* get single device web socket command
* test get single device
* add not found error
* fix handling when device doesn't exist
* add test for zha device not found
* Nightly docker builds / Hass.io dev HA
* use same style
* Finish nightly build
* Update builder version
* Fix style
* fix style part 2
* Last one
* Fix order
* Speed up history `get_states`
Adding a boundary of the start of the recorder run the point is in, significantly increases the time of the query. This speeds up the fetching of the history of 1 entity.
* Make single entity query easier
no need for joins with single entity
* Lint
I was getting the following error when i set the scan_interval
```
self.scan_interval = timedelta(seconds=config.get(CONF_SCAN_INTERVAL, 60))
TypeError: unsupported type for timedelta seconds component: datetime.timedelta
```
it turns out `config.get(CONF_SCAN_INTERVAL)` already returns `timedelta`
```('scan_interval', datetime.timedelta(seconds=180))```
* Add nws weather.
* Hassfest
* Address multiple comments
* Add NWS icon weather code link
* Add metar fallback.
Use metar code from nws observation if normal api is missing data.
* only get 1 observation - we dont use more than 1
* add mocked metar for tests
* lint
* mock metar package for all tests
* add check for metar attributes
* catch errors in setup
* add timeout error
* handle request exceptions
* check and test for missing observations
* refactor to new pynws
* change to simpler api
* Make py3.5 compatible
Remove f string
* bump pynws version
* gen_requirements
* fix wind bearing observation
* Revert "Make py3.5 compatible"
This reverts commit 4946d91779.
* Precommit black missed a file?
* black test
* add exceptional weather condition
* bump pynws version
* update requirements_all
* address comments
* move observation and forecast outside try-except-else
* Revert "move observation and forecast outside try-except-else"
This reverts commit 53b78b3283.
* remove else from update forecast block
* remove unneeded ConfigEntryNotReady import
* add scan_interval, reduce min_time_between_updates
* pytest tests
* lint test docstring
* use async await
* lat and lon inclusive in config
* define success for all branches
* add operation_mode to error when unexpected value
* fix black linting
* black linting
* fix weird black linting result
* Load user-provided descriptions for python_scripts
* Import SERVICE_DESCRIPTION_CACHE
* Use async_set_service_schema to register service descriptions
* Add python_script tests for loading service descriptions
* Use async/await in test
* Add descriptive fields to script config
* Add script descriptions to hass.data["service_description_cache"]
* Import SERVICE_DESCRIPTION_CACHE
* Register script descriptions via async_set_service_schema
* Add scripts test for loading and reloading service descriptions
* Minor cleanup
* Clean up script schema
* Added attribute support to template binary sensor with tests
Added attribute support to template binary sensor with tests
* fix dictionary update
fix dictionary update
* Fixed whitespace and line length issues
* Fixed indentation
* Simplify applying of attribute templates based on feedback
* Syntax and whitespace fixes
* Black formatting
* Black formatting on tests
* Check attribute_templates is not None
* Fixed test
* Added test for failure to render template
* Test fix
* Updated test
* Removed whitespace and applied Black formatting
* Fixed test assertion
* Updated test
* Code improvements folloing comments
Using chain to iterate over templates and attribute_templates
Replacing dict() with {}
Rmoving unused constant
* Applied Black formatting
* Fixed removed code
* Default attribute_templates to empty dict
* Black formatting
* Fixed imports