[Python-checkins] bpo-43555: Report the column offset for invalid line continuation character (GH-24939) (#24975)

[pablogsal]

https://github.com/python/cpython/commit/994a519915bff4901abaa7476e2b91682dea619a
commit: 994a519915bff4901abaa7476e2b91682dea619a
branch: 3.9
author: Miss Islington (bot) <31488909+miss-islington at users.noreply.github.com>
committer: pablogsal <Pablogsal at gmail.com>
date: 2021-03-22T19:07:05Z
summary:

bpo-43555: Report the column offset for invalid line continuation character (GH-24939) (#24975)

(cherry picked from commit 96eeff516204b7cc751103fa33dcc665e387846e)

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

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

files:
A Misc/NEWS.d/next/Core and Builtins/2021-03-19-22-49-40.bpo-43555.ZmhYSA.rst
M Lib/test/test_syntax.py
M Parser/pegen/pegen.c
M Parser/pegen/pegen.h

diff --git a/Lib/test/test_syntax.py b/Lib/test/test_syntax.py
index f0c9c988e2441..650a03173a8f3 100644
--- a/Lib/test/test_syntax.py
+++ b/Lib/test/test_syntax.py
@@ -973,6 +973,11 @@ def func2():
 """
         self._check_error(code, "invalid syntax")
 
+    def test_invalid_line_continuation_error_position(self):
+        self._check_error(r"a = 3 \ 4",
+                          "unexpected character after line continuation character",
+                          lineno=1, offset=9)
+
     def test_invalid_line_continuation_left_recursive(self):
         # Check bpo-42218: SyntaxErrors following left-recursive rules
         # (t_primary_raw in this case) need to be tested explicitly
diff --git a/Misc/NEWS.d/next/Core and Builtins/2021-03-19-22-49-40.bpo-43555.ZmhYSA.rst b/Misc/NEWS.d/next/Core and Builtins/2021-03-19-22-49-40.bpo-43555.ZmhYSA.rst
new file mode 100644
index 0000000000000..55a2fe22aa8ab
--- /dev/null
+++ b/Misc/NEWS.d/next/Core and Builtins/2021-03-19-22-49-40.bpo-43555.ZmhYSA.rst	
@@ -0,0 +1,2 @@
+Report the column offset for :exc:`SyntaxError` for invalid line
+continuation characters. Patch by Pablo Galindo.
diff --git a/Parser/pegen/pegen.c b/Parser/pegen/pegen.c
index 4e742a5ec7175..0a26275b23e0d 100644
--- a/Parser/pegen/pegen.c
+++ b/Parser/pegen/pegen.c
@@ -313,6 +313,7 @@ tokenizer_error(Parser *p)
 
     const char *msg = NULL;
     PyObject* errtype = PyExc_SyntaxError;
+    Py_ssize_t col_offset = -1;
     switch (p->tok->done) {
         case E_TOKEN:
             msg = "invalid token";
@@ -346,16 +347,14 @@ tokenizer_error(Parser *p)
             msg = "too many levels of indentation";
             break;
         case E_LINECONT:
+            col_offset = strlen(strtok(p->tok->buf, "\n")) - 1;
             msg = "unexpected character after line continuation character";
             break;
         default:
             msg = "unknown parsing error";
     }
 
-    PyErr_Format(errtype, msg);
-    // There is no reliable column information for this error
-    PyErr_SyntaxLocationObject(p->tok->filename, p->tok->lineno, 0);
-
+    RAISE_ERROR_KNOWN_LOCATION(p, errtype, p->tok->lineno, col_offset, msg);
     return -1;
 }
 
diff --git a/Parser/pegen/pegen.h b/Parser/pegen/pegen.h
index a2f524a597d23..224c5cbea6349 100644
--- a/Parser/pegen/pegen.h
+++ b/Parser/pegen/pegen.h
@@ -138,8 +138,9 @@ void *_PyPegen_raise_error_known_location(Parser *p, PyObject *errtype,
 void *_PyPegen_dummy_name(Parser *p, ...);
 
 Py_LOCAL_INLINE(void *)
-RAISE_ERROR_KNOWN_LOCATION(Parser *p, PyObject *errtype, int lineno,
-                           int col_offset, const char *errmsg, ...)
+RAISE_ERROR_KNOWN_LOCATION(Parser *p, PyObject *errtype,
+                           Py_ssize_t lineno, Py_ssize_t col_offset,
+                           const char *errmsg, ...)
 {
     va_list va;
     va_start(va, errmsg);