diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py index 9be8281ce5a..328a47b2e37 100644 --- a/Lib/test/test_codecs.py +++ b/Lib/test/test_codecs.py @@ -2197,6 +2197,18 @@ class CharmapTest(unittest.TestCase): ("", len(allbytes)) ) + self.assertRaisesRegex(TypeError, + "character mapping must be in range\\(0x110000\\)", + codecs.charmap_decode, + b"\x00\x01\x02", "strict", {0: "A", 1: 'Bb', 2: -2} + ) + + self.assertRaisesRegex(TypeError, + "character mapping must be in range\\(0x110000\\)", + codecs.charmap_decode, + b"\x00\x01\x02", "strict", {0: "A", 1: 'Bb', 2: 999999999} + ) + def test_decode_with_int2int_map(self): a = ord('a') b = ord('b') diff --git a/Misc/NEWS.d/next/Library/2020-10-17-23-17-18.bpo-42065.85BsRA.rst b/Misc/NEWS.d/next/Library/2020-10-17-23-17-18.bpo-42065.85BsRA.rst new file mode 100644 index 00000000000..83c86c0799e --- /dev/null +++ b/Misc/NEWS.d/next/Library/2020-10-17-23-17-18.bpo-42065.85BsRA.rst @@ -0,0 +1,3 @@ +Fix an incorrectly formatted error from :meth:`_codecs.charmap_decode` when +called with a mapped value outside the range of valid Unicode code points. +PR by Max Bernstein. diff --git a/Objects/unicodeobject.c b/Objects/unicodeobject.c index 01e5c728b38..c4e73ebd45d 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -8304,7 +8304,7 @@ charmap_decode_mapping(const char *s, goto Undefined; if (value < 0 || value > MAX_UNICODE) { PyErr_Format(PyExc_TypeError, - "character mapping must be in range(0x%lx)", + "character mapping must be in range(0x%x)", (unsigned long)MAX_UNICODE + 1); goto onError; }