mirror of
				https://github.com/python/cpython.git
				synced 2025-11-03 23:21:29 +00:00 
			
		
		
		
	Issue #15379: Fix passing of non-BMP characters as integers for the charmap decoder (already working as unicode strings).
Patch by Serhiy Storchaka.
This commit is contained in:
		
						commit
						a1f7655fa7
					
				
					 3 changed files with 123 additions and 3 deletions
				
			
		| 
						 | 
				
			
			@ -1692,6 +1692,15 @@ def test_decode_with_string_map(self):
 | 
			
		|||
            ("abc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict", "\U0010FFFFbc"),
 | 
			
		||||
            ("\U0010FFFFbc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertRaises(UnicodeDecodeError,
 | 
			
		||||
            codecs.charmap_decode, b"\x00\x01\x02", "strict", "ab"
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "replace", "ab"),
 | 
			
		||||
            ("ab\ufffd", 3)
 | 
			
		||||
| 
						 | 
				
			
			@ -1718,6 +1727,113 @@ def test_decode_with_string_map(self):
 | 
			
		|||
            ("", len(allbytes))
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
    def test_decode_with_int2str_map(self):
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: 'a', 1: 'b', 2: 'c'}),
 | 
			
		||||
            ("abc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: 'Aa', 1: 'Bb', 2: 'Cc'}),
 | 
			
		||||
            ("AaBbCc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: '\U0010FFFF', 1: 'b', 2: 'c'}),
 | 
			
		||||
            ("\U0010FFFFbc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: 'a', 1: 'b', 2: ''}),
 | 
			
		||||
            ("ab", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertRaises(UnicodeDecodeError,
 | 
			
		||||
            codecs.charmap_decode, b"\x00\x01\x02", "strict",
 | 
			
		||||
                                   {0: 'a', 1: 'b'}
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "replace",
 | 
			
		||||
                                  {0: 'a', 1: 'b'}),
 | 
			
		||||
            ("ab\ufffd", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "replace",
 | 
			
		||||
                                  {0: 'a', 1: 'b', 2: None}),
 | 
			
		||||
            ("ab\ufffd", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
 | 
			
		||||
                                  {0: 'a', 1: 'b'}),
 | 
			
		||||
            ("ab", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
 | 
			
		||||
                                  {0: 'a', 1: 'b', 2: None}),
 | 
			
		||||
            ("ab", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        allbytes = bytes(range(256))
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(allbytes, "ignore", {}),
 | 
			
		||||
            ("", len(allbytes))
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
    def test_decode_with_int2int_map(self):
 | 
			
		||||
        a = ord('a')
 | 
			
		||||
        b = ord('b')
 | 
			
		||||
        c = ord('c')
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: a, 1: b, 2: c}),
 | 
			
		||||
            ("abc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        # Issue #15379
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: 0x10FFFF, 1: b, 2: c}),
 | 
			
		||||
            ("\U0010FFFFbc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "strict",
 | 
			
		||||
                                  {0: sys.maxunicode, 1: b, 2: c}),
 | 
			
		||||
            (chr(sys.maxunicode) + "bc", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertRaises(TypeError,
 | 
			
		||||
            codecs.charmap_decode, b"\x00\x01\x02", "strict",
 | 
			
		||||
                                   {0: sys.maxunicode + 1, 1: b, 2: c}
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertRaises(UnicodeDecodeError,
 | 
			
		||||
            codecs.charmap_decode, b"\x00\x01\x02", "strict",
 | 
			
		||||
                                   {0: a, 1: b},
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "replace",
 | 
			
		||||
                                  {0: a, 1: b}),
 | 
			
		||||
            ("ab\ufffd", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
        self.assertEqual(
 | 
			
		||||
            codecs.charmap_decode(b"\x00\x01\x02", "ignore",
 | 
			
		||||
                                  {0: a, 1: b}),
 | 
			
		||||
            ("ab", 3)
 | 
			
		||||
        )
 | 
			
		||||
 | 
			
		||||
 | 
			
		||||
class WithStmtTest(unittest.TestCase):
 | 
			
		||||
    def test_encodedfile(self):
 | 
			
		||||
        f = io.BytesIO(b"\xc3\xbc")
 | 
			
		||||
| 
						 | 
				
			
			
 | 
			
		|||
		Loading…
	
	Add table
		Add a link
		
	
		Reference in a new issue