[PATCH 5/6] base-bearer: setup periodic stats loading

[Dan Williams]

On Mon, 2015-11-30 at 23:25 +0100, Aleksander Morgado wrote:
> On Mon, Nov 30, 2015 at 5:33 PM, Dan Williams <dcbw at redhat.com>
> wrote:
> > > +static gboolean
> > > +stats_update_cb (MMBaseBearer *self)
> > > +{
> > > +    /* If the implementation knows how to update stat values,
> > > run it
> > > */
> > > +    if (MM_BASE_BEARER_GET_CLASS (self)->reload_stats &&
> > > +        MM_BASE_BEARER_GET_CLASS (self)->reload_stats_finish) {
> > > +        MM_BASE_BEARER_GET_CLASS (self)->reload_stats (
> > > +            self,
> > > +            (GAsyncReadyCallback)reload_stats_ready,
> > > +            NULL);
> > > +        return TRUE;
> > > +    }
> > > +
> > > +    /* Otherwise, just update duration and we're done */
> > > +    mm_bearer_stats_set_duration (self->priv->stats, (guint32)
> > > g_timer_elapsed (self->priv->duration_timer, NULL));
> > > +    mm_bearer_stats_set_bytes_tx (self->priv->stats, 0);
> > > +    mm_bearer_stats_set_bytes_rx (self->priv->stats, 0);
> > > +    bearer_update_interface_stats (self);
> > > +    return TRUE;
> > > 
> > I'm a fan of G_SOURCE_CONTINUE/G_SOURCE_REMOVE these days, but
> > that's
> > personal preference.
> 
> Ah! I had to fire up devhelp to see what those were :). Yeah, I guess
> the names are a bit easier to remember than the true/false logic. How
> about a follow up patch changing the TRUE/FALSE values in all
> GSourceFuncs everywhere in the code?

Works for me.

Dan