[totem-pl-parser] plparser: Print illegal char byte sequence info during utf-8 conversion error

[Bastien Nocera <hadess src gnome org>]

commit e84cf43c7080774f5148a82e33118ad3f902a003
Author: crvi <crvisqr gmail com>
Date:   Thu Mar 18 19:03:47 2021 +0530

    plparser: Print illegal char byte sequence info during utf-8 conversion error
    
    This is useful for debugging invalid XML documents. Provides the
    following debug information.
    
    Invalid byte sequence in conversion input: byte offset 22493, byte:
    '\xe2', byte context: 'ience from Valentine\xe2'

 plparse/totem-pl-parser.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)
---
diff --git a/plparse/totem-pl-parser.c b/plparse/totem-pl-parser.c
index 534d268..4656f6e 100644
--- a/plparse/totem-pl-parser.c
+++ b/plparse/totem-pl-parser.c
@@ -159,6 +159,7 @@
 
 #define READ_CHUNK_SIZE 8192
 #define RECURSE_LEVEL_MAX 4
+#define ILLEGAL_CONTEXT_LENGTH 20
 
 #define D(x) if (debug) x
 
@@ -1885,7 +1886,7 @@ totem_pl_parser_parse_xml_relaxed (char *contents,
        g_autoptr(GError) error = NULL;
        g_autofree char *encoding = NULL;
        g_autofree char *new_contents = NULL;
-       gsize new_size;
+       gsize new_size, bytes_read;
        xml_parser_t *xml_parser;
 
        totem_pl_parser_cleanup_xml (contents);
@@ -1921,10 +1922,23 @@ totem_pl_parser_parse_xml_relaxed (char *contents,
 
        xml_parser_free_tree (doc);
 
-       new_contents = g_convert (contents, size, "UTF-8", encoding, NULL, &new_size, &error);
+       new_contents = g_convert (contents, size, "UTF-8", encoding, &bytes_read, &new_size, &error);
        if (new_contents == NULL) {
-               g_warning ("Failed to convert XML data from '%s' to '%s': %s",
-                          encoding, "UTF-8", error->message);
+               g_autofree char *message = NULL;
+               message = g_strdup_printf ("Failed to convert XML data from '%s' to '%s': %s",
+                                          encoding, "UTF-8", error->message);
+
+               if (error->code == G_CONVERT_ERROR_ILLEGAL_SEQUENCE) {
+                       int context_length = MIN (bytes_read, ILLEGAL_CONTEXT_LENGTH);
+
+                       g_warning ("%s: byte offset %" G_GSIZE_FORMAT ", byte: '%.1s', byte context: '%.*s'",
+                                  message, bytes_read, contents + bytes_read,
+                                  context_length + 1,
+                                  contents + bytes_read - context_length);
+               } else {
+                       g_warning ("%s", message);
+               }
+
                return NULL;
        }