kuhyx/wake-alarm
1
0
Fork 0
mirror of https://github.com/kuhyx/wake-alarm.git synced 2026-11-11 13:07:32 +01:00
Code Issues Actions 6 Packages Projects Releases Wiki Activity

Record the phone alarm verified end to end, by the user, on the Pixel

Browse Source 
The dismiss path is no longer "covered by widget tests": the user ran it and
confirmed it. tag discovered: 0465dab3ce2a81 on the ring screen, challenge
solved, alarm silent.

Records the four defects that only a hardware run could have found — the
onPause reader-mode teardown, the cumulative step sensor that could never move
off 0/25, the ON_LOCKED NFC gating, and the self-hiding flash challenge — and
the two things still genuinely unverified: the QR scanner UI (which does not
exist) and the locked-keyguard dismiss (implemented, never run to completion).

Also records the process lesson, because it cost the most time here: after
hardware confirms a path, the next change must be the smallest delta on that
exact state, never a restructure bundled with a fix.

Co-Authored-By: Claude Opus 5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01EckWmyfetYq3Mi8MAnUudh
This commit is contained in:
Krzysztof Rudnicki 2026-08-27 22:47:20 +02:00
parent a845cac883
commit f53003ee4d
1 changed files with 42 additions and 35 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

77
TODO-hardening-continue.md
View File

@ -20,7 +20,7 @@ starting; this file deliberately does not duplicate it.
| 2 power button | **Done.** `_inhibit.py` (logind block, self-terminating, stale cleanup by pid via `/proc`) + `_alarm_session.py` fire-time record, re-fired by `_trigger` outside the catch-up window. |
| 10 vmbox | **Done.** `scripts/vmbox/harness.sh` + `DOCS-vmbox.md`. 8 cases, fails closed. Found 2 live bugs the host could not reach. |
| 5 curtains | **Done.** `_curtains.py` driver protocol + no-op default, called from `prepare_hardware()`. `DOCS-curtains-hardware.md` has the buy list. |
| 1/4 phone alarm | **Done.** Five Kotlin files behind one MethodChannel + the Dart ring screen. The phone arms and rings its own alarm; no Clock-app confirmation. Verified on the API 36 AVD (screen Asleep→Awake, AlarmActivity over the lockscreen) and deployed to the Pixel 6a. |
| 1/4 phone alarm | **Done, verified on the phone.** Five Kotlin files behind one MethodChannel + the Dart ring screen. Fires from Doze on time, wakes the screen over the lockscreen, rings through DND on STREAM_ALARM, and dismisses via an NFC tap + challenge. Confirmed by the user on the Pixel 6a, not inferred. |
| 3 room proofs | **Mostly done.** NFC works end to end: `NfcReader.kt` (hand-written), Settings → "Room proof" registers the tag, and a tap satisfies the gate. Steps+elapsed is the fallback. QR is specified, persisted and unit-tested but has **no scanner UI** — `mobile_scanner` is the droppable extra. |
## Remaining, in order
@ -35,47 +35,54 @@ starting; this file deliberately does not duplicate it.
3. **Item 13, black screen.** Different repo (`~/testsAndMisc/
linux_configuration`), host-only, blocks nothing.
### Measured on the Pixel 6a, 2026-08-27
### Verified end to end on the Pixel 6a, 2026-08-27
**Verified for real:** the alarm fires from Doze on time (22:03:01 for a
22:03 arm), wakes the screen (`Asleep`→`Awake`), puts `AlarmActivity` on top
over the lockscreen, posts the full-screen-intent notification, and **rings
through DND** — `zen_mode=1`, player `state:started` on
`usage=USAGE_ALARM`/`STREAM_ALARM(4)`, `mutedState:none`. The STREAM_ALARM
claim is no longer an assumption. NFC tag registration works end to end
(`0465dab3ce2a81` via Settings → Room proof).
**The whole path is confirmed on hardware, by the user, not inferred:**
**Still not verified: the actual dismiss.** Three real defects were found and
fixed in the attempt (see `08b44b4`), but none of the fixes has been confirmed
on hardware yet — the last test run ended before a successful tap. What needs
one more run: register/confirm the tag, fire a ring, unlock at the keyguard
prompt, tap the tag, solve the challenge, and confirm the sound stops **and**
the screen closes.
- Fires from Doze on time (22:03:01 for a 22:03 arm), registered by the system
as `RTC_WAKEUP` / "Next wake from idle".
- Wakes the screen (`Asleep`→`Awake`) and puts `AlarmActivity` on top over the
lockscreen, with the full-screen-intent notification posted.
- **Rings through DND** — `zen_mode=1`, player `state:started` on
`usage=USAGE_ALARM`/`STREAM_ALARM(4)`, `mutedState:none`. The STREAM_ALARM
claim is measured, no longer assumed.
- NFC tag registers (Settings → Room proof) and **a tap dismisses**:
`tag discovered: 0465dab3ce2a81` on the ring screen, challenge solved,
alarm silent.
- Step counter reaches the gate.
**`requestDismissKeyguard` does not silently unlock a secured phone.** It
raises the PIN/biometric prompt and returns to `ON_LOCKED` until the user
answers, so the NFC tap cannot work before that point. That is the agreed
behaviour, not a bug — but it means "tap the tag without touching the phone"
was never achievable on a secured device, and the docs should not imply it.
**Four defects were found only because it was run on the phone**, every one
invisible to the emulator (no NFC, no step sensor) and to the widget tests
(which mock the channel):
### Not yet verified on the phone
1. `AlarmActivity.onPause` called `NfcReader.stop()`. `enableReaderMode` is
per-activity but `disableReaderMode` is global, so anything briefly in
front of the ring screen killed polling for the rest of the ring.
2. `TYPE_STEP_COUNTER` is cumulative and on-change: it emits nothing until the
user has already walked, and the first emission was being eaten as the
baseline, so the screen sat at `0/25` forever. `TYPE_STEP_DETECTOR` is now
primary — one event per stride, no baseline to miss.
3. NFC does not poll while the ring shows over a locked keyguard
(`mScreenState=ON_LOCKED`). `requestDismissKeyguard()` is called on fire.
It does **not** silently unlock a secured phone — it raises the PIN prompt
— so "tap without touching the phone" was never achievable on a secured
device. Say that plainly; don't let docs imply otherwise.
4. The **flash** challenge hides its code after 4s, which is unsolvable when
the screen is reached by walking to another room. Removed from the phone's
rotation, with a test asserting no self-hiding challenge is ever shown.
The ring itself has only been fired on the API 36 AVD. On the Pixel 6a the
build installs and the runtime permission prompts appear, but **an actual
alarm has never rung on it** — that needs an alarm armed through the picker
(which needs a sync backend connected) and a wait to the fire instant. Ring
with **DND on**, since "STREAM_ALARM ignores DND" is the load-bearing claim
and has not been observed on hardware.
**A process lesson worth more than any of them:** after hardware confirmed the
path worked, the next change bundled a MethodChannel restructure with a bug
fix and regressed both NFC and steps. The user caught it. When hardware has
just gone green, make the smallest possible delta on top of *that exact state*
— never a restructure and a fix together.
The **dismiss path** has never run on any device either: the AVD has neither a
step sensor nor NFC, so the room gate correctly refuses to open there. It is
covered by widget tests (NFC tap → challenge → `dismiss` + `closeRingScreen`),
which is weaker than a real tap on a real tag. Register a tag via Settings →
"Room proof" and dismiss a real ring before calling item 3 finished.
### Still unverified
**Mute the host before firing a ring on the emulator.** The AVD routes
`STREAM_ALARM` to the PC's speakers, and `raiseAlarmVolume()` sets it to max —
this went off out loud during the session (2026-08-27).
- The **QR** proof has no scanner UI (`mobile_scanner`); `checkQr` and its
persistence are unit-tested only.
- The **locked-keyguard** dismiss was not run to completion: the confirmed run
was on an unlocked phone. The PIN-prompt path is implemented but unproven.
## Decisions already made — do not re-litigate