* avion light: Bump python-avion dependency version
The dependencies in python-avion weren't sufficiently strict. This is now
fixed, but means we need to depend on a new version.
* decora light: Bump python-decora dependency
There's a new version of python-decora with a reliability fix, so depend on
that.
* wierd pylint complaint
* Workaround for Graber csz1 shades
* logging
* Try direct
* Try direct
* Use workaround
* Review changes and tests
* test
* reset test
* Use Bright and Dim also as open and close is
* Added the pushbullet sensor component
* Updated requirements_all.txt and .coveragerc with the new sensor
* Updated acording to houndci-bots comments
* Some more changes
* Final change by the hound (?)
* Fixes from balloobs review and from houndci-bot
This changes the sensors information to only contain one attribute
as information, and the rest as device_state_attributes.
* Added leading space to comments
* Added docstrings, removed API_KEY from log, changed imports
* The hound is at it again
* Fix remaining issues
* Fix pylint issue
* updated pylgtv module to fix problems with timeouts
* - update pylgtv to 0.1.6
- handle new TimeoutError exception from pylgtv
* used full name for exception handling of concurrent.futures._base.TimeoutError
* the exception handling should now follow the rules
* float typecasting should not be necessary
* use asyncio for TimeoutError it’s an alias for concurrent.futures.TimeoutError
The arwn platform was refactored to be asyncio friendly, however in
doing so one thing was missed which was explicitly telling hass when
something interesting has happened. This led to the very interesting
to debug issue that the state cards were all out of date, even though
the graphs were not.
* Update ios.py
as discussed. the part:
if battery_state == ios.ATTR_BATTERY_STATE_FULL:
returning_icon_level = DEFAULT_ICON_LEVEL
kinda screws up the charging icon.
i might just miss a logical solution for that though.
let me know what you think. it might not be beautiful but i think its an overall improve over the current "double battery" solution
* Update ios.py
chound fix and full_battery_charge fix
* Update ios.py
removed new line
* Update ios.py
* Update ios.py
* Update ios.py
* Update ios.py
* Update ios.py
* Update ios.py
* merged request from robbie
* Update ios.py
* Update ios.py
* Update ios.py
* Refactor into find_hsbk
This will be useful for new methods that also have to find passed in colors.
* Add AwaitAioLIFX
This encapsulates the callback and Event that aiolifx needs and thus avoids an
explosion of those when new calls are added.
The refresh_state is now generally useful, so move it into its own method.
* Initial effects support for LIFX
These effects are useful as notifications. They mimic the breathe and pulse
effects from the LIFX HTTP API:
https://api.developer.lifx.com/docs/breathe-effecthttps://api.developer.lifx.com/docs/pulse-effect
However, this implementation runs locally with the LIFX LAN protocol.
* Saturate LIFX no color value
Now the color is "full saturation, no brightness". This avoids a lot of
temporary white when fading from the "no color" value and into a real color.
* Organize LIFX effects in classes
This is to move the setup/restore away from the actual effect, making it quite
simple to add additional effects.
* Stop running LIFX effects on conflicting service calls
Turning the light on/off or starting a new effect will now stop the running
effect.
* Present default LIFX effects as light.turn_on effects
This makes the effects (with default parameters) easily accessible from
the UI.
* Add LIFX colorloop effect
This cycles the HSV colors, so that is added as an internal way to set a
color.
* Move lifx to its own package and split effects into a separate file
* Always show LIFX light name in logs
The name is actually the easiest way to identify a bulb so just using it
as a fallback was a bit odd.
* Compact effect getter
* Always use full brightness for random flash color
This is a stopgap. When a bit more infrastructure is in place, the intention
is to turn the current hue some degrees. This will guarantee a flash color
that is both unlike the current color and unlike white.
* Clear effects concurrently
We have to wait for the bulbs, so let us wait for all of them at once.
* Add lifx_effect_stop
The colorloop effect is most impressive if run on many lights. Testing
this has revealed the need for an easy way to stop effects on all lights
and return to the initial state of each bulb. This new call does just that.
Calling turn_on/turn_off could also stop the effect but that would not
restore the initial state.
* Always calculate the initial effect color
To fade nicely from power off, the breathe effect needs to keep an
unchanging hue. So give up on using a static start color and just find the
correct hue from the target color.
The colorloop effect can start from anything but we use a random color
just to keep things a little interesting during power on.
* Fix lint
* Update .coveragerc
* Refactored mvglive.py
This pull requests builds on the first work with the mvglive sensor:
- Refactoring the code so that multiple sensors for departures can be added
- Rewrites the transport mode restrictions ("products") to be more modular
- Fixes bugs, such as missing implementation of line restriction
- Other improvements, such as including data attribution
* Further improvements to mvglive sensor
- The API returns the property 'direction', which can be used to filter U-Bahn trains by direction without having to enter all final destinations
- The sensor icon now corresponds to the mode of transport of the next departure
* UBahnDirection refactored
U-Bahn SEV (bus replacement services) have unexpected direction values, fixed resulting bug and hound issues
* ready for PR
* minor fix
* another minor fix
* new platform mqtt_json
* using ATTR constants
* voluptuous check on JSON payload
* voluptuous check on JSON payload
* Add condition for API failure
If you are not running the latest ve3rsion of ZM this will cause exceptions to fire. This fix handles a response from ZM but a non successful attempt.
This resolves the issue https://github.com/home-assistant/home-assistant/issues/7178
* Fixed houndci-bot issues
* added new service
* fixed basic test in kodi platform
* Added new method async_get_albums
* Added new methods in module kodi
* added method find_song in kodi module
* method add_song_to_playlist made
async
* Added media type to method async_play_media
* added methods async_clear_playlist
and play_song
* methods play_song and find_song
made async
* added new service play_song
* Improved kodi._find
now it find for whole words only
* added possibility to specify artist in
kodi.async_find_artist
* added kodi.async_find_album
* added new optional input to play_song service
* In async_play_song added handling of no song found
* default artist value changed to ''
* async_add_song_to_playlist now can also
search for musinc
* added service add_song_to_playlist
* Added new service add_album_to_playlist
* added services to switch shuffle mode
* added service add_all_albums_to_playlist
* handled error in async_unset_shuffle
and async_set_shuffle
* Added abstract methods to media_player
* _server substituted with server property
* style made consistent with requirements
* Fixed issue with pylint
* Services moved to kodi platform
* service play_song removed
* removed service unset_shuffle
* all add services merged into one
* removed service get_artists
* added kodi_ to service names
* Fixed some style issues
* Removed changes in media_player __init__
* Implemented requested changes
* Fixed pylint problem
* Value of 0 should not be considered `unknown`.
* Reflect disconnect state in entity states.
* Due to adding unknown state on disconnect, the amount of reconnects can sometimes be more. Test for at least 2 reconnect attempts.
* Fixes issue with id mismatch when multiple devices are connected to the lutron bridge
* Updates labels
* removes no longer needed config values.
* removes no longer needed imports
* Add support for tradfri color temp
* Only return rgb_color if supported features indicate support.
* Return color_temp if supported features indicate support and dict
with allowed color temperatures exist for manufacturer.
* Add manufacturer specific supported features, with default.
* Color temperature is set from a dict of allowed pairs per
manufacturer, where kelvin is mapped to hex_color. When a user sets a
color temperature, the closest matching temperature is selected and
set.
* Clean up
* exposed content_type in rest_command, which allows for manually specifying the content_type for more-strict api endpoints
* fixed up column length
Length was 86 chars, and it needed to be 79
* double import of HTTP_HEADER_CONTENT_TYPE
Removed the accidental double-import of HTTP_HEADER_CONTENT_TYPE
* moved rest_command-specific config value into component
* if no content_type, default to None
* unit test
* newline
* unused CONTENT_TYPE_TEXT_PLAIN
* removed the http-agnostic abstraction hass provided in favor of aiohttps hdrs constant
* Cache the name of LIFX lights
After #7031 the LIFX device will change during an unregister/register
transition. This has the user-visible effect of the new device missing
a friendly name until the next poll.
We now cache the name internally and it will then transfer to the new
device when it registers.
* Allow LIFX logging even without an available device
This will allow us to set the device to None when it unregisters.
* Calculate LIFX availability from the existence of a device
This has become possible because the device is no longer needed
to provide the name of the light when it is unavailable.
We just have to forget the device when it unregisters.
* Always return True/False from is_state and is_state_attr
These functions are documented to always return True/False but the
short-circuit evaluation would return None if the entity_id did not exist.
* Reword into a single statement
* Make Tradfri discoverable
* Fix lint errors
* Fix bugs and clean up calls to light_control
* Add more color util tests
* Add coap client to dockerfile
* After rebase and all fixes
* Added color_rgb_to_hex to util.color
* Added test_color_rgb_to_hex
* Changed reference to color_rgb_to_hex
* Bumped to pytradfri 0.5, having support for retry
* Bumped to pytradfri 0.5, having support for retry
* Bumped to pytradfri 0.5, having support for retry
* Bumped to pytradfri 0.5, having support for retry
* Rolled back to 0.4
* Rolled back to 0.4
When pinging an inaccessible device, OS errors like
ping: sendto: No route to host
can occur. For a ping tracker this is not an error but rather a normal
situation. Thus, it makes sense to hide the error.
* Added initial support for IKEA Tradfri Gateway
* Pinned requirement
* Fixed lint-errors
* Added file to .coveragerc
* Trying to fix commit
* Fixed requirements_all again
* Minor reorder of code
* Minor reorder of code
* Made the changes suggested by @balloob
* Made the changes suggested by @balloob and removed debug
* Update tradfri.py
* Responsiveness
* Delay was not needed as commands does not return until done.
* Add support for cleaning maps and cleaning data
* Hound
* Docstring
* Update requirements
* Review changes
* External lib now returns the raw data.
* debug
* Sensor did not refresh
* Error handling
* Issue warning on connection error
* Update requirements
* Review changes
* Managing binary payloads
Hello,
background: I wrote a HA camera component that gets the image from a binary payload. I'm testing it with Zanzito (https://play.google.com/store/apps/details?id=it.barbaro.zanzito) and it works apparently well: it gets the image and correctly displays it in the front-end.
But I had to make the changes I'm proposing here: the message was being blocked because the utf-8 decoding failed.
As far as I know, the utf-8 encoding is required for the topic, not for the payload. What I did here was try the utf-8 decoding, but even if unsuccessful, it dispatches the message anyway.
Is there anything else I'm missing?
thanks
Gianluca
* Update __init__.py
* Update __init__.py
* Update __init__.py
* git test - ignore
* Should work
* minor fixes
* updated mqtt/services.yaml
* added two tests, modified threaded subscribe
* removing polymer
* requested changes
* requested changes - minor fix
* security wrap around payload_file_path
* services.yaml updated
* removed file publishing
* minor fix
* Added AlarmDecoder platform
* remove try/catch for generic execption
* Changes for @pvizeli, thanks for the review!
Removed _ prefix from normal function variables
Removed _hass as it will be set via .hass for us
Broke out the three config (socket, serial, usb) and use vol.Any
Added support for USB I think, don't have device, but should work
Removed components dictionary, was form old group all code that didn't work
* Fix hass string handling
* Fix mysensors callback
* All messages was not triggering proper updates. Fix by checking all
child value types each update.
* Upgrade mysensors dep
* Fix pickle persistence when upgrading.
* Plug file leak on LIFX unregister
The aiolifx 0.4.4 release closes its socket when the unregister callback is
called. This plugs a file descriptor leak but also means that we must be
careful to not use the device after it goes unavailable.
Also, when a light reappears, it has a new device that must be used.
* Do not test self.available in service calls
The core will learn to handle that.
* Fix#6783, remove a test that makes no sense anymore
* Fix#6784
* Fix typo in docstring
* Fix handling of known->unknown state, extended test, fix lint errors
* Better handling of mismatch in unit of measurement.
Set state to "unkown" and unit of measurement to "ERR" if unit of measurement differs between aggregatet states.
Add entity_id to logged warning.
* Make icon configurable
* Fix typo
* Fix lint
* Fix lint
* Fix lint
* Add option to set entity_id on min_max sensor
* Fix lint logging-not-lazy
* Revert "Add option to set entity_id on min_max sensor"
This reverts commit 4685f26647.
* Revert "Make icon configurable"
This reverts commit fe45aec82d.
* Fixes
* Adding exception handling when fetching the camera image to avoid python errors when host is not reachable or any url errors to camera
* Added exception as ConnectionError instead of plain except
* Added exception as ConnectionError instead of plain except. Removed the unused error handle
* Initial import for HassIO
* Cleanup api code for views
* First unittest for view
* Add test for edit view
* Finish unittest
* fix addons test
* cleanup service.yaml
* Address first round with ping command
* handle timeout dynamic
* fix lint
* Added average temperature for the day before and the current period
* Fixed "line too long" warnings
* Fixed "indentation contains tabs" and "indentation contains mixed spaces and tabs" warnings
* Fixed "trailing whitespace" warnings
* upgrade pyhydroquebec requirements to version 1.1.0
* Fire EVENT_HOMEASSISTANT_START automations off right away while starting
* Actually have core state be set to 'starting' during boot
* Fix correct start implementation
* Test and deprecate event automation platform on start
* Fix doc strings
* Remove shutting down exception
* More strict when to mark an instance as finished
* Add automation platform to listen for start/shutdown
* When we stop we should wait till it's all done
* Fix testing
* Fix async bugs in tests
* Only set UVLOOP when hass starts from CLI
* This hangs normal asyncio event loop
* Clean up Z-Wave node entity test
* Added eddystone_temperature platform.
* Fixed style issues.
* Fixed style issues #2.
* Fixed style issues #3.
* Added new platform to .coveragerc
* Refactored platform to use the beacontools package.
* Fixed style issues and added beacontools to excluded requirements.
* Removed obsolete constants and added pylint exception.
* Added blank line
* Updated beacontools to version 1.0.0
* Updated beacontools to version 1.0.1
* Forgot to regenerate requirements_all
* Minor changes
* Input Boolean: initial state > restore state
* Input select: initial state overrules restored state
* Input slider: initial state overrule restore state
* Lint
* Lint
* total connect alarm support
* linting fixes
* linting fixes
* docstring
* docstring
* use sync and update coveragerc
* remove unused import
* changes as per notes
* Update HA code style
The Hook switch component currently uses username/password to get a token
from the Hook web app to use for subsequent requests. This adds an option
to specify the token directly in config.
Makes the 'password' and 'token' options mutually exclusive since
otherwise one would have to take precedence, and it seems worth
preventing people from filling their config with passwords that don't
even need to be there.
* Automation: initial state > restore state
* Clean up code
* Ensure MQTT defaults are used.
* Ensure failed platforms always return None
* Automation: write state to state machine after start
* Added Met Office weather and sensor components
* Removed unnecessary dependancy
* Generated requirements
* Fix time interval
* Updated coverage
* Some review changes
* Allow user to specify lat and lon in component
* Added missing import
* Fixed unit
* Fixed import indent
* Updated condition to use CONDITION_CLASSES
* - Introduced Ring binary_sensor.
- Added unittest for Ring binary_sensor.
- Bumped ring_doorbell 3rd party module.
* Updated requirements
* Added correct file for unittest
* - Introduced Ring binary_sensor.
- Added unittest for Ring binary_sensor.
- Bumped ring_doorbell 3rd party module.
* Updated requirements
* Added correct file for unittest
* Added extra sensors last_ding and last_motion
* Modified Ring binary_sensor and sensor to inherit DOMAIN configuration
* Moved static to top ring.py
* Fixed requirements
* Bump version ring_doorbell to 0.1.2
* testing unittests
* Use hass.data dict instead GLOBALS
* Fixed unittests
* Bump ring_doorbell to 0.1.3
* Updated unittest and coverted to use decorator @requests_mock.Mocker()
* Updated ring_session with lower case
When the user exceeds the request limit for google APIs, the response status stays at 200 but the response body is different:
```
{
"error_message" : "You have exceeded your daily request quota for this API. We recommend registering for a key at the Google Developers Console: https://console.developers.google.com/apis/credentials?project=_",
"results" : [],
"status" : "OVER_QUERY_LIMIT"
}
```
This prevents HA from creating the initial configuration
* Add legacy LIFX platform for Windows support
The async platform introduced in 9ef084d903 has
turned out to use Python functionality that is not available in Windows.
This commit restores the previous implementation, now named lifx_legacy.
* Add a comment about the platform being a legacy implementation
* Warn when using unsupported lifx platform on Windows
* Update .coveragerc
* Add support for group commands (allon/alloff).
Add 'group_aliasses' config attribute that only respond to group commands.
Add nogroup_aliases that only respond to 'on' 'off' commands.
Allow settings device id group behaviour.
* Fix linting.
* Fix lint.
* yeelight: adjust supported features on update()
Earlier we checked for the type only during the initialization,
but this won't work when the bulb is disconnected during the init,
causing failures to adjust rgb&color temperature even if those should be supported.
fixes#6692
* Use reassign instead of OR for updating the supported features
* fix WOL in docker/jail
add ip_address parameter to send_magic_packet if host specific.
in docker/jail, WOL doesn't works due to subnet broadcast issues.
* Update wake_on_lan.py
lint
* Add link to docs and remove comments which are obvious
* Update docstrings
* Repleace conf details with link to docs
* Add link to docs
* Update docstrings
* Update import
* Update ordering
* Update ordering
* Update docstring
* Update ordering
* Update ordering
Addressing #6382 . Feedback from github & forums is the bridge_led feature never worked and defining the bridge LED as another group+bulb type is the right way to do this in the limitlessled component.
* New indexes for states table
* Added recorder_runs indexes
* Created a new function for compound indexes.
A new function was created because it makes it a little cleaner when creating
a single-field index since one doesn't have to create a list. This is mostly
when creating the name of the index so with a bit more logic it's possible
to combine it into one function. Given how often migration changes are run,
I thought that code bloat was probably a worthy trade-off for now.
* Adjusted indexes, POC for ref indexes by name.
* Corrected lint errors
* Fixed pydocstyle error
* Moved create_index function outside apply_update
* Moved to single line (just barely)
* Adding expire_after to mqtt sensor to expire outdated values
* Extending test case
* mqtt: refactoring expire_after to use timed events instead of polling; lint
* refactor to reset unused trigger
* Fix: handler must be set to None after execution or removal to avoid warning
* Commenting out non-working test
* Fix lint
* Commit to trigger new build
* Commit to trigger new build
* Make testcase work
* Undo unnecessary change
* Remove default value, add extra check
* Add sensor for Lyft time and price (based on Uber sensor)
* Minor fixes to lyft sensor
- use add_devices(...,True) instead of explicitly calling update
- move sensor name check into constructor
* lyft sensor: disable sandbox mode
* Adds support for the Lutron Caseta family of devices
* Added external requirement
* Removes unuse import
* Adds requirement to requirements_all.txt
* Removes unuse import
* fixes requirement_all.txt. by regenerating
* Cleans up syantax
* Cleans up syantax
* Cleans up syantax
* Cleans up syantax
* Shortens long line
* adds lutron_caseta component to .coveragerc
* Merges into light, removes component
* Fixes long lines
* Fixes requirement.txt
* Removes 'ERROR' log statements. Adds missing dependency
* savig work to pick up on other machine
* Enables Lutron Caseta component with Light and Switch platforms
* Add missing file to .coveragerc
* Changes based on PR review
* fixes requirements file
* Fixes install of paramiko dependency.
* Moves callback registration to
* comment changes
* Platform have no return value
* Change style for guard
* fix logic
* fix lint