* Add PG LAB Electronics integration
* Add time from last boot sensor diagnostic
* Limit the initial new pglab integration to only one platform
* Update FlowHandler with the new return type ConfigFlowResult
* Fix docstring file with the right integration name to PG LAB.
* There is no need for default value in the callback definition.
* Move all mqtt callbacks to be global and also renamed with a better name.
* Removed unused member variables.
* Renaming functions with a better name.
* Adding miss docstring to __build_device.
* Renamed CreateDiscovery with a better name.
* Removing not so meaning comment.
* Avoid to populate hass.data with pglab discovery information. Use hass.data[DOMAIN] instead.
* Revert "Removed unused member variables."
This reverts commit 4193c491ec.
* Removed unused member variables.
* Refactoring of const. Be sure to have in const.py constant that are used in at least two other modules
* Restoring back the process to unregister the plaform when unload the integration.
* fix spelling mistake
* Revert "Move all mqtt callbacks to be global and also renamed with a better name."
This reverts commit d94d8010d5.
* Main refactoring to avoid to store PG Lab discovery in hass.data
* Change class name BaseEntity in PGLabEntity. And named PyPGLab... what imported from external python module pypglab.
* Avoid to use dict to create DeviceInfo
* Removing unused parameter
* Removing not necessary call to base class
* Update entity name/id to be compatible with the new integration policy.
* Upate test to new entity id
* Add new line after file description
* avoid to store in local variable data for calling function
* Move PGLABConfigEntry in __init__.py
* change function to pure callback
* to avoid hang, dont' trust the split of the discovery topic... introduce a max split count
* rename method with a more meaning name
* use assignment operator
* rename variable with a better name
* removing unecessary test
* Raise exception in case of unexpected error during discovery
* Review comments all other the intergration.
* Rename classes to be consistent in integration
* Using new feature single_config_entry to allow single instance integration
* rename class FlowHandler to PGLabFlowHandler
* using __package__ to initialize integration logger
* missing to catch the exception when for some reason is not possible to create the discovery instance. This can happen when the discovery MQTT message is not in valid json format.
* using ATTR_ENTITY_ID instead of the string
* using SOURCE_MQTT, SOURCE_USER instead of config_entries.SOURCE_MQTT, config_entries.SOURCE_USER
* Using FlowResultType.ABORT instead of the string value
* Code refactoring for tests of configuration from USER and MQTT
* Remove to the user the possibility to add PGLab integration manually, and remove not needed tests.
* Change test_device_update to use snapshot to check test result
* Raise exeception in case of unexpected device and entity_id
* Avoid to log on info channel.
* Renamed _LOGGER in LOGGER
* Propage the call to the base class
* Remove not needed code because from the manifest it's only allows a single instance
* Using specific type for result test instead of string value
* Code refactoring, avoid not necessary function
* update to the new way to import mqtt components
* Avoid runtime check
* add err variable for catching the exception
* add doc string to mqtt_publish
* add doc string to mqtt_subscribe
* Rename DiscoverDeviceInfo.add_entity_id in add_entity
* add doc string
* removing not meaning documentation string
* fix spelling
* fix wrong case in docstring
* fix spelling mistake in PyPGLab callback name
* rename mqtt message received callback
* Avoid to store hard coded discovery_prefix
* Removing unused strings from strings.json
* Give to the user more information during config_flow, and add the possibility to add manually the integration
* Fix to avoid fails of auto test
* update discovery test
* Be sure to always subscribe to MQTT topic when entity is added to HA
* Update codeowner of PGLAB integration and test
* Add control to check if mqtt is available during integration setup
* New test for check no state change for disable entity switch
* Remore not more used file
* update pypglab to version 0.0.3 and improve the symmetry to subscribe/unsubscribe to mqtt entity topic and to register/deregister the status update callback
* Update codeowner of pglab integration
* Adding quality_scale
* removing async_setup
* Fix spelling mistake
* Added test to cover config_flow.async_step_user
---------
Co-authored-by: Pierluigi <p.garaventa@gmail.com>
* Add Google Drive integration for backup
* Add test_config_flow
* Stop using aiogoogle
* address a few comments
* Check folder exists in setup
* fix test
* address comments
* fix
* fix
* Use ChunkAsyncStreamIterator in helpers
* repair-issues: todo
* Remove check if folder exists in the reatuh flow. This is done in setup.
* single_config_entry": true
* Add test_init.py
* Store into backups.json to avoid 124 bytes per property limit
* Address comments
* autouse=True on setup_credentials
* Store metadata in description and remove backups.json
* improvements
* timeout downloads
* library
* fixes
* strings
* review
* ruff
* fix test
* Set unique_id
* Use slugify in homeassistant.util
* Fix
* Remove RefreshError
* review
* push more fields to the test constant
---------
Co-authored-by: Joostlek <joostlek@outlook.com>
* Add the Model Context Protocol Server integration
* Remove unusued code in init
* Fix comment wording
* Use util.uild for unique ids
* Set config entry title to the LLM API name
* Extract an SSE parser and update comments
* Update comments and defend against already closed sessions
* Shorten description
* Update homeassistant/components/mcp_server/__init__.py
Co-authored-by: Paulus Schoutsen <paulus@home-assistant.io>
* Change integration type to service
---------
Co-authored-by: Paulus Schoutsen <paulus@home-assistant.io>
* Add NASweb integration
* Fix DeviceInfo import
* Remove commented out code
* Change class name for uniquness
* Drop CoordinatorEntity inheritance
* Rename class Output to more descriptive: RelaySwitch
* Update required webio-api version
* Implement on-the-fly addition/removal of entities
* Set coordinator name matching device name
* Set entities with too old status as unavailable
* Drop Optional in favor of modern typing
* Fix spelling of a variable
* Rename commons to more fitting name: helper
* Remove redundant code
* Let unload fail when there is no coordinator
* Fix bad docstring
* Rename cord to coordinator for clarity
* Remove default value for pop and let it raise exception
* Drop workaround and use get_url from helper.network
* Use webhook to send data from device
* Deinitialize coordinator when no longer needed
* Use Python formattable string
* Use dataclass to store integration data in hass.data
* Raise ConfigEntryNotReady when appropriate
* Refactor NASwebData class
* Move RelaySwitch to switch.py
* Fix ConfigFlow tests
* Create issues when entry fails to load
* Respond when correctly received status update
* Depend on webhook instead of http
* Create issue when status is not received during entry set up
* Make issue_id unique across integration entries
* Remove unnecessary initializations
* Inherit CoordinatorEntity to avoid code duplication
* Optimize property access via assignment in __init__
* Use preexisting mechanism to fill schema with user input
* Fix translation strings
* Handle unavailable or unreachable internal url
* Implement custom coordinator for push driven data updates
* Move module-specific constants to respective modules
* Fix requirements_all.txt
* Fix CODEOWNERS file
* Raise ConfigEntryError instead of issue creation
* Fix entity registry import
* Use HassKey as key in hass.data
* Use typed ConfigEntry
* Store runtime data in config entry
* Rewrite to be more Pythonic
* Move add/remove of switch entities to switch.py
* Skip unnecessary check
* Remove unnecessary type hints
* Remove unnecessary nonlocal
* Use a more descriptive docstring
* Add docstrings to NASwebCoordinator
* Fix formatting
* Use correct return type
* Fix tests to align with changed code
* Remove commented code
* Use serial number as config entry id
* Catch AbortFlow exception
* Update tests to check ConfigEntry Unique ID
* Remove unnecessary form abort