gh-97556: Raise null bytes syntax error upon null in multiline string (GH-104136)

diff --git a/Lib/test/test_cmd_line_script.py b/Lib/test/test_cmd_line_script.py
index d98e23855e0c194077ba9628ddb337893620dc01..8bf299382e9ca4ee80d72baacde5d8e42019193c 100644 (file)
--- a/Lib/test/test_cmd_line_script.py
+++ b/Lib/test/test_cmd_line_script.py
@@ -669,6 +669,19 @@ class CmdLineTest(unittest.TestCase):
                 ],
             )
 
+    def test_syntaxerror_null_bytes_in_multiline_string(self):
+        scripts = ["\n'''\nmultilinestring\0\n'''", "\nf'''\nmultilinestring\0\n'''"] # Both normal and f-strings
+        with os_helper.temp_dir() as script_dir:
+            for script in scripts:
+                script_name = _make_test_script(script_dir, 'script', script)
+                _, _, stderr = assert_python_failure(script_name)
+                self.assertEqual(
+                    stderr.splitlines()[-2:],
+                    [   b"    multilinestring",
+                        b'SyntaxError: source code cannot contain null bytes'
+                    ]
+                )
+
     def test_consistent_sys_path_for_direct_execution(self):
         # This test case ensures that the following all give the same
         # sys.path configuration:

diff --git a/Parser/tokenizer.c b/Parser/tokenizer.c
index d2f9fee110ebf5595e9de51a8454b3b692c504a3..7c07d2011fda616a4965384a06e0f451b27279ff 100644 (file)
--- a/Parser/tokenizer.c
+++ b/Parser/tokenizer.c
@@ -2301,8 +2301,12 @@ tok_get_normal_mode(struct tok_state *tok, tokenizer_mode* current_tok, struct t
         /* Get rest of string */
         while (end_quote_size != quote_size) {
             c = tok_nextc(tok);
-            if (tok->done == E_DECODE)
+            if (tok->done == E_ERROR) {
+                return MAKE_TOKEN(ERRORTOKEN);
+            }
+            if (tok->done == E_DECODE) {
                 break;
+            }
             if (c == EOF || (quote_size == 1 && c == '\n')) {
                 assert(tok->multi_line_start != NULL);
                 // shift the tok_state's location into
@@ -2554,6 +2558,9 @@ f_string_middle:
 
     while (end_quote_size != current_tok->f_string_quote_size) {
         int c = tok_nextc(tok);
+        if (tok->done == E_ERROR) {
+            return MAKE_TOKEN(ERRORTOKEN);
+        }
         if (c == EOF || (current_tok->f_string_quote_size == 1 && c == '\n')) {
             if (tok->decoding_erred) {
                 return MAKE_TOKEN(ERRORTOKEN);