* manual: make it completely async
Restoring the timers on startup cannot use track_point_in_time,
because the restoring code runs from the async_added_to_hass
method. Rewrite everything to run in the event loop instead
of using threaded wrappers, this way the code can be reused in
async_added_to_hass.
* manual_mqtt: replace async function with @callback
* manual_mqtt: make it completely async
Restoring the timers on startup cannot use track_point_in_time,
because the restoring code runs from the async_added_to_hass
method. Rewrite everything to run in the event loop instead
of using threaded wrappers, this way the code can be reused in
async_added_to_hass when state restore is added.
Co-authored-by: Jan Bouwhuis <jbouwh@users.noreply.github.com>
* Trigger reauth flow when HomeWizard API was disabled
* Add tests for reauth flow
* Fix typo in test
* Add parallel updates constant
* Improve error message when device in unreachable during config
* Set quality scale
* Remove quality scale
* Throw error instead of abort when setup fails
* Adjust test for new setup behaviour
* Trigger reauth flow when API is disabled and continue retrying
* Reload entry and raise AuthFailed during init
* Abort running config flow
* Listen for coordinator updates to trigger reload
* Use build-in backoff system
* Fix failing test
* Test reauth flow is active after disable-api init
* Test reauth flow removal
* Remove unused function in withings
* Move to tests
* Remove reference to attribute.platform
* Move WITHINGS_MEASUREMENTS_MAP to tests
* measute_type > measure_type
* One more
* Recognize more SSDP friendly names as "ours"
* Trigger on SoyeaLink devices
* Pass discovered URL through context, it does not persist via user_input
* Use manufacturer from SSDP
* Actually use serial number if available, update URL for configured
* Remove heuristic friendly name filtering against non-LTE devices
* Add language to conversation and intent response
* Move language to intent response instead of speech
* Extend intent response for voice MVP
* Add tests for error conditions in conversation/process
* Move intent response type data into "data" field
* Move intent response error message back to speech
* Remove "success" from intent response
* Add id to target in intent response
* target defaults to None
* Update homeassistant/helpers/intent.py
* Fix test
* Return conversation_id and multiple targets
* Clean up git mess
* Fix linting errors
* Fix more async_handle signatures
* Separate conversation_id and IntentResponse
* Add unknown error code
* Add ConversationResult
* Don't set domain on single entity
* Language is required for intent response
* Add partial_action_done
* Default language in almond agent
* Remove partial_action_done
* Fix linting
* Rename success/fail targets
Co-authored-by: Paulus Schoutsen <balloob@gmail.com>
NyroandGitHub
epenetandGitHub
Marc MuellerandGitHub
GitHub Action
Daniel Hjelseth HøyerandGitHub
Duco SebelandGitHub
muppet3000andGitHub
Joakim PlateandGitHub
Franck NijhofandGitHub
Bram KragtenandGitHub
ollo69andGitHub
Erik MontnemeryandGitHub
Aaron BachandGitHub
Ville SkyttäandGitHub
Philip AllgaierandGitHub
Robert SvenssonandGitHub
Rami MoslehandGitHub
majussandGitHub
Øyvind Matheson WergelandandGitHub
