Hello everyone,
My first request is about « virtual devices ». I know that with MQTT, we can do a good part of what I am going to explain, but… it’s not fully integrated into Gladys…
Being able to create virtual devices for more flexibility:
Create your own designs for devices that cannot be edited (e.g., Z-Wave)
Manage more complex scenes (e.g., a device with multiple states instead of the device itself being non-editable)
Trigger a scene by a device on the dashboard
And the Jeedom documentation:
Consolidate information or actions from multiple physical/virtual devices into a single device;
Create a device powered by an external source (Node-RED…)
Duplicate a device to split it into two, for example
The devices created in the MQTT integration are quite normal, and creating an integration that would just be a copy/paste of the MQTT integration with a different name won’t change much, except the name of the integration
You can run a mosquitto container with this integration, it’s very lightweight, mosquitto.
After that, from what I remember, I don’t think the MQTT integration configuration is mandatory to create MQTT devices…
In any case, even if you run a mosquitto container and there’s nothing on the other side listening to this container, it doesn’t change anything Hence the fact that it is already a « virtual » integration in a sense.
I partially agree with the observation about the non-integrated side of Gladys regarding certain uses of the MQTT integration.
In my case, I have rather used this integration and its « virtual » devices to make « variables ».
I set a state to 0 or 1 in a scene, and I control this state in another scene.
I really use them as variables, and therefore this requires having an MQTT bus available.
I don’t know if there is a notion of a variable (not user-related nor linked to integrations) in the backlog?
Perhaps this has already been discussed and decided since I have not followed the exchanges on the v4 development this year.