gh-99581: Fix a buffer overflow in the tokenizer when copying lines that fill the available buffer (GH-99605)

(cherry picked from commit e13d1d9dda8c27691180bc618bd5e9bf43dfa89f)

Co-authored-by: Pablo Galindo Salgado <Pablogsal@gmail.com>

1 files changed, 16 insertions, 0 deletions

diff --git a/Lib/test/test_tokenize.py b/Lib/test/test_tokenize.py
index 47f2c06..63c2501 100644
--- a/Lib/test/test_tokenize.py
+++ b/Lib/test/test_tokenize.py
@@ -10,6 +10,8 @@ from textwrap import dedent
 from unittest import TestCase, mock
 from test.test_grammar import (VALID_UNDERSCORE_LITERALS,
                                INVALID_UNDERSCORE_LITERALS)
+from test.support import os_helper
+from test.support.script_helper import run_test_script, make_script
 import os
 import token
 
@@ -2631,5 +2633,19 @@ async def f():
         self.assertEqual(get_tokens(code), get_tokens(code_no_cont))
 
 
+class CTokenizerBufferTests(unittest.TestCase):
+    def test_newline_at_the_end_of_buffer(self):
+        # See issue 99581: Make sure that if we need to add a new line at the
+        # end of the buffer, we have enough space in the buffer, specially when
+        # the current line is as long as the buffer space available.
+        test_script = f"""\
+        #coding: latin-1
+        #{"a"*10000}
+        #{"a"*10002}"""
+        with os_helper.temp_dir() as temp_dir:
+            file_name = make_script(temp_dir, 'foo', test_script)
+            run_test_script(file_name)
+
+
 if __name__ == "__main__":
     unittest.main()