Created attachment 271466 [details] [review]
power: handle WakeUpScreen from gnome-shell

This signal should be handled in the same way as resetting
IDLETIME, but we don't really reset the counter, so other
parts of the code (for example the automatic suspend) are
unaffected.

Created attachment 271470 [details] [review]
power: don't unblank the screen for notifications

The Shell already does that.

Review of attachment 271466 [details] [review]:

::: plugins/power/gsd-power-manager.c
@@ +1974,3 @@
+        if (manager->priv->temporary_unidle_on_ac_id != 0 &&
+            manager->priv->previous_idle_mode == mode)
+                return;

The second condition is already checked in idle_set_mode(). Any reason to not check the first one there too?

Review of attachment 271470 [details] [review]:

ok

(In reply to comment #3)
> Review of attachment 271466 [details] [review]:
> 
> ::: plugins/power/gsd-power-manager.c
> @@ +1974,3 @@
> +        if (manager->priv->temporary_unidle_on_ac_id != 0 &&
> +            manager->priv->previous_idle_mode == mode)
> +                return;
> 
> The second condition is already checked in idle_set_mode(). Any reason to not
> check the first one there too?

Mh? idle_set_mode() checks current_idle_mode, not previous_idle_mode

Comment on attachment 271466 [details] [review]
power: handle WakeUpScreen from gnome-shell

(In reply to comment #5)
> Mh? idle_set_mode() checks current_idle_mode, not previous_idle_mode

Oops, overlooked that, it all makes sense now ;-)

The function name is a bit unfortunate but it's a small helper and I can't think of anything better either.

Attachment 271466 [details] pushed as 468e142 - power: handle WakeUpScreen from gnome-shell
Attachment 271470 [details] pushed as fc2f698 - power: don't unblank the screen for notifications