I’m currently using the bridge api to unlatch the front. But I only want to do this when the lock is unlocked.
Wouldn’t it make sense to have a unlatch when unlocked action, without the need to actively check the state before?
I’m thinking about upgrading to the 4. Generation of the lock, but if I read correctly it’s not possible to explicitly query the lock state over Mqtt. How can I make sure I’m not unlocking and unlatching the door if I missed a locker event?
There has been no demand for this yet. Most likely because your use case is not a broad enough one. However, you can of course add a feature request and start collecting votes for it.
That’s correct, because MQTT works exclusively with pushes (which are stored and retained on server side). i.e. the server always remembers the last state that was sent by the client / lock. In addition there is a timestamp of the last update also retained by the server and the server reports if the lock is currently connected.
The lock sends its current state whenever it reconnects via MQTT and also while it is connected and the server retains the last state. The likeliness that events get lost is very low, but there is no 100% guarantee. For this you would need the “unlatch only when unlocked” command, which would check the state right at the time when the unlatch would be performed directly inside the lock.
Thanks for the long explanation.
I’ll add a feature request.