* Add time to charge and type checking
* Revert drive_state_shift_state change
* Use original name
* Use function instead of lambda
* Update homeassistant/components/tessie/sensor.py
Co-authored-by: Jan-Philipp Benecke <github@bnck.me>
* Fix callback
* Avoid having to test None
* Go back to if
* Use minutes instead of hours
---------
Co-authored-by: Jan-Philipp Benecke <github@bnck.me>
* Use snapshot testing in Airly sensor
* Apply suggestions from code review
Co-authored-by: Robert Resch <robert@resch.dev>
* Fix tests
---------
Co-authored-by: Robert Resch <robert@resch.dev>
* Add helper function to update and reload config entry to config flow
* Use async_create_task
* Remove await
* Reload only when update & add task name
* Rename function
* Add typing to Lutron platforms
* Add devices to Lutron
* Add devices to Lutron
* Fix typing
* Fix
* Add name
* Fix lights
* Comment out ESA
* Fix domain
* Fix domain
* Fix
* Make generic keypad base class
Fix unifi test_tracked_clients client
This test relied on the sensor platform getting set up and creating
the device before the device_tracker platform was setup since the
device_tracker platform will disable the entity because there is
not matching device entry for it via
https://github.com/home-assistant/core/blob/4ee6735cbbf13c130ee7cf2ca99274141fafc0fc/homeassistant/components/device_tracker/config_entry.py#L336
There is no guarantee the sensor platform will get set up before
the device tracker platform so the test was subject to a race
where it would fail if the device tracker platform was setup first
I recently came up with an idea to look for callback functions
that get created over and over frequently by adding logging to
homeassistant.core.callback when its called to decorate
a function. This one happens a lot at runtime.
* Significantly reduce websocket api connection auth phase latancy
Since the auth phase has exclusive control over the websocket
until ActiveConnection is created, we can bypass the queue and
send messages right away. This reduces the latancy and reconnect
time since we do not have to wait for the background processing
of the queue to send the auth ok message.
* only start the writer queue after auth is successful
* Send recurrence data when updating a task in todoist
* Update tests/components/todoist/test_todo.py
Co-authored-by: Allen Porter <allen.porter@gmail.com>
* Move logic into _task_api_data.
* Add comment about sending potentinally stale data.
---------
Co-authored-by: Allen Porter <allen.porter@gmail.com>
J. Nick Koston
Jan-Philipp Benecke
Brett Adams
jmwaldrip
Joost Lekkerkerker
Michael
Franck Nijhof
Bruno Pantaleão Gonçalves
Martin Hjelmare
Erik Montnemery
Simone Chemelli
Robert Resch
Marc Mueller
Teemu R
Joakim Sørensen
Cody C
Aaron Godfrey
David Bonnes
Ståle Storø Hauknes
Jonas Fors Lellky