Issue #13333: The UTF-7 decoder now accepts lone surrogates

(the encoder already accepts them).
This commit is contained in:
Antoine Pitrou 2011-11-15 01:42:21 +01:00
parent c2fe57762b
commit 5418ee0b9a
3 changed files with 19 additions and 12 deletions

View File

@ -1091,10 +1091,18 @@ class UnicodeTest(string_tests.CommonTest,
for (x, y) in utfTests:
self.assertEqual(x.encode('utf-7'), y)
# Unpaired surrogates not supported
self.assertRaises(UnicodeError, str, b'+3ADYAA-', 'utf-7')
# Unpaired surrogates are passed through
self.assertEqual('\uD801'.encode('utf-7'), b'+2AE-')
self.assertEqual('\uD801x'.encode('utf-7'), b'+2AE-x')
self.assertEqual('\uDC01'.encode('utf-7'), b'+3AE-')
self.assertEqual('\uDC01x'.encode('utf-7'), b'+3AE-x')
self.assertEqual(b'+2AE-'.decode('utf-7'), '\uD801')
self.assertEqual(b'+2AE-x'.decode('utf-7'), '\uD801x')
self.assertEqual(b'+3AE-'.decode('utf-7'), '\uDC01')
self.assertEqual(b'+3AE-x'.decode('utf-7'), '\uDC01x')
self.assertEqual(str(b'+3ADYAA-', 'utf-7', 'replace'), '\ufffd\ufffd')
self.assertEqual('\uD801\U000abcde'.encode('utf-7'), b'+2AHab9ze-')
self.assertEqual(b'+2AHab9ze-'.decode('utf-7'), '\uD801\U000abcde')
# Issue #2242: crash on some Windows/MSVC versions
self.assertEqual(b'+\xc1'.decode('utf-7'), '\xc1')

View File

@ -10,6 +10,9 @@ What's New in Python 3.2.3?
Core and Builtins
-----------------
- Issue #13333: The UTF-7 decoder now accepts lone surrogates (the encoder
already accepts them).
- Issue #13342: input() used to ignore sys.stdin's and sys.stdout's unicode
error handler in interactive mode (when calling into PyOS_Readline()).

View File

@ -2282,21 +2282,17 @@ PyObject *PyUnicode_DecodeUTF7Stateful(const char *s,
*p++ = outCh;
#endif
surrogate = 0;
continue;
}
else {
*p++ = surrogate;
surrogate = 0;
errmsg = "second surrogate missing";
goto utf7Error;
}
}
else if (outCh >= 0xD800 && outCh <= 0xDBFF) {
if (outCh >= 0xD800 && outCh <= 0xDBFF) {
/* first surrogate */
surrogate = outCh;
}
else if (outCh >= 0xDC00 && outCh <= 0xDFFF) {
errmsg = "unexpected second surrogate";
goto utf7Error;
}
else {
*p++ = outCh;
}
@ -2306,8 +2302,8 @@ PyObject *PyUnicode_DecodeUTF7Stateful(const char *s,
inShift = 0;
s++;
if (surrogate) {
errmsg = "second surrogate missing at end of shift sequence";
goto utf7Error;
*p++ = surrogate;
surrogate = 0;
}
if (base64bits > 0) { /* left-over bits */
if (base64bits >= 6) {