[GreenGrassBlueOcean]

Problem

test_issue_746_switch_04_wall_and_ha_toggling_and_scenario_6_events failed once on the HA Compatibility (beta) job of #781 (run 39998271332):

assert 'pushbutton_short_release' == 'pushbutton_short_press'   (tests/test_issue_746_mh200n_trace.py:183)

Both CEN events were fired (the count assertion passed) but landed in the wrong order. That PR only changed comments. The same test passed on its previous run, on v2-phase1-architecture seven minutes earlier, and on the stable and dev jobs of the same run.

Cause

The test registered cen_events.append (and the scene / general-light lists) as plain event-bus listeners. Home Assistant runs a listener that is neither a coroutine function nor marked @callback as an executor job, so two events fired back to back can be appended in either order.

Fix

Register the listeners through a small @callback helper. They then run on the event loop, in firing order.

Verification

• The test passes 25/25 runs locally with the change.
• I could not reproduce the flake locally. The unchanged test also passed 25/25, so the executor ordering is the explanation from how Home Assistant dispatches listeners, not something I observed failing. A timing-dependent failure like this one only shows up under CI load.
• Two other tests register list.append listeners (test_access_policy.py, test_issue_368_resync.py). They wait for a single event, so order cannot matter there, and I left them alone.

🤖 Generated with Claude Code

[codecov-commenter]

⚠️ Please install the to ensure uploads and comments are reliably processed by Codecov.

Codecov Report

✅ All modified and coverable lines are covered by tests.

📢 Thoughts on this report? Let us know!