Issue #15379: Fix passing of non-BMP characters as integers for the charmap decoder (already working as unicode strings).
Patch by Serhiy Storchaka.
diff --git a/Lib/test/test_codeccallbacks.py b/Lib/test/test_codeccallbacks.py
index c5b1e25..e656d2f 100644
--- a/Lib/test/test_codeccallbacks.py
+++ b/Lib/test/test_codeccallbacks.py
@@ -744,7 +744,7 @@
raise ValueError
self.assertRaises(UnicodeError, codecs.charmap_decode, b"\xff", "strict", {0xff: None})
self.assertRaises(ValueError, codecs.charmap_decode, b"\xff", "strict", D())
- self.assertRaises(TypeError, codecs.charmap_decode, b"\xff", "strict", {0xff: sys.maxunicode+1})
+ self.assertRaises(TypeError, codecs.charmap_decode, b"\xff", "strict", {0xff: 0x110000})
def test_encodehelper(self):
# enhance coverage of:
diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py
index 3426a4d..f342d88 100644
--- a/Lib/test/test_codecs.py
+++ b/Lib/test/test_codecs.py
@@ -1546,6 +1546,10 @@
("abc", 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)
@@ -1572,6 +1576,107 @@
("", 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.assertRaises(TypeError,
+ codecs.charmap_decode, b"\x00\x01\x02", "strict",
+ {0: 0x110000, 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")