libtracker-extract: Add tracker_guarantee_resource_utf8_string()

This is a wrapper to tracker_resource_set_string(), which additionally
validates the string to be UTF-8, and inserts only as much text as
could be validated.

Can be thought of as an internal replacement to
tracker_sparql_builder_object_unvalidated(), the fact that we deal with
non utf-8 data sources is an implementation detail to extractors, so
it doesn't make as much sense to make this a public function in the
shiny new TrackerResource API.

2 files changed, 25 insertions, 0 deletions

diff --git a/src/libtracker-extract/tracker-guarantee.c b/src/libtracker-extract/tracker-guarantee.c
index 9b8460a08..5c595c9ad 100644
--- a/src/libtracker-extract/tracker-guarantee.c
+++ b/src/libtracker-extract/tracker-guarantee.c
@@ -348,3 +348,25 @@ tracker_guarantee_resource_date_from_file_mtime (TrackerResource *resource,
 
 	return success;
 }
+
+gboolean
+tracker_guarantee_resource_utf8_string (TrackerResource *resource,
+                                        const gchar     *key,
+                                        const gchar     *value)
+{
+	const gchar *end;
+	gchar *str;
+
+	if (!g_utf8_validate (value, -1, &end)) {
+		if (end == value)
+			return FALSE;
+
+		str = g_strndup (value, end - value);
+		tracker_resource_set_string (resource, key, str);
+		g_free (str);
+	} else {
+		tracker_resource_set_string (resource, key, value);
+	}
+
+	return TRUE;
+}
diff --git a/src/libtracker-extract/tracker-guarantee.h b/src/libtracker-extract/tracker-guarantee.h
index 9b58a634a..cdb28c24a 100644
--- a/src/libtracker-extract/tracker-guarantee.h
+++ b/src/libtracker-extract/tracker-guarantee.h
@@ -49,6 +49,9 @@ gboolean tracker_guarantee_resource_date_from_file_mtime (TrackerResource  *reso
                                                           const gchar      *key,
                                                           const gchar      *current_value,
                                                           const gchar      *uri);
+gboolean tracker_guarantee_resource_utf8_string (TrackerResource *resource,
+                                                 const gchar     *key,
+                                                 const gchar     *value);
 
 G_END_DECLS