The PC keeps a dropout moving at 1 m/s to 25 km/h; the app zeroed
every gap over 30 s. Bridge unpaused dropouts, pauses stay zero.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWXcjiPpgRfxzrRfoky
Pure-Dart TCX codec pinned to the PC's RunnerUp parser by a shared
golden fixture (Dart byte-compare + pytest through the real verifier),
GPS filters and moving-time maths mirroring the PC walk rule, runs /
run_points tables (schema v9), PositionSource seam over geolocator,
location permissions and foreground service type, map/cache deps.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01WzxdWZ9CKZsNxtD5Jr9bYD
A manual workout may be dated today..today+7 days (no past dates). A future
entry is held out of log.json and earns nothing until its date; the first
sync/lock/reset pass on that day ingests and credits it like a workout logged
that day.
- ingest_manual_records defers records dated after the local day and counts
the budget (2/7d, 10/30d) against the workout's own date over logged credits
plus pending future records (a pending record never rejects itself)
- PC-logged future entries (Tk dialog, --log-manual-workout --date) go to a
durable pending ledger re-published on every push (a one-shot publish to
devices/pc is overwritten by the next push); tombstones win, entries leave
the ledger once logged or deleted
- one shared date helper (validate_workout_date) used by the dialog and CLI;
the form's date field rides in the sport row to keep the lock screen fitting
- poke start_the_day also ingests due local entries after the daily reset
- PC admission rule pinned to the phone's by the shared budget fixture
(planned.expected_by_date)
Co-Authored-By: Claude Sonnet 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_015mmTBiMpvyfLuEAXc6nDYT
The app said "manual budget exhausted for this window" (2/2w, 4/10m) while
the PC read 1/2w, 2/10m for the same records. The PC was right; two
independent Dart defects each inflated a window by one.
`_readMergedPayloads` never looked at `Record.deleted`, so the tombstoned
duplicate `manual:2026-08-21T21:00` -- the second of two logged on 08-21,
already deleted -- kept consuming a weekly slot. Suppression is collected
across the whole device union and applied after the merge, not per file:
one device tombstoning a record while another still holds it live must
delete it, and a per-file skip would just let the live copy win. This
mirrors the PC's `_tombstoned_ids`.
`countManualBudget` also counted `manual_workout:2026-08-15`, a bare
`{type, kind, date}` stub with no evidence fields. The PC refuses to ingest
those ("no workout data to ingest; skipping permanently"), so the phone was
charging budget for a workout neither device ever accepted. The filter lives
in the counter rather than the shared reader because `runnerup_verified` and
`pc_workout_verified` payloads legitimately carry no `start_time`, and the
history view must keep showing them.
Covered by a shared literal fixture both languages read, rather than two
independent round-trips: each side's own tests passed while the pair
disagreed, which is exactly how this shipped. Verified by reverting each
guard in turn and watching the twin fail.
Co-Authored-By: Claude Opus 5 (1M context) <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_01AWq9MRemX9ZWvNmnhNWUP2
