I think I was thinking in another language, since this was not valid Python. Result was that the thread that maintained the SocketIO connection to the table would die early on, so no status updates were actually flowing from the table.
* Bring the Sisyphus integration to silver quality
Checklist:
- [x] (N/A - push integration) Set an appropriate SCAN_INTERVAL (if a polling integration)
- [x] Raise PlatformNotReady if unable to connect during platform setup
- [x] (N/A - no credentials) Handles expiration of auth credentials. Refresh if possible or print correct error and fail setup. If based on a config entry, should trigger a new config entry flow to re-authorize.
- [x] (N/A - local integration) Handles internet unavailable. Log a warning once when unavailable, log once when reconnected.
- [x] Handles device/service unavailable. Log a warning once when unavailable, log once when reconnected.
- [x] Set available property to False if appropriate
- [x] Entities have unique ID (if available)
* Feedback (fix a couple verbose places)
* Use a task instead of a lock
* Initialize field in constructor
* Revert package upgrade.
springstan
Jonathan Keljo
Paulus Schoutsen
cgtobi
Fabian Affolter
Paulus Schoutsen