Hi Daniel, On Mon, Dec 13, 2010 at 03:07:57PM +0100, Daniel Wagner wrote: > From: Daniel Wagner <[email protected]> > > --- > plugins/openvpn.c | 82 > +++++++++++++++++++++++++++++++++++++++++++++++++++++ > 1 files changed, 82 insertions(+), 0 deletions(-) > > diff --git a/plugins/openvpn.c b/plugins/openvpn.c > index 0a71f5d..9f8989d 100644 > --- a/plugins/openvpn.c > +++ b/plugins/openvpn.c > @@ -42,12 +42,74 @@ > > static DBusConnection *connection; > > +struct ov_route { > + char *host; > + char *netmask; > + char *gateway; > +}; Ok, so you're keeping a routing hash table within openvpn.c in order to protect your code against the route_* openvpn environment variables being sent unordered. Although I think this is very unlikely, I still belive that it makes sense. I would appreciate if you could just comment on that here, so that we could still easily maintain this code once you become BMW new CTO ;)
> @@ -115,6 +181,18 @@ static int ov_notify(DBusMessage *msg, struct > connman_provider *provider) > g_strfreev(options); > } > > + route = ov_route_lookup(key, "route_network_", routes); > + if (route != NULL) > + route->host = g_strdup(value); So here I'd like to keep the whole internal parsing and prefix stuff into the actual lookup routine. It would be renamed ov_append_route(key, routes), and would check if key has a valid prefix, and if so do its magic. Cheers, Samuel. -- Intel Open Source Technology Centre http://oss.intel.com/ _______________________________________________ connman mailing list [email protected] http://lists.connman.net/listinfo/connman
