Issue #6922: Fix an infinite loop when trying to decode an invalid
UTF-32 stream with a non-raising error handler like "replace" or "ignore".
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index 06cab1c..1893311 100644
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -305,6 +305,12 @@
]
)
+ def test_handlers(self):
+ self.assertEqual((u'\ufffd', 1),
+ codecs.utf_32_decode('\x01', 'replace', True))
+ self.assertEqual((u'', 1),
+ codecs.utf_32_decode('\x01', 'ignore', True))
+
def test_errors(self):
self.assertRaises(UnicodeDecodeError, codecs.utf_32_decode,
"\xff", "strict", True)
@@ -422,6 +428,12 @@
]
)
+ def test_handlers(self):
+ self.assertEqual((u'\ufffd', 1),
+ codecs.utf_16_decode('\x01', 'replace', True))
+ self.assertEqual((u'', 1),
+ codecs.utf_16_decode('\x01', 'ignore', True))
+
def test_errors(self):
self.assertRaises(UnicodeDecodeError, codecs.utf_16_decode, "\xff", "strict", True)
diff --git a/Misc/NEWS b/Misc/NEWS
index 7796957..dc37a37 100644
--- a/Misc/NEWS
+++ b/Misc/NEWS
@@ -12,6 +12,9 @@
Core and Builtins
-----------------
+- Issue #6922: Fix an infinite loop when trying to decode an invalid
+ UTF-32 stream with a non-raising error handler like "replace" or "ignore".
+
- Issue #6713: Improve performance of integer -> string conversions.
- Issue #1590864: Fix potential deadlock when mixing threads and fork().
diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c
index c4b4902..6164510 100644
--- a/Objects/unicodeobject.c
+++ b/Objects/unicodeobject.c
@@ -2321,7 +2321,7 @@
if (unicode_decode_call_errorhandler(
errors, &errorHandler,
"utf32", errmsg,
- starts, size, &startinpos, &endinpos, &exc, &s,
+ starts, size, &startinpos, &endinpos, &exc, (const char **)&q,
&unicode, &outpos, &p))
goto onError;
}