* Fix Google Calendar caching when offline
Events from Google Calendar were not firing under the following
circumstances:
1. Start ha as normal with Google Calendar configured as per
instructions.
2. ha loses network connectivity to Google
3. ha attempts update of Google Calendar
4. calendar/google component throws uncaught Exception causing update
method to not return
5. (cached) Google Calendar event does not fire, remains "Off"
Catching the Exception and returning False from the update() method
causes the correct behavior (i.e., the calendar component firing the
event as scheduled using cached data).
* Add requirements
* Revert code cleanup
* Remove explicit return value from update()
* Revert "Remove explicit return value from update()"
This reverts commit 7cd77708af658ccea855de47a32ce4ac5262ac30.
* Use MockDependency decorator
No need to whitelist google-python-api-client for a single unit test at
this point.
* Fix Google Calendar caching when offline
Events from Google Calendar were not firing under the following
circumstances:
1. Start ha as normal with Google Calendar configured as per
instructions.
2. ha loses network connectivity to Google
3. ha attempts update of Google Calendar
4. calendar/google component throws uncaught Exception causing update
method to not return
5. (cached) Google Calendar event does not fire, remains "Off"
Catching the Exception and returning False from the update() method
causes the correct behavior (i.e., the calendar component firing the
event as scheduled using cached data).
* Add requirements
* Revert code cleanup
* Remove explicit return value from update()
* Revert "Remove explicit return value from update()"
This reverts commit 7cd77708af658ccea855de47a32ce4ac5262ac30.
* Use MockDependency decorator
No need to whitelist google-python-api-client for a single unit test at
this point.
DarkSky delivers relative humidity from 0-100% as a 0-1 decimal value,
so we convert it by multiplying by 100.0. Unfortunately, due to floating
point math, the display of a raw value of 0.29 ends up looking like
28.999999999999996% relative humidity.
This change rounds the value to two decimal places.
DarkSky delivers relative humidity from 0-100% as a 0-1 decimal value,
so we convert it by multiplying by 100.0. Unfortunately, due to floating
point math, the display of a raw value of 0.29 ends up looking like
28.999999999999996% relative humidity.
This change rounds the value to two decimal places.
Accounts with multiple LaMetric devices at unreachable IPs (for example
at a different location, on a different/unroutable subnet, etc.) may
cause the notify.lametric service to fail. This update wraps the message
sending routine in a try/except clause and outputs log messages
indicating the problem.
Fixes#12450