bpo-36311: Fixes decoding multibyte characters around chunk boundarie… · python/cpython@f93c15a (original) (raw)

`@@ -3075,13 +3075,13 @@ def test_mbcs_alias(self):

3075

`self.assertEqual(codec.name, 'mbcs')

3076

3077

`@support.bigmemtest(size=2**31, memuse=7, dry_run=False)

3078

def test_large_input(self):

3078

def test_large_input(self, size):

3079

`# Test input longer than INT_MAX.

3080

`# Input should contain undecodable bytes before and after

3081

`# the INT_MAX limit.

3082

encoded = (b'01234567' * (2**28-1) +

3082

encoded = (b'01234567' * ((size//8)-1) +

3083

`b'\x85\x86\xea\xeb\xec\xef\xfc\xfd\xfe\xff')

3084

self.assertEqual(len(encoded), 2**31+2)

3084

self.assertEqual(len(encoded), size+2)

3085

`decoded = codecs.code_page_decode(932, encoded, 'surrogateescape', True)

3086

`self.assertEqual(decoded[1], len(encoded))

3087

`del encoded

`@@ -3092,6 +3092,20 @@ def test_large_input(self):

3092

`'\udc85\udc86\udcea\udceb\udcec'

3093

`'\udcef\udcfc\udcfd\udcfe\udcff')

3094

3095

@support.bigmemtest(size=2**31, memuse=6, dry_run=False)

3096

def test_large_utf8_input(self, size):

3097

Test input longer than INT_MAX.

3098

Input should contain a decodable multi-byte character

3099

surrounding INT_MAX

3100

encoded = (b'0123456\xed\x84\x80' * (size//8))

3101

self.assertEqual(len(encoded), size // 8 * 10)

3102

decoded = codecs.code_page_decode(65001, encoded, 'ignore', True)

3103

self.assertEqual(decoded[1], len(encoded))

3104

del encoded

3105

self.assertEqual(len(decoded[0]), size)

3106

self.assertEqual(decoded[0][:10], '0123456\ud10001')

3107

self.assertEqual(decoded[0][-11:], '56\ud1000123456\ud100')

3108

+

3095

3109

3096

3110

`class ASCIITest(unittest.TestCase):

3097

3111

`def test_encode(self):