adaptive-lighting/tests
Joe Neuman b1729d906c Add restore_manual_control option to keep manual control across restarts
A restart currently clears manual control for every light: the startup
adaptation resets the flags and force-adapts, so a light that was dimmed by
hand before the restart comes back at the adaptive values. With the new
opt-in option, manual control is written to storage as it changes (debounced,
one store per Home Assistant instance) and restored at startup. The restored
flags are resolved once at EVENT_HOMEASSISTANT_STARTED: kept while the light
is on and at least one profile controlling it has the option without
autoreset_control_seconds, dropped otherwise, and forgotten when no profile
controls the light anymore, so the outcome does not depend on the order the
profiles are set up in. Only a full restart restores; a config-entry reload
still clears manual control. Unloading the last entry writes a pending save.
Defaults stay unchanged and nothing is read or written unless a profile opts in.

Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EdXCHuaz4Yj932oEwtgouW
2026-09-08 18:00:23 -07:00
..
__init__.py Copy tests from https://github.com/home-assistant/core/pull/40626 2022-08-28 12:05:31 -07:00
conftest.py Bump to 1.27.0 for fixes in 2025.12 (#1293) 2025-11-27 09:06:41 -08:00
README.md Set conservative line and branch coverage floors for stable HA versions (#1566) 2026-09-06 17:18:47 +02:00
test_adaptation_utils.py Fix missing follow-up commands during multi-light interception (#1560) 2026-09-06 14:03:08 +02:00
test_automation_examples.py fix: synchronize sleep mode and mixed-light split timing (#1579) 2026-09-06 21:50:32 +02:00
test_color_and_brightness.py fix: keep adapting during polar night and midnight sun (#1489) 2026-09-06 00:23:21 -07:00
test_config_flow.py Add restore_manual_control option to keep manual control across restarts 2026-09-08 18:00:23 -07:00
test_diagnostics.py Add on-demand Home Assistant diagnostics (#1575) 2026-09-06 19:11:55 +02:00
test_hass_utils.py Test HA core v2023.6 until v2024.2 (#942) 2024-04-06 10:14:54 +02:00
test_init.py Register service actions in async_setup for Bronze tier compliance (#1403) 2026-09-06 13:04:18 +02:00
test_switch.py Add restore_manual_control option to keep manual control across restarts 2026-09-08 18:00:23 -07:00

Developer notes for the tests directory

To run the tests, check out the CI configuration to see how they are executed in the CI pipeline. Alternatively, you can use the provided Docker image to run the tests locally or run them with VS Code directly in the dev container.

Coverage reports

Open a pytest workflow run in GitHub Actions to see line and branch coverage in each job's summary. Download its coverage-<Home Assistant version>-py<Python version> artifact for the XML and JSON reports and the browsable HTML report. After extracting it, open htmlcov/index.html to inspect missing lines and branches. Supported stable Home Assistant versions require at least 89% line coverage and 80% branch coverage. The dev job reports coverage without enforcing these floors.

Coverage measures executed code, not whether assertions would catch a bug. Add tests for observable behavior: emitted light commands, final states, manual-control events, and timer expiry. The integration suite runs inside Home Assistant with simulated lights; it does not establish physical-device behavior. It also does not execute every documentation generator included in the package's coverage total.

The tests in test_automation_examples.py load YAML directly from README.md and execute it through Home Assistant's automation and script engines. Edit the README source when changing those examples, then run ./scripts/update-generated-content to update the documentation pages.

Prerequisites

Before running tests with Docker, you need a local Home Assistant core checkout with symlinks:

# Clone HA core (one-time setup)
git clone --depth 1 https://github.com/home-assistant/core.git core

# Setup symlinks (one-time setup)
./scripts/setup-symlinks

Running tests with Docker

Navigate to the adaptive-lighting repo folder and execute the following command.

Important: Mount the entire repo (-v $(pwd):/app), not individual directories, or the symlinks will break.

Linux / MacOS / Windows PowerShell:

docker run -v ${PWD}:/app basnijholt/adaptive-lighting:latest
  • In windows command prompt, the command is:
    docker run -v %cd%:/app basnijholt/adaptive-lighting:latest
    

This command will download the Docker image from the adaptive-lighting Docker Hub repo and run the tests.

If you prefer to build the image yourself, use the following command:

docker build -t basnijholt/adaptive-lighting:latest --no-cache --progress=plain .

This might be necessary if the image on Docker Hub is outdated or if the test_dependencies.py file is updated.

Passing arguments to pytest

You can pass arguments to pytest by appending them to the command:

For example, to run the tests with a custom log format, use the following command (this also gets rid of the captured stderr output):

docker run -v $(pwd):/app basnijholt/adaptive-lighting:latest --show-capture=log --log-format="%(asctime)s %(levelname)-8s %(name)s:%(filename)s:%(lineno)s %(message)s" --log-date-format="%H:%M:%S" tests/components/adaptive_lighting/