* Add ws get, set card
* lint+fix test
* Add test for set
* Added more tests, catch unsupported yaml constructors
Like !include will now give an error in the frontend.
* lint
* zone trigger supports entity id pattern
* fixed lint error
* fixed test code
* initial version of new geo_location trigger
* revert to original
* simplified code and added tests
* refactored geo_location trigger to be based on a source defined by the entity
* amended test cases
* small refactorings
Will set the available property to False if unable to communicate with August lock or doorbell.
HTTP request errors (i.e. timeout, connection error, HTTP error) will not result in traceback. Instead an error will be logged.
This prevents the following traceback that will otherwise occur.
Traceback (most recent call last):
File "/usr/local/lib/python3.6/concurrent/futures/thread.py", line 56, in run
result = self.fn(*self.args, **self.kwargs)
File "/usr/local/lib/python3.6/site-packages/homeassistant/components/device_tracker/bluetooth_le_tracker.py", line 107, in update_ble
see_device(address, devs[address], new_device=True)
File "/usr/local/lib/python3.6/site-packages/homeassistant/components/device_tracker/bluetooth_le_tracker.py", line 47, in see_device
see(mac=BLE_PREFIX + address, host_name=name.strip("\x00"),
AttributeError: 'NoneType' object has no attribute 'strip'
* Switch mailgun to use webhook api
* Generalize webhook_config_entry_flow
* Add tests for webhook_config_entry_flow
* Add tests for mailgun
* Remove old mailgun file from .coveragerc
* Refactor WebhookFlowHandler into config_entry_flow
* Remove test of helper func from IFTTT
* Lint
* Bug fix
Current version causes 500 error since it is sending an array of from numbers to ClickSend. Changing the from number to 'hass' identifies all messages as coming from Home Assistant making them more recognisable and removes the bug.
* Amendment
Changed it to use 'hass' as the default instead of defaulting to the recipient which is the array. Would have worked if users set their own name but users who were using the default were experiencing the issue.
* Added DEFAULT_SENDER variable
* Resolves /home-assistant/home-assistant#17433
Away mode temperature issue fix for generic_thermostat
* Debug messages removed from generic_thermostat.py
* Test for repeat away_mode set
Test for fix of generic thermostat issue when away_mode was set several times in a row.
* Code style fix in generic_thermostat
* Remove blank line in the end of generic_thermostat
* Fix style
* zone trigger supports entity id pattern
* fixed lint error
* fixed test code
* initial version of new geo_location trigger
* revert to original
* simplified code and added tests
* refactored geo_location trigger to be based on a source defined by the entity
* amended test cases
* small refactorings
* Add ws get, set card
* lint+fix test
* Add test for set
* Added more tests, catch unsupported yaml constructors
Like !include will now give an error in the frontend.
* lint
* New rtorrent sensor
* Fix lint issue
* Fix another lint issue
* Fix pylint issue
* how many python linters do you guys use
* Cleanup code
* python linting
* newline
* Added support for dyson hot+cool fan as climate device
* Removed decimal place in kelvin units conversion
Minor edits to be consistent with Dyson's internal conversion of temperature from kelvin to celsius. It does not include decimal place to convert between kelvin and celsius.
* made changes according to comments
* Refactored target temp logics, fixed enum issues
* changed name of component to entity
* removed temperature conversion for min/max property
* changed back to 644 permission
* added extra tests for almost-all coverage
* changed assert method to avoid lack of certain method in py35
* added test_setup_component
* shorten line length
* fixed mock spec and added checking of message listener is called
* added doc string and debug msg
* shorten line length
* removed pending target temp
* Enable brightness slider for RGB
If we are using RGB with no brightness topic, the brighness slider
should still be visible, as we can scale the RGB amount to give us the
brightness.
* Output RGB scaled by brightness
If we are outputting to an RGB device, but do not have a dedicated
brightness topic set, when the brightness slider is changed, we should
output the current colour's HS, with the V coming from the brightness
slider.
* Brightness from RGB when we're not using a brightness topic
When we aren't using a brightness topic, set the brightness slider based
on the received value from an RGB -> HSV conversion.
* Test for new brightness state scaled by RGB
This adds a test to make sure the brightness stored in the state is
being computed correctly from the RGB value when a dedicated brightness
topic is not set.
* Changes from review
Fixes formatting of supported features flags, and checks HS colour
hasn't been set when operating in RGB-only mode
* Set optimistic brightness correctly in rgb mode
When we're using rgb mode to set the brightness, we want to set
optimistic brightness if:
we are running in optimistic mode
OR
the brightness state topic isn't set and we have a brightness command topic
OR
the rgb state topic isn't set and we don't have a brightness command topic
* Add test for turn_on in RGB brightness mode
* Fix linky sensor login error
* Make platform fail-safe
Adding following enhancements:
* Make sure the platform loads correctly by making the first API request in setup_platform.
* Close the session after each API call.
* Use timeout parameter everywhere.
* Fix Hound CI error: line too long.
* Update pylinky library
* Remove LinkyClient from update()
* hass.async_add_executor_job
* Fix accessories.run -> async_track_state_change
* Fixed media_player test
* Flags are now local vars
* consistent use of " and '