Commit d73aca76 authored by Benjamin Peterson's avatar Benjamin Peterson

do not call into python api if an exception is set (#24022)

parent 6de708fd
import math import math
import os
import unittest import unittest
import sys import sys
import _ast import _ast
import tempfile
import types import types
from test import support from test import support, script_helper
class TestSpecifics(unittest.TestCase): class TestSpecifics(unittest.TestCase):
...@@ -492,6 +494,16 @@ if 1: ...@@ -492,6 +494,16 @@ if 1:
self.assertInvalidSingle('f()\nxy # blah\nblah()') self.assertInvalidSingle('f()\nxy # blah\nblah()')
self.assertInvalidSingle('x = 5 # comment\nx = 6\n') 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 @support.cpython_only
def test_compiler_recursion_limit(self): def test_compiler_recursion_limit(self):
# Expected limit is sys.getrecursionlimit() * the scaling factor # Expected limit is sys.getrecursionlimit() * the scaling factor
......
...@@ -10,6 +10,8 @@ Release date: tba ...@@ -10,6 +10,8 @@ Release date: tba
Core and Builtins 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 - 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 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 tries to use the same I/O object (typically stdout or stderr). A fatal
......
...@@ -1301,6 +1301,8 @@ verify_identifier(struct tok_state *tok) ...@@ -1301,6 +1301,8 @@ verify_identifier(struct tok_state *tok)
{ {
PyObject *s; PyObject *s;
int result; int result;
if (tok->decoding_erred)
return 0;
s = PyUnicode_DecodeUTF8(tok->start, tok->cur - tok->start, NULL); s = PyUnicode_DecodeUTF8(tok->start, tok->cur - tok->start, NULL);
if (s == NULL || PyUnicode_READY(s) == -1) { if (s == NULL || PyUnicode_READY(s) == -1) {
if (PyErr_ExceptionMatches(PyExc_UnicodeDecodeError)) { if (PyErr_ExceptionMatches(PyExc_UnicodeDecodeError)) {
...@@ -1469,11 +1471,8 @@ tok_get(struct tok_state *tok, char **p_start, char **p_end) ...@@ -1469,11 +1471,8 @@ tok_get(struct tok_state *tok, char **p_start, char **p_end)
c = tok_nextc(tok); c = tok_nextc(tok);
} }
tok_backup(tok, c); tok_backup(tok, c);
if (nonascii && if (nonascii && !verify_identifier(tok))
!verify_identifier(tok)) {
tok->done = E_IDENTIFIER;
return ERRORTOKEN; return ERRORTOKEN;
}
*p_start = tok->start; *p_start = tok->start;
*p_end = tok->cur; *p_end = tok->cur;
return NAME; return NAME;
......
Markdown is supported
0%
or
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment