Patch #1031213: Decode source line in SyntaxErrors back to its original
source encoding. Will backport to 2.5.
diff --git a/Parser/parsetok.c b/Parser/parsetok.c
index c951396..f3d8462 100644
--- a/Parser/parsetok.c
+++ b/Parser/parsetok.c
@@ -218,16 +218,24 @@
err_ret->error = E_EOF;
err_ret->lineno = tok->lineno;
if (tok->buf != NULL) {
+ char *text = NULL;
size_t len;
assert(tok->cur - tok->buf < INT_MAX);
err_ret->offset = (int)(tok->cur - tok->buf);
len = tok->inp - tok->buf;
- err_ret->text = (char *) PyObject_MALLOC(len + 1);
- if (err_ret->text != NULL) {
- if (len > 0)
- strncpy(err_ret->text, tok->buf, len);
- err_ret->text[len] = '\0';
+#ifdef Py_USING_UNICODE
+ text = PyTokenizer_RestoreEncoding(tok, len, &err_ret->offset);
+
+#endif
+ if (text == NULL) {
+ text = (char *) PyObject_MALLOC(len + 1);
+ if (text != NULL) {
+ if (len > 0)
+ strncpy(text, tok->buf, len);
+ text[len] = '\0';
+ }
}
+ err_ret->text = text;
}
} else if (tok->encoding != NULL) {
node* r = PyNode_New(encoding_decl);