|
 |
ae2451 |
|
|
|
ae2451 |
# HG changeset patch
|
|
|
ae2451 |
# User Serhiy Storchaka <storchaka@gmail.com>
|
|
|
ae2451 |
# Date 1382204269 -10800
|
|
|
ae2451 |
# Node ID 214c0aac7540947d88a38ff0061734547ef86710
|
|
|
ae2451 |
# Parent c207ac413457a1b834e4b7dcf1a6836cd6e036e3
|
|
|
ae2451 |
Issue #19279: UTF-7 decoder no more produces illegal unicode strings.
|
|
|
ae2451 |
|
|
|
ae2451 |
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
|
|
|
ae2451 |
--- a/Lib/test/test_codecs.py
|
|
|
ae2451 |
+++ b/Lib/test/test_codecs.py
|
|
|
ae2451 |
@@ -611,6 +611,35 @@ class UTF7Test(ReadTest):
|
|
|
ae2451 |
]
|
|
|
ae2451 |
)
|
|
|
ae2451 |
|
|
|
ae2451 |
+ def test_errors(self):
|
|
|
ae2451 |
+ tests = [
|
|
|
ae2451 |
+ ('a\xffb', u'a\ufffdb'),
|
|
|
ae2451 |
+ ('a+IK', u'a\ufffd'),
|
|
|
ae2451 |
+ ('a+IK-b', u'a\ufffdb'),
|
|
|
ae2451 |
+ ('a+IK,b', u'a\ufffdb'),
|
|
|
ae2451 |
+ ('a+IKx', u'a\u20ac\ufffd'),
|
|
|
ae2451 |
+ ('a+IKx-b', u'a\u20ac\ufffdb'),
|
|
|
ae2451 |
+ ('a+IKwgr', u'a\u20ac\ufffd'),
|
|
|
ae2451 |
+ ('a+IKwgr-b', u'a\u20ac\ufffdb'),
|
|
|
ae2451 |
+ ('a+IKwgr,', u'a\u20ac\ufffd'),
|
|
|
ae2451 |
+ ('a+IKwgr,-b', u'a\u20ac\ufffd-b'),
|
|
|
ae2451 |
+ ('a+IKwgrB', u'a\u20ac\u20ac\ufffd'),
|
|
|
ae2451 |
+ ('a+IKwgrB-b', u'a\u20ac\u20ac\ufffdb'),
|
|
|
ae2451 |
+ ('a+/,+IKw-b', u'a\ufffd\u20acb'),
|
|
|
ae2451 |
+ ('a+//,+IKw-b', u'a\ufffd\u20acb'),
|
|
|
ae2451 |
+ ('a+///,+IKw-b', u'a\uffff\ufffd\u20acb'),
|
|
|
ae2451 |
+ ('a+////,+IKw-b', u'a\uffff\ufffd\u20acb'),
|
|
|
ae2451 |
+ ]
|
|
|
ae2451 |
+ for raw, expected in tests:
|
|
|
ae2451 |
+ self.assertRaises(UnicodeDecodeError, codecs.utf_7_decode,
|
|
|
ae2451 |
+ raw, 'strict', True)
|
|
|
ae2451 |
+ self.assertEqual(raw.decode('utf-7', 'replace'), expected)
|
|
|
ae2451 |
+
|
|
|
ae2451 |
+ def test_nonbmp(self):
|
|
|
ae2451 |
+ self.assertEqual(u'\U000104A0'.encode(self.encoding), '+2AHcoA-')
|
|
|
ae2451 |
+ self.assertEqual(u'\ud801\udca0'.encode(self.encoding), '+2AHcoA-')
|
|
|
ae2451 |
+ self.assertEqual('+2AHcoA-'.decode(self.encoding), u'\U000104A0')
|
|
|
ae2451 |
+
|
|
|
ae2451 |
class UTF16ExTest(unittest.TestCase):
|
|
|
ae2451 |
|
|
|
ae2451 |
def test_errors(self):
|
|
|
ae2451 |
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
|
|
|
ae2451 |
--- a/Objects/unicodeobject.c
|
|
|
ae2451 |
+++ b/Objects/unicodeobject.c
|
|
|
ae2451 |
@@ -1671,6 +1671,7 @@ PyObject *PyUnicode_DecodeUTF7Stateful(c
|
|
|
ae2451 |
(base64buffer >> (base64bits-16));
|
|
|
ae2451 |
base64bits -= 16;
|
|
|
ae2451 |
base64buffer &= (1 << base64bits) - 1; /* clear high bits */
|
|
|
ae2451 |
+ assert(outCh <= 0xffff);
|
|
|
ae2451 |
if (surrogate) {
|
|
|
ae2451 |
/* expecting a second surrogate */
|
|
|
ae2451 |
if (outCh >= 0xDC00 && outCh <= 0xDFFF) {
|
|
|
ae2451 |
@@ -1737,6 +1738,7 @@ PyObject *PyUnicode_DecodeUTF7Stateful(c
|
|
|
ae2451 |
inShift = 1;
|
|
|
ae2451 |
shiftOutStart = p;
|
|
|
ae2451 |
base64bits = 0;
|
|
|
ae2451 |
+ base64buffer = 0;
|
|
|
ae2451 |
}
|
|
|
ae2451 |
}
|
|
|
ae2451 |
else if (DECODE_DIRECT(ch)) { /* character decodes as itself */
|
|
|
ae2451 |
|