node-red-contrib-ccu 4.5.3

v4.5.3

on Node.js NPM

Fixed

• Home Assistant showed every entity of a device as "unavailable" for
  good (B-38, reported by mail). The discovery configs name the device's
  …:0/UNREACH topic as its availability, and Home Assistant treats an
  entity as unavailable until that topic delivers a value. Only ccu-mqtt
  published UNREACH: when it changed, or once in its start-up burst of
  cached values - which is off unless the node's Cache option is ticked,
  and lost when the MQTT broker is not connected yet at that moment (Node-RED's
  mqtt out drops what it gets while disconnected). A device that never went
  unreachable therefore never got a value on that topic, while its state
  updates arrived normally. The ccu-homeassistant node now publishes each
  device's current UNREACH itself, retained, right after the discovery
  configs (asking the CCU for it) and every change after that; republish
  discovery sends it again. The example flow also wires Home Assistant's
  birth message (homeassistant/status) to the node, so a Home Assistant
  restart republishes discovery and availability; an offline there is
  ignored.

Commits since v4.5.2

• 4.5.3 - Home Assistant entities no longer stay unavailable when UNREACH never changed (4870459)
• Home Assistant: publish the devices' availability with the discovery (B-38) (a6435c6)