import math
+import os
import unittest
import sys
import _ast
+import tempfile
import types
-from test import support
+from test import support, script_helper
class TestSpecifics(unittest.TestCase):
self.assertInvalidSingle('f()\nxy # blah\nblah()')
self.assertInvalidSingle('x = 5 # comment\nx = 6\n')
+ def test_particularly_evil_undecodable(self):
+ # Issue 24022
+ src = b'0000\x00\n00000000000\n\x00\n\x9e\n'
+ with tempfile.TemporaryDirectory() as tmpd:
+ fn = os.path.join(tmpd, "bad.py")
+ with open(fn, "wb") as fp:
+ fp.write(src)
+ res = script_helper.run_python_until_end(fn)[0]
+ self.assertIn(b"Non-UTF-8", res.err)
+
@support.cpython_only
def test_compiler_recursion_limit(self):
# Expected limit is sys.getrecursionlimit() * the scaling factor
Core and Builtins
-----------------
+- Issue #24022: Fix tokenizer crash when processing undecodable source code.
+
- Issue #23309: Avoid a deadlock at shutdown if a daemon thread is aborted
while it is holding a lock to a buffered I/O object, and the main thread
tries to use the same I/O object (typically stdout or stderr). A fatal
{
PyObject *s;
int result;
+ if (tok->decoding_erred)
+ return 0;
s = PyUnicode_DecodeUTF8(tok->start, tok->cur - tok->start, NULL);
if (s == NULL || PyUnicode_READY(s) == -1) {
if (PyErr_ExceptionMatches(PyExc_UnicodeDecodeError)) {
c = tok_nextc(tok);
}
tok_backup(tok, c);
- if (nonascii &&
- !verify_identifier(tok)) {
- tok->done = E_IDENTIFIER;
+ if (nonascii && !verify_identifier(tok))
return ERRORTOKEN;
- }
*p_start = tok->start;
*p_end = tok->cur;
return NAME;