* Switch russound to pypi

* Switch pymysensors to pypi

* Switch pocketcasts to pypi

* Generate requirements file for tests

* Update tox

* Update validate

* Lint

* Tweak order in travis.yml to run longest job first

* Update to pyunifi 2.12

* Update requirements_all.txt

* Fix plant MIN_TEMPERATURE, MAX_TEMPERATURE validation

small_float only allows values from 0 to 1 so we should use float instead

* Do not use vol.All for a single validation

* Fix plant MIN_TEMPERATURE, MAX_TEMPERATURE validation

small_float only allows values from 0 to 1 so we should use float instead

* Do not use vol.All for a single validation

* Update to pyunifi 2.12

* Update requirements_all.txt

0.44.1

* Version bump to 0.44.2

* Update frontend

The hue is now a float but the hsbk conversion still believed it to be
an integer that could not be larger than 359. The float can in fact be,
for example, 359.9 and this would cause an out-of-bounds error in the
set_color call.

For completeness, the initial hue is also changed to a float.

A small bug in the python-rflink library caused packets to be printed. This update prevents this from happening.

* Refactor sun component for correctness

* Convert datetimes to dates for astral

* Fix tests for updated code

* Fix times now that calcs are fixed

* Move sun functions to helpers

* Fix flake on new file

* Additional tweaks from review

* Update requirements

Returned on the current playback response itself, not the device

* Add support for automation config

* Build fromtend

* Lint

* # This is a combination of 3 commits.
# The first commit's message is:
Add seperate zwave panel

# The 2nd commit message will be skipped:

#	unused import

# The 3rd commit message will be skipped:

#	Use get for config

* Add seperate zwave panel

* more info

* Add usercodeview

* Improve api

* Improve api

* Separate api into own file.

* disable missing import

* review changes

* Tests 1

* Verify that we fetch data from groups

* Tests groups

* config 1

* usercode 1

* Api mods

* Tweak API

* docstrings

* 100% api testing

* Update docker dev environment to python3.6

* comment out disable switches again

…#7438)

* sensor.envirophat: do not set up platform if hardware is not attached

Fixes comment from:
#7427 (comment)

* Fix update logic.

…services (#7454)

* telegram_bot and notify.telegram enhancements:
- Receive callback queries and produce `telegram_callback` events.
- Custom reply_markup (keyboard or inline_keyboard) for every type of message (message, photo, location & document).
- `disable_notification`, `disable_web_page_preview`, `reply_to_message_id` and `parse_mode` optional keyword args.
- Line break between title and message fields: `'{}\n{}'.format(title, message)`
- Move Telegram notification services to `telegram_bot` component and forward service calls from the telegram notify service to the telegram component, so now the `notify.telegram` platform depends of `telegram_bot`, and there is no need for `api_key` in the notifier configuration. The notifier calls the new notification services of the bot component:
	- telegram_bot/send_message
	- telegram_bot/send_photo
	- telegram_bot/send_document
	- telegram_bot/send_location
	- telegram_bot/edit_message
	- telegram_bot/edit_caption
	- telegram_bot/edit_replymarkup
	- telegram_bot/answer_callback_query
- Added descriptions of the new notification services with a services.yaml file.
- CONFIG_SCHEMA instead of PLATFORM_SCHEMA for the `telegram_bot` component, so only one platform is allowed.
- Async component setup.

* telegram_bot and notify.telegram enhancements: change in requirements_all.txt.

* Added rpi_pfio component supporting the PiFace I/O module

* Fixed some code style issues

* Removed global listener

* Update rpi_pfio.py

* Add not-context-manager

* Add missing comma

* Make Configurator thread safe, get_instance timing issues breaking configurator working on multiple devices

* No blank lines allowed after function docstring

* Fix comment Tox

* Generate a new updateDate with every call

This should fix #7459
Tests need to be updated in another commit.

* Replace STATIC_TIME with datetime object check

Removing the "DATE" argument from the Alexa component's configuration (because it is now dynamically generated) requires this commit's changes to the test cases to check that the updateDate data is a datetime type rather than a specific hardcoded value ('2016-10-10T19:51:42.0Z').

* Fix brackets

* ps - do not install all dependencies

* Comment out blinkt because it depends on GPIO

* Add pip upgrade check back

* Disable import error blinkt

* Update comment

* Fix comment

#7532)

* Support adding different server locations

* Rename variables and move CONF_ const into component as requested in review

* Fix unittests

* Forgot to add tests for microsoft_face_identify

* Added Axis hub, binary sensors and camera

* Added Axis logo to static images

* Added Axis logo to configurator
Added Axis mdns discovery

* Fixed flake8 and pylint comments

* Missed a change from list to function call
V5 of axis py

* Added dependencies to requirements_all.txt

* Clean up

* Added files to coveragerc

* Guide lines says to import function when needed, this makes Tox pass

* Removed storing hass in config until at the end where I send it to axisdevice

* Don't call update in the constructor

* Don't keep hass private

* Unnecessary lint ignore, following Baloobs suggestion of using NotImplementedError

* Axis package not in pypi yet

* Do not catch bare excepts. Device schema validations raise vol.Invalid.

* setup_device still adds hass object to the config, so the need to remove it prior to writing config file still remains

* Don't expect axis.conf contains correct values

* Improved configuration validation

* Trigger time better explains functionality than scan interval

* Forgot to remove this earlier

* Guideline says double qoutes for sentences

* Return false from discovery if config file contains bad data

* Keys in AXIS_DEVICES are serialnumber

* Ordered imports in alphabetical order

* Moved requirement to pypi

* Moved update callback that handles trigger time to axis binary sensor

* Renamed configurator instance to request_id since that is what it really is

* Removed unnecessary configurator steps

* Changed link in configurator to platform documentation

* Add not-context-manager (#7523)

* Add not-context-manager

* Add missing comma

* Threadsafe configurator (#7536)

* Make Configurator thread safe, get_instance timing issues breaking configurator working on multiple devices

* No blank lines allowed after function docstring

* Fix comment Tox

* Added Axis hub, binary sensors and camera

* Added Axis logo to static images

* Added Axis logo to configurator
Added Axis mdns discovery

* Fixed flake8 and pylint comments

* Missed a change from list to function call
V5 of axis py

* Added dependencies to requirements_all.txt

* Clean up

* Added files to coveragerc

* Guide lines says to import function when needed, this makes Tox pass

* Removed storing hass in config until at the end where I send it to axisdevice

* Don't call update in the constructor

* Don't keep hass private

* Unnecessary lint ignore, following Baloobs suggestion of using NotImplementedError

* Axis package not in pypi yet

* Do not catch bare excepts. Device schema validations raise vol.Invalid.

* setup_device still adds hass object to the config, so the need to remove it prior to writing config file still remains

* Don't expect axis.conf contains correct values

* Improved configuration validation

* Trigger time better explains functionality than scan interval

* Forgot to remove this earlier

* Guideline says double qoutes for sentences

* Return false from discovery if config file contains bad data

* Keys in AXIS_DEVICES are serialnumber

* Ordered imports in alphabetical order

* Moved requirement to pypi

* Moved update callback that handles trigger time to axis binary sensor

* Renamed configurator instance to request_id since that is what it really is

* Removed unnecessary configurator steps

* Changed link in configurator to platform documentation

* No blank lines allowed after function docstring

* No blank lines allowed after function docstring

* Changed discovery to use axis instead of axis_mdns

* Travis CI requested rerun of script/gen_requirements_all.py

* Move service helpers to LifxManager

* Add lifx_set_color

This is a synonym for light.turn_on except it does not actually turn on the
light unless asked to.

Thus, turn_on can be implemented just by asking.

* Rename set_color to set_state

* Support power=False with lifx_set_state

* Added raspihats binary_sensor platform

* Updated .coveragerc to ommit raspihats platforms.

* Using vol.Coerce(int) for validation and casting of I2CHat config address

* Initialize sun with unknown values.

Initial values should be `unknown` instead of `0`. Otherwise on HA restart the value of `0` is pushed to metrics databases (graphite/influx/recorder).

* Update sun position before emitting initial update

* Simplify based on armills comment.

* Use provided time for calculation.

* Adds support for the Lutron Caseta Serena shades hardware

* fixes typos

* Tests for zwave discovery logic

* Simplify patching

* Test ignored node

* Tests for zwave setup features

* Add test for frontend panel register

Bumped blink version to support automatic reauthorization when tokens expire. Changed the battery sensor call to a string version so that the battery reports back "Low" or "OK" rather than a cryptic integer

* Fix Kodi specific services, add descriptions, add more handled exceptions

 - Fixes issue #7528
 - Add descriptions for Kodi specific services in services.yaml.
 - Error handling in Kodi API errors.
 - Make compatible the existent specific service `media_player.kodi_set_shuffle` with the general `media_player.shuffle_set` service (both use the same method but with different named parameter, I think the Kodi specific service should be eliminated, since it is not)

* fix line too long

* removed new services (for another PR); removed `kodi_set_shuffle` service

* requested changes

 - Removed `kodi_set_shuffle` service.
 - Optional `media_name` and `artist_name` parameters. `media_name` defaults to 'ALL'.
 - Guard clause to check if the services are already registered.

* Add Kira component to sensor and remote platforms

* Test cases for Kira component and platforms

* Additional Wink lock features

* Websocket_api: avoid parallel drain

* Remove send_message method

* No longer require pyunify during tests

* No longer require cast during tests

* No longer required dependency for tests

* No longer require pymochad for tests

* Astral is a core dependency

* Avoid having to install datadog dependency during tests

* CMUS test doesn't test anything

* Frontier Silicon doesn't test anything

* No longer require mutagen

* Update requirements_test_all.txt

* Remove stale comment

* Fix websocket api reaching queue

* Fix outside task message sending

* Fix Py34 tests

…in email headers (#7533)

* SMTP notify enhancements: HTML emails and customization options

- Send full HTML emails, with or without inline attached images.
- Custom `timeout`.
- Custom `product_name` identifier for the `FROM` and `X-MAILER` email headers.
- New HTML email test

* `sender_name` instead of product_name

 - Change `sender_name` instead of `product_name`.
 - No changes in `X-Mailer` header.
 - `From` header as always unless you define the new `sender_name` parameter.

* snmp sensor protocol version configuration option

* fixed lint findings.

* Add sonos alarm clock update service

* Add tests and break lines

* Fix style errors

* Make test work with python<3.6

* Fix last two pylint errors

* fix new line to long errors

* Add File sensor

* Use None and return

* Remove I/O

* Use less memory

* No traceback if file is empty

* Make miflora monitored_conditions parameter optional.

* Use default keyword instead.

* Use list instead of tuple

* Simplify even more

* Add Content-type: image/jpeg for camera proxy

* Set content_type in constructur

* Refactor color profiles to a class

* Refactor into preprocess_turn_on_alternatives

* LIFX: use light.preprocess_turn_on_alternatives

This avoids the color_name duplication and gains support for profile.

* Add kelvin parameter to light.turn_on

* Add brightness_pct parameter to light.turn_on

* LIFX: accept brightness_pct in effects

* Add test of kelvin/brightness_pct conversions

* Add support for disabling tradfri groups

* Fixed styleguide problems

* Fix style problems

* Use default for groups when setting up in UI

* Kodi specific services to call Kodi API methods

 - new service: `kodi_execute_addon` to run a Kodi Addon with optional parameters. Results of the Kodi API call, if any, are redirected in a Home Assistant event: `kodi_execute_addon_result`.
 - new service: `kodi_run_method` to run a Kodi JSONRPC API method with optional parameters. Results of the Kodi API call are redirected in a Home Assistant event: `kodi_run_method_result`.
 - Add descriptions in services.yaml.
 - Add `timeout` parameter to yaml config (needed to make slow queries to the JSONRPC API, default timeout is set to 5s).
 - Trigger events with the results of the Kodi API calls, with:
 ```
 event_data = {
   'result': api_call_results,
   'result_ok': boolean,
   'input': api_call_parameters,
   'entity_id': 'media_player.kodi'}
```

* no need to clean OrderedDicts; no need for the `kodi_execute_addon` service

* no need for the `kodi_execute_addon` service

* unused import

* naming changes

self._brightness is assigned with the returned value of the
set_luminance() function, which is always equal to None.

* Add initial seven segments OCR image processing

* Fix typo

* Implemented ffmpeg option on Amcrest camera and upgraded to version 1.2.0

* Added ffmpeg arguments and binary options to Amcrest camera

* Added ffmpeg as dependencies

* Makes lint happy and fixed requirements_all.txt

* Inherent the ffmpeg.binary configuration from ffmpeg component

* Update amcrest.py

* update usps

* fix doc

* # This is a combination of 3 commits.
# The first commit's message is:
Add seperate zwave panel

# The 2nd commit message will be skipped:

#	unused import

# The 3rd commit message will be skipped:

#	Use get for config

* Add seperate zwave panel

* Modify set_config_parameter to accept setting string values

* descriptions

* Tweaks

* Tweaks

* Tweaks

* Tweaks

* lint

* Fallback if no config parameteres are available

* Update services.yaml

* review changes

* # This is a combination of 3 commits.
# The first commit's message is:
Add seperate zwave panel

# The 2nd commit message will be skipped:

#	unused import

# The 3rd commit message will be skipped:

#	Use get for config

* Add seperate zwave panel

* Modify set_config_parameter to accept setting string values

* descriptions

* Tweaks

* Tweaks

* Tweaks

* Tweaks

* lint

* Fallback if no config parameteres are available

* Update services.yaml

* review changes

* Add network_key as a config option

* Update __init__.py

* Add network_key as a config option

* Update __init__.py

* Remove global hass

* Http.auth test no longer spin up server

* Remove server usage from http.ban test

* Remove setupModule from test device_sun_light_trigger

* Update common.py

* Added support for openhome devices using transport service

* Style cleanup

* Ignore attribute changes in automation trigger from/to

* Quote names in deprecation warnings

This makes it somewhat easier to read if the suggestion happens to be
named "to".

* Add test with same state, new attribute value

* Implemented ffmpeg option on Amcrest camera and upgraded to version 1.2.0

* Added ffmpeg arguments and binary options to Amcrest camera

* Added ffmpeg as dependencies

* Makes lint happy and fixed requirements_all.txt

* Inherent the ffmpeg.binary configuration from ffmpeg component

* Update amcrest.py

0.45

Prior to this change the zoneminder component was attempting to
use logging.exception outside of exception handling code. This
would lead to the traceback module throwing an exception when
trying to work out the traceback for the exception.

This fixes the issue by changing the exception call into a
plain error logging call.

* Apply refresh_node workaround on 1st instance only

* Add another test

* Add reset service for command_class meters.

* Add reset service for command_class meters.

* cast index to const.py

* bugfix for Telegram chat_ids

- Negative `chat_id`s for groups.
- Include `chat_id` in event data.
- Handle KeyError when receiving other types of messages, as `new_chat_member` ones, and send them as text.

* unused import

* fix double quote style, fix boolean expr, change warning msg

* mistake

* some more fixes

- fix if condition for msg bad fields.
- return True for a correct but not allowed or not recognized message: if not, the message arrives continuously.
- Allow to receive messages from unauthorized users if they come from authorized groups.

* support for `edited_message`s

- They come as normal messages, except for the 'edited_message' field instead of 'message'.

Web streams can't be paused and resumed later. That's why volumio stops them instead of pausing them.

* Sort entities in default groups by name

* Cleanups from review

…t installed (#7690)

If OpenWRT isn't running the DHCP server then some OpenWRT hardware,
such as TP-Link TL-WDR3600 v1, can't determine the host
corresponding to an associated wifi client. This change handles that
by returning None when the request has no data in the result.

* bugfix for Telegram chat_ids

- Negative `chat_id`s for groups.
- Include `chat_id` in event data.
- Handle KeyError when receiving other types of messages, as `new_chat_member` ones, and send them as text.

* unused import

* fix double quote style, fix boolean expr, change warning msg

* mistake

* some more fixes

- fix if condition for msg bad fields.
- return True for a correct but not allowed or not recognized message: if not, the message arrives continuously.
- Allow to receive messages from unauthorized users if they come from authorized groups.

* support for `edited_message`s

- They come as normal messages, except for the 'edited_message' field instead of 'message'.

Web streams can't be paused and resumed later. That's why volumio stops them instead of pausing them.

If OpenWRT isn't running the DHCP server then some OpenWRT hardware,
such as TP-Link TL-WDR3600 v1, can't determine the host
corresponding to an associated wifi client. This change handles that
by returning None when the request has no data in the result.

0.45.1

…drop from apcaccess (#7722)

Add an optional extended description…

* Make ssh and telnet connections continuous in asuswrt

* Refactored SSH and Telnet connections into respective classes.

* Fixed several copy-paste typos and errors.

* More typos fixed.

* Small changes to arguments, to pass automated tests.

* Removed unsupported named arguments.

* Fixed a couple of mistakes in Telnet, and other lint errors.

* Added Telnet tests, and added lint exceptions.

* Removed comments from tests, as they irritated the hound.

… (#7716)

…7625)

* Don't initialize mqtt components which have already been discovered

* Fix string length

* Fix blank lines, fix constant name

* Remove globals. Remove JSON dump

* Add tests. Update grammar

* PEP8 style issue

* Add hyphen to object_id regex

* PEP8 style fix

The product type is already established in order to decide the Kelvin range
so just reuse that information to disable color features for white-only lights.

Also change the breathe/pulse effects to be more useful for white-only
bulbs. For consistency, color bulbs set to a desaturated (i.e. white-ish)
color get the same default treatment as white-only bulbs.

#7739)