* Initial monzo implementation
* Tests and fixes
* Extracted api to pypi package
* Add app confirmation step
* Corrected data path for accounts
* Removed useless check
* Improved tests
* Exclude partially tested files from coverage check
* Use has_entity_name naming
* Bumped monzopy to 1.0.10
* Remove commented out code
* Remove reauth from initial PR
* Remove useless code
* Correct comment
* Remove reauth tests
* Remove device triggers from intial PR
* Set attr outside constructor
* Remove f-strings where no longer needed in entity.py
* Rename field to make clearer it's a Callable
* Correct native_unit_of_measurement
* Remove pot transfer service from intial PR
* Remove reauth string
* Remove empty fields in manifest.json
* Freeze SensorEntityDescription and remove Mixin
Also use list comprehensions for producing sensor lists
* Use consts in application_credentials.py
* Revert "Remove useless code"
Apparently this wasn't useless
This reverts commit c6b7109e47202f866c766ea4c16ce3eb0588795b.
* Ruff and pylint style fixes
* Bumped monzopy to 1.1.0
Adds support for joint/business/etc account pots
* Update test snapshot
* Rename AsyncConfigEntryAuth
* Use dataclasses instead of dictionaries
* Move OAuth constants to application_credentials.py
* Remove remaining constants and dependencies for services from this PR
* Remove empty manifest entry
* Fix comment
* Set device entry_type to service
* ACC_SENSORS -> ACCOUNT_SENSORS
* Make value_fn of sensors return StateType
* Rename OAuthMonzoAPI again
* Fix tests
* Patch API instead of integration for unavailable test
* Move pot constant to sensor.py
* Improve type safety in async_get_monzo_api_data()
* Update async_oauth_create_entry() docstring
---------
Co-authored-by: Erik Montnemery <erik@montnemery.com>
* Add support for RR DNS
* 🧪 Update tests for DNS IP round-robin
* 🤖 Configure DNS IP round-robin automatically
* 🐛 Sort IPv6 addresses correctly
* Limit returned IPs and cleanup test class
* 🔟 Change max DNS results to 10
* Rename IPs to ip_addresses
* Bump ondilo to 0.5.0 and handle errors retrieving data
* Bump ondilo to 0.5.0 and handle errors retrieving data
* Updated ruff recommendation
* Refactor
* Refactor
* Added exception log and updated call to update data
* Updated test cases to test through state machine
* Updated test cases
* Updated test cases after comments
* REnamed file
---------
Co-authored-by: Joostlek <joostlek@outlook.com>
* Bump google-generativeai to v0.5.2
* Avoid exceptions when Gemini responses are blocked
* pytest --snapshot-update
* set error response
* add test
* ruff
* Switch out aiohttp-isal for aiohttp-fast-zlib to make isal optional
aiohttp-isal does not work on core installs where the system has 32bit userland and a 64bit kernel because we have no way to detect this configuration or handle it.
fixes#116681
* Update homeassistant/components/isal/manifest.json
* Update homeassistant/components/isal/manifest.json
* hassfest
* isal
* fixes
* Apply suggestions from code review
* make sure isal is updated before http
* fix tests
* late import
We only did discovery for the key at setup time. If it
changed, a reloaded of the integration was needed to
update the key. We now update it every time we update
the lock detail.
* Turn on thread safety checks in async_dispatcher_send
We keep seeing issues where async_dispatcher_send is called from
a thread which means we call the callback function on the other
side in the thread as well which usually leads to a crash
* Turn on thread safety checks in async_dispatcher_send
We keep seeing issues where async_dispatcher_send is called from
a thread which means we call the callback function on the other
side in the thread as well which usually leads to a crash
* adjust
* Refactor entity_platform polling to avoid double time fetch
Replace async_track_time_interval with loop.call_later
to avoid the useless time fetch every time the listener
fired since we always throw it away
* fix test
* Improve recorder and worker thread matching in RecorderPool
Previously we would look at the name of the threads. This
was a brittle if because other integrations may name their
thread Recorder or DbWorker. Instead we now use explict thread
ids which ensures there will never be a conflict
* fix
* fixes
* fixes
