I also got a compiler warning about the following: diff --git i/libnm-util/nm-setting.c w/libnm-util/nm-setting.c index 1c7e4c9..6247acd 100644 --- i/libnm-util/nm-setting.c +++ w/libnm-util/nm-setting.c @@ -1171,7 +1171,7 @@ nm_setting_get_virtual_iface_name (NMSetting *setting) gboolean nm_setting_set_virtual_iface_name (NMSetting *setting, const char *iface_name) { - g_return_val_if_fail (NM_IS_SETTING (setting), NULL); + g_return_val_if_fail (NM_IS_SETTING (setting), FALSE); if (NM_SETTING_GET_CLASS (setting)->set_virtual_iface_name) return NM_SETTING_GET_CLASS (setting)->set_virtual_iface_name (setting, iface_name); On Wed, 2013-11-06 at 13:17 +0100, Thomas Haller wrote:
And in the following 3 patches, I think set_virtual_iface_name should return FALSE, if it did not change anything. So basically I would add if (!g_strcmp0 (priv->interface_name, iface_name)) return FALSE;
I am not sure about this ^^^. Probably it's better always to return TRUE(?).
Attachment:
signature.asc
Description: This is a digitally signed message part