Announcements can use Piper, a free local neural TTS, instead of the browser's system
voice. No account, no paid service, and it keeps working with no internet at the track.
- npm run tts:install (or Settings -> Audio -> Install) downloads the standalone Piper
release for Linux/Windows/macOS plus Swedish and English voices into piper/ once
- server/tts.js generates each line once, one at a time, and caches it as WAV under
data/tts/; a new line takes ~0.3 s with the standalone build
- src/piper_speech.js plays it and falls back to the browser voice if Piper is missing,
errors, or takes over 2.5 s, so no announcement is lost and timing is never held up
- "Pre-generate phrases and driver names" renders every fixed phrase and name in both
languages before race day
Browser voice stays the default. Generation routes fall under the admin PIN.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Off by default. Once a PIN is set in Settings -> System & backup, every non-GET /api/*
request (plus GET /api/ammc/config) needs a signed 30-day HttpOnly cookie, which also
closes the unauthenticated /api/update/run restart trigger. A 401 opens a PIN dialog
and the pending save is retried after login. Reads stay open because the public overlay
hydrates through GET /api/state. Optional RC_ADMIN_ALLOWED_CIDRS restricts the same
routes by client address.
The PIN hash and signing secret live in data/, outside the app_state blob, so they
never leak through GET /api/state, exports or backups.
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The page had grown to ~20 stacked panels on one very long scroll - the same problem
already fixed twice this session for Settings and Event Manager. Grouped into 5 tabs
(Getting started, Create a race, Run a race, Training & results, Video guides) using
the same activeGuideTab-as-module-state approach so the selection survives an in-place
re-render (e.g. a language switch). The always-visible overview cards at the top are
unchanged.
Also investigated a "feels slow" concern raised alongside this: measured the page's
actual render cost in a real browser (median 0.6ms, on par with or faster than Settings/
Timing) and confirmed the periodic tickClock/updateSyncStatus intervals never re-render
an idle Guide page. No performance issue found - the perceived slowness was the page's
length, which this change addresses directly.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
These three UIs (all shipped earlier this session) had only been verified with throwaway
scratch scripts during development. Adds tests/browser_settings_tabs.js,
browser_event_manager_tabs.js and browser_event_schedule.js, wired into the existing
test:browser suite.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A "practice" session used as a scheduled slot in a structured
race/sponsor event's session plan silently skipped the entire
countdown/lights/start-tone ceremony and jumped straight to running,
unlike qualifying/heat/final. Narrowed the ceremony-skip check to
only the standalone "Quick Open Practice" shortcut and free_practice
(both rolling/best-lap ranked with no fixed roster) - "practice" is
already ranked like a real timed session in every other respect and
now gets the same start sequence.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Events can now be given a start time and a gap between sessions on a
new "Schedule" panel; every session's planned clock time cascades
automatically in running order (practice, qualifying, finals), and
once a session actually starts or finishes, everything after it
re-anchors to that real time - a late start or a session that ran
long (including any pause) pushes the rest of the day forward with
no manual reschedule step. The scheduled time now also shows on the
printed/PDF start lists.
Printable start lists and manually moving a driver to start last in
their own heat (for a mechanic-time request) already existed, so
this only builds the missing piece: the schedule itself.
Also fixes a real pre-existing bug in the same session-ordering logic
this feature relies on: getSessionSortWeight used `order[type] || 99`,
which silently treated open_practice's own weight of 0 as falsy and
sorted practice sessions dead last instead of first in print/PDF
output.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
getEventDrivers followed the drivers directory's internal creation
order instead of the order participants were actually selected for
the event, so the fallback qualifying/finals grid, the Start List
display, and the staggered-release announcer all read out drivers in
a scrambled order relative to how the event was actually entered.
Fixed to preserve raceConfig.driverIds' own order.
Also, a staggered heat's shared clock/auto-finish target only ever
counted the flat session duration, not the staggered release schedule
itself - so the last-released driver's own personal window could
still be running when the session force-finished. The shared clock
now widens by (driverCount - 1) * staggerGapSec on top of the base
duration for staggered sessions; each driver's own personal quota is
unchanged.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Staggered-start qualifying heats now announce each car's transponder
number in grid order (~3s apart) right after the official start,
matching what a track announcer does when releasing cars one at a
time; mass/position starts are unaffected. The Timing leaderboard and
every overlay variant now also show a green "Done" flag next to a
driver's name once their own personal timing window (which already
started at their own first crossing) has elapsed, even while the
session is still running for other drivers.
Also confirmed two things already worked correctly: finals already
always start everyone together with the shared countdown/lights/tone
sequence, and practice's time-up announcement already says "finish
the lap you are on" - only a "2 minutes remaining" cue is genuinely
missing from the existing 1min/30s/10s ladder, left as an open
question for the user.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Registering or editing a driver with a transponder already assigned
to a different driver (e.g. a wrong-class typo) previously saved
silently - the only symptom was that driver's laps never registering
during their heat, with no indication why. Now a confirm dialog names
the conflicting driver and their class before saving; the same
person's own multiple class entries sharing one physical transponder
are correctly never flagged.
Also confirms two things the user asked to verify: the race clock
already counts down for all timed session types (only open practice
counts up unbounded), and a driver not entered in a heat already has
their laps fully discarded rather than recorded - both were already
correct, no changes needed there.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The user asked whether a lone driver left in the last heat (e.g. 7
drivers at 3 per heat) could be moved into another heat manually.
Investigated first: neither "Reseed" (redistributes every not-yet-run
heat by current standings, not a targeted move) nor the grid editor
(reorders start position within one heat, never moves a driver
between heats) could do this - there was genuinely no path to it.
moveDriversIntoQualifyingHeat (src/event_race_logic.js) adds the same
"checking them elsewhere is the move" pattern already used for team
rosters: a new driver checklist on the qualification session edit
form pulls a checked driver out of whichever other heat they're
currently on, scoped to the same round only ("Q2 H2" never touches
"Q1"/"Q3", since heat composition can legitimately differ round to
round after a reseed). The checklist names the driver's current heat
before saving, mirroring the team-roster warning.
Also confirmed, per the user's earlier questions, already correct
with no changes needed: practice's per-driver "best 3 laps"/"best
5-minute result" stats, and that heats are already seeded fastest-
first from practice results. Flagged one real, deliberately unbuilt
gap while investigating: a true per-driver rolling "5 minutes from
your own first crossing" qualifying *heat* clock isn't supported -
only the practice stats work that way today.
Verified with new unit tests for moveDriversIntoQualifyingHeat and
the extracted parseQualifyingHeatName helper, and a real headless-
browser run reproducing the exact reported scenario end to end
through the actual UI.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The user asked whether the DNS fix had actually been tested against
the different points systems, not just the default one. It hadn't -
buildQualifyingStandings had never been exercised end to end with
qualifyingPointsTable "field_desc"/"ifmar" (high-points-win scoring)
or qualifyingScoring "best" (ranked by time, not points) at all,
DNS or otherwise.
Added two tests: a DNS driver scored under the high-points field_desc
table (confirms the DNS row still ranks last under an inverted sort
direction, and isn't scored as either zero or the winner's points),
and one under "best time" qualifying (confirms a bad DNS heat can't
contaminate a driver's genuine best result, since bestRank/
bestRoundMetric take the driver's single best heat, not an average).
Both passed immediately, confirming the fix already handled these
correctly - this closes a real, previously-untested gap in confidence
rather than fixing a new bug.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The user asked specifically to confirm this: a driver who DNS'd in one
qualifying heat but raced normally in others must still be fully
counted across all of them, not dropped. Traced through the fix
(buildQualifyingStandings backfills a DNS row per finished heat,
accumulating into the same per-driver entry across every heat they're
in) and confirmed it already handles this correctly - added a test
that exercises exactly this scenario (DNS in heat 1, wins heat 2 and
3) to lock it in, since this specific multi-heat aggregation path had
no prior coverage.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A thorough audit of practice/qualifying/heats/finals, bump-up,
judging and pause/timer code (requested by the user) found several
real, concrete bugs, fixed together:
- Qualifying reseed sorted heats as text ("H10" before "H2"),
scrambling grids at 10+ heats; now sorts numerically and reports a
clear warning instead of silently dropping/ghosting heats when the
participant count changes.
- A driver with zero laps in a finished heat/final (DNS) vanished
entirely - not just from that round, but silently from every future
reseeded round and the final grid. Now backfilled as a real,
last-place "DNS" row.
- Bump-up did nothing at all with "reserve bump slots" off (mains
were already seeded full, so the old capacity check always
blocked it), and a promoted driver stayed double-entered in their
old main's upcoming leg. An oversized bump count could also blow
the main-letter naming past "Z". All three fixed.
- Judging undo restored an absolute snapshot instead of reversing its
own delta - undoing two penalties out of order could leave a live
penalty behind two "undone" log entries. Now always nets correctly
regardless of undo order.
- A competitor's elapsed time didn't exclude paused time (only the
session clock did), making a paused heat's results incomparable to
an unpaused one in qualifying ranking. Finishing while paused also
dropped the final pause segment, so the displayed elapsed time kept
growing after finish. Both fixed.
Verified with new unit tests for every fix (including a new
session_control.test.js - that file had zero prior coverage) and a
real headless-browser run confirming the judging-undo and bump-up
fixes end to end through the actual UI.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Reported by the user: creating a team, adding the wrong driver, then
moving that driver to another team by ticking them on the new team's
edit form left the driver on both teams at once. Root cause: there
was never a "move" action - the edit form only ever added a driver to
the team being saved, never removed them from wherever they already
were. Live laps from a driver on two teams were then silently
credited to only the first-matching team, with no clear error.
reassignTeamMembers (event_race_logic.js) now strips a driver/car
being saved onto one team out of every other team in the same event,
and the team checkbox list shows which team a driver is already on
before you move them.
Also fixed two related gaps found while investigating, at the user's
request to fix everything: getTeamDriverPool had regressed to
ignoring the event's class/participant scoping entirely (showing
every driver in the club), and deleting a driver or car left them
behind in team rosters forever. Added a duplicate_team_members Race
Day preflight check as defense in depth for any already-corrupted
saved state.
Verified with new unit tests for reassignTeamMembers, getTeamDriverPool
and the new preflight check, plus a new permanent browser test
reproducing the user's exact reported scenario end to end.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
The user pointed out that track layout changes matter most for Open
Practice/training, not races - so the admin Training page and the
driver's own mobile profile (server/public_training_profile.js, the
actual thing most viewed after practice) needed the same layout
scoping already applied to the live "track record" audio cue.
Every stat now splits into two groups: lifetime/commitment numbers
(total laps, sessions, training days, milestone badges) still read a
driver's whole history, since a track rebuild doesn't erase past
practice. Pace/record/consistency numbers (personal best, recent-pace
trend, consistency spread, the "stable pace" control zone, stint
achievements, "best stint") now read only laps tagged with whichever
layout is currently active. Each ledger entry is tagged with
trackLayoutId at the moment it's synced from a session, so it survives
the source session later being deleted.
Verified with new unit tests covering both implementations and a real
headless-browser run confirming both the admin page and the actual
public mobile profile show the current-layout best lap (never a
faster but incomparable old-layout one) while keeping the correct
full lifetime lap count in both places.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Clubs that reconfigure their track (e.g. running indoors in the same
building) can now add named track layouts under Settings, each with
just a start date/time - layouts are contiguous, so a layout's end is
implicitly the next one's start, with no separate field to keep in
sync. Every session freezes which layout was active when it started,
so the "track record" audio cue and the Results Archive (which now
shows each session's layout) only compare laps run on the same
physical track. With no layouts configured, everything behaves
exactly as before - the feature is fully inert until a club uses it.
Design settled through conversation with the user: global scope (not
per class), start-date-only (no redundant stop date), and no changes
needed to driver/transponder mapping, since that's already fully
independent of track geometry in this app.
Deliberately out of scope for this pass: the mobile/driver training
profile's own cross-session personal-best trend stats still compare
across a driver's whole lap ledger regardless of layout - logged in
ROADMAP.md as a real remaining gap for a later pass.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Unlike Open Practice, these session types run the real countdown/light/
audio sequence before a session transitions to "running" - this was
untested beyond Open Practice's instant-start shortcut. Confirms the
Start button shows a real ticking countdown, the right audio events
fire per type (qualifying_start only for qualification), and pause/
resume/finish behave the same as they already do for Open Practice.
All audio effects/voice are silenced via raceAudio.events so the test
depends only on the sequence's own timers, not on real audio/TTS
playback completing in a headless browser with no sound hardware.
Also logs a new roadmap item: indoor clubs that reconfigure the track
layout in the same venue have no way to tell the app a layout change
happened, so PBs/track records/pace baselines silently keep comparing
lap times across incomparable layouts. No design decided yet - open
questions recorded for a future dedicated conversation.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Settings now shows when a newer version is available, with the real
version number read from the remote's package.json (via the git
binary already required for this deployment model, so it works
against any remote, not just GitHub), and an "Uppdatera nu" button
that stops the server, pulls, reinstalls dependencies and restarts
automatically - warning explicitly if a race is currently active.
The existing app-version poller already reloads the page once a
restart is detected, so no new client-side polling was needed.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
buildSimplePdf's font declared no /Encoding at all, so PDF viewers fell
back to a standard font's built-in encoding, which has no å/ä/ö glyphs
- and separately, the content stream was written as UTF-8 instead of a
single-byte encoding, so even a correctly-encoded font would receive
the wrong bytes (a multi-byte UTF-8 sequence where PDF expects exactly
one byte per character): the classic "åäö" mojibake pattern.
Fixed by declaring /Encoding /WinAnsiEncoding and writing the content
stream through a proper WinAnsi (~CP1252) byte mapper instead of UTF-8.
This is the single shared renderer behind every PDF export (race
results, start lists, heat sheets, team reports, the results archive),
so the fix applies everywhere at once, matching what was reported.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A driver completing the practice timer, "Jämna varv", or "Stegträning"
now gets a permanent history entry instead of the outcome only existing
as a one-off notification - visible on both their mobile profile and
the admin Training page, so progress over past sessions is actually
reviewable. Each completion earns a prominent emoji-badged credit.
The completion is verified server-side against the driver's real
recorded laps before being saved, reusing the same progress builders
live tracking already uses - the client can only ask for a completion
to be checked and recorded, never supply the outcome itself, so a
driver can't fabricate a completed session (and the reward tied to it)
any more than they could fabricate a lap time.
Both per-driver and club-wide goals can now target "complete N
sessions of one exercise kind" as an alternative to "N laps", sharing
the existing reward/claim mechanism and forms.
Fixes a real bug found while building this: a pre-existing two-argument
wrapper around getTrainingGoalProgress in app.js silently dropped the
new third argument, so every sessions-metric goal displayed 0 progress
regardless of how many sessions were actually completed.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Until now there was no way to find a completed session without already
knowing which event it belonged to - navigation was strictly Events ->
one event -> its sessions. A new "Resultatarkiv" view lists every
finished session across all events, searchable by event/session name,
class, driver and date range.
Opening a session reuses the existing single-session leaderboard
builder for standings and the existing per-competitor lap lookup for
lap history - no new results logic - plus a session-wide penalties/
corrections list (resultsBySession[id].adjustments), previously only
visible in the live judging action log. Export buttons reuse the
existing per-event PDF and race-package JSON export functions rather
than adding a new format.
Investigated what should happen to raw timing data when a race is
deleted. Training statistics and session replay already read from
their own independent, deliberately-retained copies that survive
deletion by existing design - unaffected either way. The one thing
that outlived deletion with no purpose was the raw `passings` SQLite
rows (write-only; nothing in the app reads them), growing forever with
driver IDs never purged. A new POST /api/passings/purge endpoint now
removes exactly the deleted session(s)' rows, called as a best-effort
side call from both delete-event and delete-session.
Also fixes a real bug found while building this: ensureSessionResult
guarded adjustments/splits/pitEvents/pitStops against being missing on
an already-existing result, but not passings itself, so viewing lap
history for such a result (e.g. an older/imported race package)
crashed outright.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
When a driver's personal 5-minute practice timer finishes, a "Pass
avslutat" card now appears immediately: how this pass's pace and
consistency compared to their recent sessions, any goal or lap-count
milestone crossed during the pass with its reward text made prominent,
and one suggested next goal (reusing the existing pace recommendation,
falling back to the next milestone when there isn't enough data yet).
Scoped to the driver's current class, and explicit about insufficient
sample sizes instead of showing a misleading number.
Adds a per-session consistency calculation to
buildPublicTrainingWindowStats server-side; everything else needed
(stats, goals, recommendation, milestone) was already sent to the
client, so the summary itself is computed there from data already in
hand - no raw laps or driver IDs added to what the public endpoint
exposes.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Investigated whether two admin browsers open at once can overwrite each
other's changes, per an explicit request to report the risk before
making bigger changes. Found that POST /api/state always replaces the
entire stored document with a client's in-memory copy, with no version
check and no per-field merge - and a browser only ever pushes, never
re-fetching server state after page load except on the Overlay view.
An idle browser's next save, even an unrelated edit, could silently
revert another browser's more recent changes, worst case a live race's
in-progress results.
Two scoped fixes instead of the bigger optimistic-locking rework (that
needs real UX decisions about conflicting saves, deferred separately):
1. server.js now remembers which session results were just written by
a real live decoder passing (/api/passings) and protects that key
for 15 seconds against being reverted by a stale full-state save,
while still applying the rest of that save normally.
2. Every non-Overlay view now quietly refreshes its data in the
background every 15 seconds (skipped while a local edit is
pending), without ever calling renderView(), so it can never
discard an open, unsaved edit form - shrinking how stale a
browser's next save can be from hours to seconds.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
A fresh Windows install reported the server failing around backups.
createBackupService created the backups subdirectory once, synchronously,
at startup with no error handling - any failure there (permissions, a
path beyond Windows' MAX_PATH, a OneDrive/cloud-sync lock, antivirus)
crashed the entire Node process before it started listening, taking
down timing with it instead of just leaving backups unavailable.
Directory creation is now retried lazily and non-fatally on each backup
attempt, and a post-backup file rename retries briefly on EBUSY/EPERM
since a freshly written file is sometimes momentarily locked by
antivirus/cloud-sync scanning on Windows.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
"Jämna varv" and "Stegträning" previously only lived on the admin
Training page. A driver's own personal mobile profile link now offers
the same self-service configure/start/stop flow already used for the
5-minute practice timer there, with live progress computed server-side
from the driver's own share token so raw lap rows and driver IDs never
reach the client.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Andra guidade träningsövningen i Open Practice: konfigurera en
stegserie (antal varv + lugnt/kontrollerat/snabbt per steg),
förvalt till klassiska pyramiden. Training-sidan visar aktuellt
steg, kvarvarande varv och en progress-pill per steg. När
sekvensen är klar jämförs de två kontrollerade stegens snitt
neutralt (t.ex. "+0.90s"), utan att kalla en försämring för
trötthet.
syncTrainingLedger sprider nu resten av det normaliserade
training-objektet istället för att lista fält manuellt, så
samma klass av bugg som drabbade Jämna varv inte kan upprepas
för framtida fält.
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Ny guidad träningsövning i Open Practice: sätt måltid, tolerans
och antal varv, se aktuell serie och dagens bästa live på
Training-sidan. Ett missat varv startar en ny serie utan att
radera bästa resultatet.
Hittade och fixade två buggar under vägen:
- syncTrainingLedger byggde om training-state vid varje sparning
utan att föra vidare det nya evenLaps-fältet, så sparade
konfigurationer försvann direkt
- Training-sidan kraschade för alla förare med färre än 3 varv i
varje stint (t.ex. de första varven i ett pass) eftersom en
dedupe-funktion läste en egenskap på ett null-värde
Co-Authored-By: Claude Sonnet 5 <noreply@anthropic.com>
Nytt browser-test för start/paus/fortsätt/finish och byte till
nästa heat i kön, mot en simulerad decoder-socket (ingen riktig
AMMC-hårdvara). Hittade och fixade två buggar under vägen:
- Open Practice på sponsor/track-event kunde inte startas trots
grönt ljus i förkontrollen (separat, striktare kontroll saknade
samma undantag)
- Race Day-vyn ritades aldrig om när decodern anslöt/kopplades ner
Testramverket hanterar nu även native alert()/confirm()-dialoger
automatiskt, annars fryser en headless körning helt.
