrogates not supported
         self.assertRaises(UnicodeError, unicode, '+3ADYAA-', 'utf-7')
 
-        self.assertEqual(unicode('+3ADYAA-', 'utf-7', 'replace'), u'\ufffd')
-
-        # Issue #2242: crash on some Windows/MSVC versions
-        self.assertRaises(UnicodeDecodeError, '+\xc1'.decode, 'utf-7')
+        self.assertEqual(unicode('+3ADYAA-', 'utf-7', 'replace'), u'\ufffd\ufffd')
+
+        # Direct encoded characters
+        set_d = "ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz0123456789'(),-./:?"
+        # Optional direct characters
+        set_o = '!"#$%&*;<=>@[]^_`{|}'
+        for c in set_d:
+            self.assertEqual(c.encode('utf7'), c.encode('ascii'))
+            self.assertEqual(c.encode('ascii').decode('utf7'), c)
+        for c in set_o:
+            self.assertEqual(c.encode('ascii').decode('utf7'), c)
 
     def test_codecs_utf8(self):
         self.assertEqual(u''.encode('utf-8'), '')

Misc/NEWS

@@ -12,6 +12,9 @@ What's New in Python 2.7 alpha 1
 Core and Builtins
 -----------------
 
+- Issue #4426: The UTF-7 decoder was too strict and didn't accept some legal
+  sequences. Patch by Nick Barnes and Victor Stinner.
+
 - Issue #1588: Add complex.__format__. For example, 
   format(complex(1, 2./3), '.5') now produces a sensible result.