Issue #8941: decoding big endian UTF-32 data in UCS-2 builds could crash
the interpreter with characters outside the Basic Multilingual Plane
(higher than 0x10000).
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index 254bc4d..4c020c3 100644
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -315,6 +315,16 @@
self.assertRaises(UnicodeDecodeError, codecs.utf_32_decode,
"\xff", "strict", True)
+ def test_issue8941(self):
+ # Issue #8941: insufficient result allocation when decoding into
+ # surrogate pairs on UCS-2 builds.
+ encoded_le = '\xff\xfe\x00\x00' + '\x00\x00\x01\x00' * 1024
+ self.assertEqual(u'\U00010000' * 1024,
+ codecs.utf_32_decode(encoded_le)[0])
+ encoded_be = '\x00\x00\xfe\xff' + '\x00\x01\x00\x00' * 1024
+ self.assertEqual(u'\U00010000' * 1024,
+ codecs.utf_32_decode(encoded_be)[0])
+
class UTF32LETest(ReadTest):
encoding = "utf-32-le"
@@ -348,6 +358,13 @@
self.assertRaises(UnicodeDecodeError, codecs.utf_32_le_decode,
"\xff", "strict", True)
+ def test_issue8941(self):
+ # Issue #8941: insufficient result allocation when decoding into
+ # surrogate pairs on UCS-2 builds.
+ encoded = '\x00\x00\x01\x00' * 1024
+ self.assertEqual(u'\U00010000' * 1024,
+ codecs.utf_32_le_decode(encoded)[0])
+
class UTF32BETest(ReadTest):
encoding = "utf-32-be"
@@ -381,6 +398,14 @@
self.assertRaises(UnicodeDecodeError, codecs.utf_32_be_decode,
"\xff", "strict", True)
+ def test_issue8941(self):
+ # Issue #8941: insufficient result allocation when decoding into
+ # surrogate pairs on UCS-2 builds.
+ encoded = '\x00\x01\x00\x00' * 1024
+ self.assertEqual(u'\U00010000' * 1024,
+ codecs.utf_32_be_decode(encoded)[0])
+
+
class UTF16Test(ReadTest):
encoding = "utf-16"