* Add ability to ignore heos discovery
* Fetch player_id, update tests
* Handle failure state
* Update tests as there are two players in the mock now
* Adjust and add more tests
* Strip out player id lookup
* reverts per review
* one more revert
* fallback to device serial number if no mac available
* make password optional to fix#35904
* update tests to reflect new flow
* fix snake case and AsyncMock
* add comments around why weird things are being done
* Add config flow to pi-hole
* Add config flow tests
* Change PlatformNotReady to ConfigEntryNotReady
* Improve config flow
* Add @shenxn as codeowner
* Use entity_id as unique id
* Remove .get with [] for required fields
* Remove unique id from config flow
* Replace some strings with references
* Fix api_key string
* Fix service api_key check
* Remove unused DuplicatedNameException
* New forked_daapd component
* Bunch of changes
Add config flow and zeroconf
Add zones on callback when added by server
Add password auth
Add async_play_media for TTS
Add media_image_url
Add support for pipe control/input from librespot-java
Improve update callbacks
* Refactor as per code review suggestions
Move config_flow connection testing to pypi library (v0.1.4)
Remove use of ForkedDaapdData class
Decouple Master and Zone data and functions
Add updater class to manage websocket and entity updates
* More changes as per code review
Avoid direct access to entities in tests
Bump pypi version
Mark entities unavailable when websocket disconnected
Move config tests to test_config_flow
Move full url creation from media_image_url to library
Move updater entity from master to hass.data
Remove default unmute volume option
Remove name from config_flow
Remove storage of entities in hass.data
Use async_write_ha_state
Use signal to trigger update_options
Use unittest.mock instead of asynctest.mock
* Yet more changes as per code review
Add more assertions in tests
Avoid patching asyncio
Make off state require player state stopped
Only send update to existing zones
Split up some tests
Use events instead of async_block_till_done
Use sets instead of lists where applicable
Wait for pause callback before continuing TTS
* Remove unnecessary use of Future()
* Add pipes and playlists as sources
* Add support for multiple servers
Change config options to add max_playlists+remove use_pipe_control
Create Machine ID in test_connection and also get from zeroconf
Modify hass.data storage
Update host for known configurations
Use Machine ID in unique_ids, entity names, config title, signals
* Use entry_id as basis for multiple entries
* Use f-strings and str.format, abort for same host
* Clean up check for same host
* Add integration for wiffi devices
wiffi devices are DIY board manufactured by stall.biz.
Several devices are available, e.g. a weather station (weatherman), an
indoor environmental sensor (wiffi-wz) and some more.
This intgration has been developed using a weatherman device, but should
also work for other devices from stall.biz.
* Fix pylint warning
* Use WIFFI / STALL WIFFI instead of wiffi to be consistent with stall.biz
* Don't update disabled entities.
* fix complains
- move wiffi specific code to pypi
- remove yaml configuration code
* incorporate various suggestions from code review
* fix remaining comments from Martin
* fix comments
* add tests for config flow
* fix comments
* add missing requirements for tests
* fix pylint warnings
* fix comments
* fix comments
remove debug log
rename .translations to translations
* rebase and adapt to latest dev branch
* Update homeassistant/components/wiffi/config_flow.py
Co-Authored-By: Martin Hjelmare <marhje52@gmail.com>
* Update homeassistant/components/wiffi/config_flow.py
Co-Authored-By: Martin Hjelmare <marhje52@gmail.com>
* fix missing import
Co-authored-by: Martin Hjelmare <marhje52@gmail.com>
* Use helper functions for exception handling
* Create a separate class to handle communication with the device
* Update manifest
* Use coroutine for service setup
* Fix sensor update
* Update tests
* Fix MP1 switch
* Add device.py to .coveragerc
* Remove unnecessary blocking from test_learn_timeout
* Change access method for entries with default values
* Make the changes suggested by MartinHjelmare
* Remove dot from debug message
* Use underscore for unused variable
* Ensure homekit_controller recieves zeroconf c# updates
If an integration has a homekit config flow step
homekit controller would not see updates for
devices that were paired with it and would not
rescan for changes.
* Only send updates to homekit controller if the device is paired
This avoids the device showing up a second time.
* remove debug
* fix refactor error
* add binary sensor platform to zwave_mqtt
* add tests for binary_sensor
* fix tests
* device class is required value
* Update homeassistant/components/zwave_mqtt/binary_sensor.py
use colon as separator
Co-authored-by: Martin Hjelmare <marhje52@gmail.com>
* extend tests
* code optimize
* add test for enabling a legacy binary sensor
Co-authored-by: Martin Hjelmare <marhje52@gmail.com>
* rewrite library
* Update strings.json
* fix updated with empty reply
* dont use entity_id
* atag_id
* use super init instead
* original ids to prevent breaking change
Each ServerBrowser currently runs in its own thread which
processes every A or AAAA record update per instance.
As the list of zeroconf names we watch for grows, each additional
ServiceBrowser would process all the A and AAAA updates on the network.
To avoid overwhemling the system we pre-filter here and only process
DNSPointers for the configured record name (type)
* Ensure camera stream can be restarted after failure
* If ffmpeg failed to start, was killed, or the iOS device
closed the stream right away, the stream could never
be started until the HomeKit bridge was restarted.
* watch ffmpeg instead of checking only once
* handle forceful shutdowns gracefully
* Increase coverage
