gh-112387: Fix error positions for decoded strings with backwards tokenize errors (#112409)

pablogsal · web-flow · commit 45d648597b11 · 2023-11-27T18:37:48.000Z
Signed-off-by: Pablo Galindo &lt;pablogsal@gmail.com&gt;
diff --git a/Lib/test/test_syntax.py b/Lib/test/test_syntax.py
@@ -2334,6 +2334,10 @@ def test_error_parenthesis(self):
 """
         self._check_error(code, "parenthesis '\\)' does not match opening parenthesis '\\['")
 
+        # Examples with dencodings
+        s = b'# coding=latin\n(aaaaaaaaaaaaaaaaa\naaaaaaaaaaa\xb5'
+        self._check_error(s, "'\(' was never closed")
+
     def test_error_string_literal(self):
 
         self._check_error("'blech", r"unterminated string literal \(.*\)$")
diff --git a/Misc/NEWS.d/next/Core and Builtins/2023-11-25-22-39-44.gh-issue-112387.AbBq5W.rst b/Misc/NEWS.d/next/Core and Builtins/2023-11-25-22-39-44.gh-issue-112387.AbBq5W.rst
@@ -0,0 +1,2 @@
+Fix error positions for decoded strings with backwards tokenize errors.
+Patch by Pablo Galindo
diff --git a/Parser/pegen_errors.c b/Parser/pegen_errors.c
@@ -282,6 +282,10 @@ get_error_line_from_tokenizer_buffers(Parser *p, Py_ssize_t lineno)
     Py_ssize_t relative_lineno = p->starting_lineno ? lineno - p->starting_lineno + 1 : lineno;
     const char* buf_end = p->tok->fp_interactive ? p->tok->interactive_src_end : p->tok->inp;
 
+    if (buf_end < cur_line) {
+        buf_end = cur_line + strlen(cur_line);
+    }
+
     for (int i = 0; i < relative_lineno - 1; i++) {
         char *new_line = strchr(cur_line, '\n');
         // The assert is here for debug builds but the conditional that

Original file line number	Diff line number	Diff line change
`@@ -0,0 +1,2 @@`
	`1`	`+Fix error positions for decoded strings with backwards tokenize errors.`
	`2`	`+Patch by Pablo Galindo`