bpo-45408: Don't override previous tokenizer errors in the second parser pass (GH...

diff --git a/Lib/test/test_ast.py b/Lib/test/test_ast.py
index 1f4257b1191baa078d4acb5a0eebf45ca325942c..e630677f796845312a462f1014060c24f50fe785 100644 (file)
--- a/Lib/test/test_ast.py
+++ b/Lib/test/test_ast.py
@@ -1075,6 +1075,14 @@ Module(
         with self.assertRaisesRegex(ValueError, msg):
             ast.literal_eval(node)
 
+    def test_literal_eval_syntax_errors(self):
+        msg = "unexpected character after line continuation character"
+        with self.assertRaisesRegex(SyntaxError, msg):
+            ast.literal_eval(r'''
+                \
+                (\
+            \ ''')
+
     def test_bad_integer(self):
         # issue13436: Bad error message with invalid numeric values
         body = [ast.ImportFrom(module='time',

diff --git a/Lib/test/test_exceptions.py b/Lib/test/test_exceptions.py
index 85dc3c0f22081d32fc4b16f26ff16bd5acb8609e..02489a2bd92150f7dc0b62a324159506a6adc0a8 100644 (file)
--- a/Lib/test/test_exceptions.py
+++ b/Lib/test/test_exceptions.py
@@ -223,7 +223,7 @@ class ExceptionTests(unittest.TestCase):
         check('x = "a', 1, 5)
         check('lambda x: x = 2', 1, 1)
         check('f{a + b + c}', 1, 1)
-        check('[file for str(file) in []\n])', 2, 2)
+        check('[file for str(file) in []\n])', 1, 11)
         check('a = « hello » « world »', 1, 5)
         check('[\nfile\nfor str(file)\nin\n[]\n]', 3, 5)
         check('[file for\n str(file) in []]', 2, 2)

diff --git a/Misc/NEWS.d/next/Core and Builtins/2021-10-07-21-26-44.bpo-45408.qUqzcd.rst b/Misc/NEWS.d/next/Core and Builtins/2021-10-07-21-26-44.bpo-45408.qUqzcd.rst
new file mode 100644 (file)
index 0000000..e4d4db9
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2021-10-07-21-26-44.bpo-45408.qUqzcd.rst
@@ -0,0 +1,2 @@
+Fix a crash in the parser when reporting tokenizer errors that occur at the
+same time unclosed parentheses are detected. Patch by Pablo Galindo.

diff --git a/Parser/pegen.c b/Parser/pegen.c
index 7e2d37caae37a33f0c1a42a98f399d5e65b0bb0f..a9896356e5b58911763a0fc4536ee1e0b02f2cd7 100644 (file)
--- a/Parser/pegen.c
+++ b/Parser/pegen.c
@@ -1342,13 +1342,16 @@ _PyPegen_run_parser(Parser *p)
 {
     void *res = _PyPegen_parse(p);
     if (res == NULL) {
+        if (PyErr_Occurred() && !PyErr_ExceptionMatches(PyExc_SyntaxError)) {
+            return NULL;
+        }
         Token *last_token = p->tokens[p->fill - 1];
         reset_parser_state(p);
         _PyPegen_parse(p);
         if (PyErr_Occurred()) {
             // Prioritize tokenizer errors to custom syntax errors raised
             // on the second phase only if the errors come from the parser.
-            if (p->tok->done != E_ERROR && PyErr_ExceptionMatches(PyExc_SyntaxError)) {
+            if (p->tok->done == E_DONE && PyErr_ExceptionMatches(PyExc_SyntaxError)) {
                 _PyPegen_check_tokenizer_errors(p);
             }
             return NULL;