From 400175ad91c1aa8abc6dfce108446e4210d0c157 Mon Sep 17 00:00:00 2001 From: "Miss Skeleton (bot)" <31488909+miss-islington@users.noreply.github.com> Date: Sat, 17 Oct 2020 23:16:52 -0700 Subject: [PATCH] bpo-42065: Fix incorrectly formatted _codecs.charmap_decode error message (GH-19940) (cherry picked from commit 3635388f52b42e5280229104747962117104c453) Co-authored-by: Max Bernstein --- Lib/test/test_codecs.py | 12 ++++++++++++ .../Library/2020-10-17-23-17-18.bpo-42065.85BsRA.rst | 3 +++ Objects/unicodeobject.c | 2 +- 3 files changed, 16 insertions(+), 1 deletion(-) create mode 100644 Misc/NEWS.d/next/Library/2020-10-17-23-17-18.bpo-42065.85BsRA.rst diff --git a/Lib/test/test_codecs.py b/Lib/test/test_codecs.py index 0fd258ca8ab..d1faf0126c1 100644 --- a/Lib/test/test_codecs.py +++ b/Lib/test/test_codecs.py @@ -2183,6 +2183,18 @@ 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 01df54da005..a038e0d4c06 100644 --- a/Objects/unicodeobject.c +++ b/Objects/unicodeobject.c @@ -7998,7 +7998,7 @@ 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; }