parser: Always perform UTF-8 validation

JSON is defined to be encoded using UTF-8, so we should not rely on
the documentation saying so when parsing, but validate the input and
eventually provide a recoverable parsing error.

1 files changed, 9 insertions, 0 deletions

diff --git a/json-glib/json-parser.c b/json-glib/json-parser.c
index 12c3ebf..edc13f0 100644
--- a/json-glib/json-parser.c
+++ b/json-glib/json-parser.c
@@ -878,6 +878,15 @@ json_parser_load (JsonParser   *parser,
 
   json_parser_clear (parser);
 
+  if (!g_utf8_validate (data, -1, NULL))
+    {
+      g_set_error_literal (error, JSON_PARSER_ERROR,
+                           JSON_PARSER_ERROR_INVALID_DATA,
+                           _("JSON data must be UTF-8 encoded"));
+      g_signal_emit (parser, parser_signals[ERROR], 0, *error);
+      return FALSE;
+    }
+
   scanner = json_scanner_create (parser);
   json_scanner_input_text (scanner, data, length);