maximunited / imou_life / 25288999887

90%

push

github

feat: implement UI device deletion support (async_remove_config_entry_device) (#32)

* feat: implement async_remove_config_entry_device for manual device deletion

Complete stale-devices Gold tier requirement with UI deletion support.

**Added:**
- `async_remove_config_entry_device()` callback in `__init__.py`
- Enables delete button in Settings ? Devices & Services UI
- Validates device belongs to config entry before allowing removal
- Returns True for safe removal, False to prevent deletion
- DeviceEntry import from helpers.device_registry

**Testing:**
- Created `test_device_removal.py` with 4 comprehensive tests
- Tests matching device removal (returns True)
- Tests non-matching device prevention (returns False)
- Tests multiple identifiers handling
- Tests edge case with missing device_id
- All 396 unit tests passing ?

**Gold Tier Impact:**
- Completes stale-devices requirement (automatic + manual)
- Users can now delete devices from UI
- Automatic detection via repair flow
- Manual deletion via device settings

**Documentation:**
- Updated `gold_tier_verification.md` with full implementation evidence
- Documents both automatic detection and manual deletion support

Co-Authored-By: Claude Sonnet 4.5 <noreply@anthropic.com>

* fix: correct device identifier comparison in async_remove_config_entry_device

**Bug Fix:**
CodeRabbit caught a critical bug - was comparing device identifier against
CONF_DEVICE_ID instead of config_entry.entry_id.

**Root Cause:**
Entities register devices with identifier tuple: (DOMAIN, config_entry.entry_id)
But the function was checking: identifier[1] == device_id (from CONF_DEVICE_ID)
These are different values, so device removal would always fail.

**Fix:**
- Changed comparison from device_id to config_entry.entry_id
- Added comment explaining identifier source
- Updated all 4 unit tests to use correct identifier tuples
- Tests now use (DOMAIN, "test_entry_id") matching config_entry.entry_id

**Testing:**
- All 396 u... (continued)

11 of 11 new or added lines in 1 file covered. (100.0%)

1384 of 1544 relevant lines covered (89.64%)

3.58 hits per line