diff --git a/Lib/test/test_pyexpat.py b/Lib/test/test_pyexpat.py index aaa91aca36e3c4..0b95547b63edcf 100644 --- a/Lib/test/test_pyexpat.py +++ b/Lib/test/test_pyexpat.py @@ -712,6 +712,19 @@ def test_change_size_2(self): parser.Parse(xml2, True) self.assertEqual(self.n, 4) + @support.requires_resource('cpu') + @support.requires_resource('walltime') + def test_large_character_data_no_buffer_overflow(self): + # See https://github.com/python/cpython/issues/148441 + parser = expat.ParserCreate() + parser.buffer_text = True + parser.buffer_size = 2**31 - 1 # INT_MAX + N = 2049 * (1 << 20) - 3 # Character data greater than INT_MAX + self.assertGreater(N, parser.buffer_size) + parser.CharacterDataHandler = lambda text: None + xml_data = b"" + b"A" * N + b"" + self.assertEqual(parser.Parse(xml_data, True), 1) + class ElementDeclHandlerTest(unittest.TestCase): def test_trigger_leak(self): # Unfixed, this test would leak the memory of the so-called diff --git a/Misc/NEWS.d/next/Library/2026-04-23-12-50-15.gh-issue-148441.zvpCkR.rst b/Misc/NEWS.d/next/Library/2026-04-23-12-50-15.gh-issue-148441.zvpCkR.rst new file mode 100644 index 00000000000000..f51fbdcd99ac9b --- /dev/null +++ b/Misc/NEWS.d/next/Library/2026-04-23-12-50-15.gh-issue-148441.zvpCkR.rst @@ -0,0 +1,4 @@ +:mod:`xml.parsers.expat`: Fix a heap buffer overflow in +:meth:`~xml.parsers.expat.xmlparser.CharacterDataHandler` +when the character data size exceeds the parser's +:attr:`buffer size `. diff --git a/Modules/pyexpat.c b/Modules/pyexpat.c index 0f0afe17513ef1..c01f7babe74527 100644 --- a/Modules/pyexpat.c +++ b/Modules/pyexpat.c @@ -393,7 +393,7 @@ my_CharacterDataHandler(void *userData, const XML_Char *data, int len) if (self->buffer == NULL) call_character_handler(self, data, len); else { - if ((self->buffer_used + len) > self->buffer_size) { + if (len > (self->buffer_size - self->buffer_used)) { if (flush_character_buffer(self) < 0) return; /* handler might have changed; drop the rest on the floor