My app structure is Win > Conformant > Naviframe

In the navframe I have added Box > Entry

When my entry component has focus the virtual keyboard appears (slides up).  And if I press the physical  back button the keyboard disappears (slides down) but the entry still has focus.  I need to respond to the keyboard dissappearing event but can’t find a signal that corresponds to this (since the entry still has focus).  I have tried listening to the following events:

evas_object_event_callback_add(entry, EVAS_CALLBACK_HIDE , keyboard_hidden_cb, data);
evas_object_event_callback_add(entry, EVAS_CALLBACK_FOCUS_OUT , keyboard_focusout_cb, data);
evas_object_smart_callback_add(entry, "unfocused", unfocused_cb, data);
evas_object_smart_callback_add(entry, "text,set,done", done_cb, data);
evas_object_smart_callback_add(conformant, "virtualkeypad,state,off", ckbdoff_cb, NULL);
evas_object_smart_callback_add(conformant, "virtualkeypad,size,changed", ckbdoff_cb, NULL);

But none of those are triggered when the keyboard disappears (some are triggered if I press BACK again and the entry disappears).  Can someone help – how can I respond to the keyboard dissappearing?  I know it’s possible because Samsung’s Reminder WIdget shows a keyboard and when BACK is pressed the widget changes views.

According to SOME documentation on the Tizen website the conformant should send a signal (virtualkeypad,state,off) when the keyboard dissappears but I can’t seem to trap that.   According to other documentation on the Tizen website (https://developer.tizen.org/ko/development/api-references/native-application?redirect=https://developer.tizen.org/dev-guide/2.3.1/org.tizen.native.wearable.apireference/group__Conformant.html) the virtualkeypad on/off signals do not exist for wearable, only a changed signal (which I can’t seem to trap either)..

I found a similar question from 4 years ago, that seemed to be unresolved:

https://developer.tizen.org/forums/native-application-development/handling-device-back-button?langswitch=en

This seems like a pretty major bug/oversight…but somehow Samsung has gotten around it.  Is this a documentation problem or a bug?  Hopefully this can be escalated.