bpo-40334: Fix error location upon parsing an invalid string literal (GH-19962)
When parsing a string with an invalid escape, the old parser used to
point to the beginning of the invalid string. This commit changes the new
parser to match that behaviour, since it's currently pointing to the
end of the string (or to be more precise, to the beginning of the next
token).
diff --git a/Parser/pegen/pegen.c b/Parser/pegen/pegen.c
index c311593..06af53b 100644
--- a/Parser/pegen/pegen.c
+++ b/Parser/pegen/pegen.c
@@ -383,7 +383,7 @@
PyObject *errstr = NULL;
PyObject *loc = NULL;
PyObject *tmp = NULL;
- Token *t = p->tokens[p->fill - 1];
+ Token *t = p->known_err_token != NULL ? p->known_err_token : p->tokens[p->fill - 1];
Py_ssize_t col_number = !with_col_number;
va_list va;
p->error_indicator = 1;
@@ -1053,6 +1053,7 @@
p->starting_col_offset = 0;
p->flags = flags;
p->feature_version = feature_version;
+ p->known_err_token = NULL;
return p;
}
@@ -1972,12 +1973,7 @@
const char *fstr;
Py_ssize_t fstrlen = -1;
- char *this_str = PyBytes_AsString(t->bytes);
- if (!this_str) {
- goto error;
- }
-
- if (_PyPegen_parsestr(p, this_str, &this_bytesmode, &this_rawmode, &s, &fstr, &fstrlen) != 0) {
+ if (_PyPegen_parsestr(p, &this_bytesmode, &this_rawmode, &s, &fstr, &fstrlen, t) != 0) {
goto error;
}