The relay's price_updated is the maximum price_last_updated across the grades it maps, so it cannot vouch for the selected grade: stop using it to score, gate or warn about Best Value.
- FuelPriceFreshness is now station-level and display-only: one age per station, fresh/outdated/unknown/invalid, "May be outdated" past 30 days, with future-tolerance and non-finite guards.
- Best Value restores the closest-station-selling-the-selected-fuel baseline and its savings ranking; eligibility gating, eligible-first sorting and the unavailable banner are gone.
- Price age appears in the More info sheet only - no listing-row warnings, no extra banners.
- Copy reverted from "closest eligible station" to "closest station"; stale price-confidence strings removed, "May be outdated" added.
- Dormant per-grade timestamp plumbing removed, and regression tests added for every fuel type, missing/invalid/future stamps and the bundled offline dump.
- Add FuelPriceFreshness: selected-fuel-aware eligibility plus a shared
comparison confidence, so a saving is presented as verified only when
both the candidate and the baseline pass the gate for the fuel on screen.
- Rank against the closest station with an eligible selected-fuel price
instead of the physically closest station, so a single stale neighbour
can no longer invalidate every comparison in the list. Ordinary Closest
sorting still includes every priced station.
- Keep ineligible prices visible, fall back to pump-price order, and label
Best Value unavailable when no comparison qualifies.
- Widget: same baseline and confidence path as the app.
- Tests: stale-closest baseline, alert baseline, eligibility policy,
boundary and missing-timestamp regressions.
Distances (the "Approx. 1.5 mi" report):
- Refuse junk CoreLocation fixes (invalid accuracy, or a coarse >300 m fix
within 3 km of the position already held; a coarse fix in another city is
real travel and is still accepted).
- Resolve ONE display basis per render through FuelStore.distanceDisplayOrigin,
used by rows, radius filter, sorting, savings figures and the Live Activity,
so a stray fix can no longer re-base the list outside the route cache's 600 m
ring while the routed values sit unused in the cache.
- Treat an incomplete route set as a state: retry it after 30 s rather than the
full 10-minute throttle, and repair a stale incomplete flag once every
in-radius station has a route (nothing left to route, flag stuck down).
- Widget extension resolves the same origin for fetch, radius, ordering and
labels, and keeps its own fix in fuelboard.widgetLocation instead of
overwriting the app's GPS fix.
Alerts:
- Best Value as a distinct alert mode on the existing geofence/notification
pipeline, with a persisted threshold; reverts to cheapest when the feature is
unconfigured and is not selectable while disabled.
Diagnostics (DEBUG, gated): FB_DIST_FOCUS now prints FB-DIST-ORIGIN (candidate,
accuracy, anchor, offset, resolved branch, served origin), FB_FAKE_FIX feeds a
synthetic fix through the real guard, FB_STALE_INCOMPLETE forces the stale-flag
state, and the route-fill status is surfaced in Settings behind debug mode.
Distance confidence
- Every distance surface takes the figure AND its "Approx." marker from one
FuelStore.stationDistanceDisplay call (station rows, detail sheet, widget,
Live Activity, watch, Siri), so a surface can never mark one figure approximate
while showing a routed one. Radius filters and Closest ordering test that same
displayed distance, so no row can read a distance outside the radius that
filtered it (the reported "14.6 mi inside a 10-mile list").
Accurate distances (Settings -> Distances, default off)
- Off: the browse layer serves the straight line, marked Approx., for every row,
so two installs agree to the metre and the app makes no route requests at all.
- On: routed distances for the whole search area, unmarked, served only from a
COMPLETE set anchored within 600 m — a partial fill reverts every row at once
rather than mixing road and straight-line bases.
- Best Value maths sits deliberately outside the toggle: it keeps using measured
road routes and its own confidence marking, so money figures never move with
the switch.
- Fills run only while the toggle is on and the app is frontmost; capacity raised
to 600 stations per pass; completeness tolerates pins Apple Maps has no route
for, which otherwise left an area permanently incomplete and the whole list
stuck on estimates.
Performance (the reported slowdown switching to Closest / launching with it)
- FuelStore.accurateDistancesEnabled was a UserDefaults(suiteName:) read per
call: a synchronous cfprefsd round-trip measured at ~1.9 us, ~200k calls per
Closest sort. Memoised, with the setter keeping it in step.
- Both keyed sorts decorate their keys — one distance per station, carried as an
index instead of copying whole stations into the decorated array — and the
comparison baseline plus per-station journeys now come from ONE pass instead of
two full scans per tab. The Stations tab also reuses its single sorted list for
the count and the TOP badge instead of sorting the country three times.
- Measured on iPhone SE (7,914-station Closest pool): sort 36.6 ms -> 17.1 ms per
list, Cheapest 6 ms, Best Value 11 ms.
- DEBUG probes: FB_PERF list timing, FB_PERF_BENCH toggle-read benchmark, and
FB_DIST_FOCUS now reports the served value, marker and mode (and re-prints when
a fill lands). FB_ACCURATE_DISTANCES / FB_SORT_MODE env forms for devices, where
devicectl eats a leading-dash argument.
Tests: 198, 0 failures. Device and simulator Debug builds green.
Optional Fuel Savings feature (off by default): vehicle MPG in UK imperial gallons, fill amount in pounds or litres, persisted through the existing keychain/app-group store. Ranks the WHOLE station pool by estimated net saving (fuel-price saving at each pump minus the extra return-trip driving cost) using MapKit road distances with a straight-line fallback marked (est.).
The closest station with a valid price for the selected fuel is the comparison baseline and is labelled "Closest station"; other rows read "Save £X.XX after travel" / "£X.XX worse off after travel" / "Break even". Labels live in the shared model so they are unit-tested. The TOP badge still marks the cheapest pump price and reads "CHEAPEST" when the list is ranked by net saving.
Road routing now prioritises the Worth-the-Drive candidates plus the baseline and merges cache entries instead of replacing them, so far-out stations get real routes rather than permanent straight-line estimates. Numeric fields gain a Done button above the keypad (decimalPad has no Return key); the fill-up control is hidden unless the feature is configured; Favourites shares one session fill amount with the station list. 131 tests.
Expanded leading/trailing, compact and minimal island views had
no Link, so tapping them just opened the app with no destination.
Every island region now links to the same tap destination as the
Lock Screen card (Maps vs in-app detail).
Tap hint captions follow the setting too: Tap for details vs
Tap for directions.
Small widgetURL and medium list Links route through
StationTapAction via FuelStation.tapDestinationURL: Open map
keeps the Maps URL, More info uses a fuelboard://station
deep link (stationID plus coordinate/name fallback).
Deep link carries fallback coords; ContentView presents the
detail sheet when the station is cached, else the map sheet.
Live Activity detailURL upgraded to the same rich link.
Notification body/action taps route through StationTapAction:
Open map opens Apple Maps (existing); More info presents the
in-app detail sheet (falls back to the map sheet when the
station is not in cache).
Live Activity Lock Screen card and island station row pick
their Link destination the same way: Maps URL for Open map,
fuelboard://station deep link for More info, handled in
FuelBoardApp and observed by ContentView.
Also adds the small non-interactive map preview above the
Directions button in the More info sheet.
GeometryReader as the layout root pinned content top-left, and forcing
maxHeight:.infinity over-claimed the whole proposed height, centring the
content below true vertical centre (bigger gap above) on the Lock Screen.
Demote GeometryReader to a background side-channel that only measures the
slot width for the rich-vs-compact branch; the content stays intrinsic-sized
so the system vertically centres it itself.
User: 'Remove the white background circle from the live activity'. Drop
the explicit Circle badge and render the fuel-tinted fuelpump.fill on its
own (sized up 19 -> 28 since it no longer sits inside a disc).
fuelpump.circle.fill's first hierarchical style hits the PUMP, not the
circle, so the earlier darkened colour landed on the pump and the circle
stayed white (user: 'I didn't want the fuel pump darker I wanted the
white circle darker'). Rebuild the brand glyph from two explicit shapes:
a light-grey (.86) Circle badge + a bright fuel-tinted fuelpump.fill, so
circle vs pump colours are unambiguous and the badge centres cleanly with
the two-line text block.
Add a DEBUG-only -qaLiveActivity <fuel> launch hook that starts a Live
Activity with a long station name so the Lock Screen can be rendered in
the Simulator and verified by pixels (pump bright green 43,193,81; badge
grey 202,201,202).
A GeometryReader hands its content the full proposed frame and pins it
top-left, so the Lock Screen card sat high instead of vertically centred
when the activity container offered extra height. Wrap the branch in
.frame(maxWidth:.infinity, maxHeight:.infinity, alignment:.center).
Also darken the pump circle to 82% of the raw fuel tint (mix 18% black) so
the white pump symbol keeps contrast against bright Lock Screen/card, esp.
for the light yellow premium / cyan diesel palette colours.
Move FuelType.tintColor (green=unleaded #30D158, yellow=premium #FFD60A,
cyan=diesel #64D2FF) into Shared/FuelStore.swift so the app, widget, and
Live Activity all read one definition (was app-only in StationsView, so the
widget couldn't see it). Tint the Live Activity rich card pump icon and the
Dynamic Island compactLeading pump by context.state.fuel.tintColor.
ViewThatFits measures each child's IDEAL (untruncated) width to decide
fit, and a flexible .frame(maxWidth:.infinity) middle column does not cap
that ideal — so a long station name (e.g. 5-mi Phoenix Filling Stations)
inflated richBody's ideal past the iPhone Lock Screen width and fell back
to the compact strip. Replace the ViewThatFits branch with a
GeometryReader that reads the ACTUAL proposed width and shows the full
three-column card at/above 280pt, compact below. Content-independent, so a
long name can no longer collapse the iPhone card.
The compact CarPlay/Watch strip's second line had escaped interpolation
(\\( instead of \\() so it rendered the literal placeholder text
"context.state.stationName · context.state.distanceText" instead of the
actual station name and distance. Normalize to single-backslash interpolation.
Restore .supplementalActivityFamilies([.small]) + a ViewThatFits compact
fallback so CarPlay's small slot renders our shared compact strip (fuel type,
bold price, station · distance) instead of falling back to the Dynamic Island
compact closure, which showed the app name and omitted the distance the user
wants. Full iPhone/iPad card is preserved via the richMinWidth gate on
richBody plus its flexible, truncating middle column (long names can no longer
collapse it).
Long station names inflated richBody's IDEAL width, so ViewThatFits (which
measures ideal, untruncated text) decided 'rich doesn't fit' and fell back to
the compact strip — that's why 5-mi / long-named activities rendered small
while 10-15-mi / short names stayed full. With .supplementalActivityFamilies
already removed, the body is only ever Lock-Screen width, so drop ViewThatFits
and compactBody entirely and always render the full three-column card; the
station caption line-limits + scale-downs + tail-truncates in place.
The active activity was rendering the narrow .small form on the iPhone Lock
Screen (squeezed, small text) because .supplementalActivityFamilies([.small])
made iOS eligible to present it that way. Remove it so iPhone/iPad always get
the full three-column card; CarPlay's small form still comes from the Dynamic
Island compact closures. Keep compactBody as a narrow-width safety fallback.
The adaptive layout picks the full three-column view when there's Lock
Screen width and a compact price strip in the small CarPlay/Watch slot.
Tighten the discriminant (rich ≥290pt, compact ≤220pt) so iPhone/iPad
always get the full view and only genuinely small slots get the compact
strip. Long station names used to clip at inconsistent points across
the rich/compact/Dynamic Island views; both bodies now scale + tail-
truncate so the caption stays consistent regardless of station name.
- Move widget entry view into Shared/FuelPriceWidgetViews.swift (FuelPriceWidgetContent,
FuelPriceWidgetEntryData) so the app can render the REAL widget faces; widget
extension now wraps it (FuelPriceEntry.data) — one source of truth for the face
- WidgetMockScreen (-widgets launch arg): app icon + small + medium widget previews
for App Store captures; AppIconPreview image set (1024 drip master)
- AppStoreScreenshots: 5-8 dark-mode tab captures with real Halifax data, 9 widgets
- 95/95 tests green; Release IPA rebuilt + re-served
The LAN relay is unreachable from any phone outside the developer's LAN, so
for App Store users a relay fallback was dead weight AND harmful: attempting
the private IP fires the iOS Local Network prompt on a stranger's phone and
adds a timeout before the bundled dump. GitHub raw is reachable anywhere the
relay would be, and more reliably.
- LiveChainProvider: GitHub mirror first EVERYWHERE (focused alert path no
longer relay-first); LAN relay only joins behind the hidden dev flag
fuelboard.relayFallback (Settings → Debug, keychain-first, off by default)
- MirrorFuelProvider: focused() projection replaces the relay's server-side
radius endpoint (fuel filter + within-radius + nearest-first + limit 500)
— the alert path re-checks radius/fuel itself, so behaviour is identical
- Widget: fetchFocused/fetchFuelOnly now GitHub-first via the chain (day-cache
shared through the app group); relay + widget-diag GET gated on the same
dev flag — off-LAN widgets no longer attempt local network at all
- FuelBeacon: fires only when the dev flag is on — consumers make zero
local-network attempts; /stats app-hit spike = GitHub down + dev flag on
- Onboarding: Local Network machinery removed entirely (no probe, no prompt,
no 12 s stall) — data page is now informational: prices download from the
internet, no permissions needed. Replay on cellular can no longer hang.
- Info.plist: NSLocalNetworkUsageDescription copy → optional dev-only wording
- 95 tests (6 new: relay-skip-when-disabled x2, focused prefer-mirror, mirror
focused projection x3, beacon dev-flag gate)
The iOS 18+ long-press app-icon menu (turn icon into a widget) shows
the supported SIZES of the FIRST-registered widget kind only — it
does not list widget kinds and is not keyed by intent type (two
distinct intents in 6375af4 changed nothing on-device). The bundle's
first kind (FuelPriceSmallWidget, small-first since 38b47ee) therefore
decided the menu offered only the small tile.
The early build worked because FuelPriceWidget was registered first
back then. Reordering to medium-first would fix the menu but flip the
gallery order (38b47ee). Instead: this first kind now supports BOTH
.systemSmall and .systemMedium, so the icon menu offers both size
tiles (ChatGPT/Safari behave the same way — their first kind supports
several families). The gallery still lists both widget kinds in bundle
order; the Small entry just gains a medium variant. The shared view
already branches on family, so medium renders the station list.
iOS keys the long-press app-icon contextual menu by configuration-intent
TYPE, so sharing one intent across kinds collapsed the menu to the
first-registered kind (small) — the medium 'FuelBoard Prices' vanished
from the icon menu while the + gallery (kind-keyed) still listed both.
dafc96f's skeleton fix was never the problem; the extra favourite
AppEntity params were.
FIX: FuelBoardSmallWidgetConfigurationIntent — a distinct intent TYPE
with the IDENTICAL minimal parameter set (fuel/sort/distance only, no
AppEntity params). Fresh-widget default-config resolution stays intact;
the appex metadata now registers two WidgetConfiguration actions.
Verified: build green, 45/45 tests, extract.actionsdata lists both
intents, fresh IPA packaged + signed per component.
The pump glyph in the widget header (both the small single-station face and
the medium list face) was hardcoded green; it now uses the app's fuel
colour wheel — green #30D158 unleaded, yellow #FFD60A premium, cyan
#64D2FF diesel — mirroring the per-fuel pumps in the app's segmented
picker. Palette duplicated in the widget target (Shared is Foundation-only
by design); cross-referenced to StationsView.swift's FuelType.tintColor.
ROOT CAUSE (proven by A/B test on-device): the small widget's custom
intent (FuelBoardSmallWidgetConfigurationIntent with favouriteFuel +
favourite AppEntity params) fails fresh-widget default-config resolution
at the system level — the whole default config fails, no timeline is
ever requested, makeEntry never runs, widget stays on the skeleton.
A minimal small widget sharing the MEDIUM intent (fuel/sort/distance
only) populated immediately in the same extension on the same device.
Neither removing defaultResult() (4d35b08) nor the
@IntentParameterDependency (8a9e775) helped — the extra entity params
themselves break resolution.
FIX: FuelPriceSmallWidget now uses FuelBoardWidgetConfigurationIntent
exactly like the working test widget. FuelBoardSmallWidgetConfiguration
Intent, FavouriteFuel(+Query), WidgetFavourite(+Query) deleted; the
pinned-favourite picker is gone (approved tradeoff). Favourites sort
still works on both faces via the list branch — the small face renders
the cheapest favourite of the chosen fuel. Beacon tag distinguishes the
small widget in diagnostics; test widget removed, gallery order
restored (small, medium, live activity).
Temporary FuelBoardTestSmallWidget kind: a small-face widget driven by
the PROVEN medium intent (fuel/sort/distance only — no favourite params,
no dependency, no custom entity query), registered first in the gallery.
Isolates small SIZE vs small INTENT: if it populates while
FuelPriceSmallWidget skeletons, the small intent's extra machinery is the
culprit; if it also skeletons, the failure is small-family/system-level.
Provider gains a beaconTag so the test widget's beacons are
distinguishable in the relay log and keychain slots. Removed once the
root cause is confirmed.
Final structural difference from the working medium widget. The
dependent-picker machinery resolves the favourite parameter through the
hidden favouriteFuel parameter during fresh-widget default-config
resolution (default Sort = Cheapest hides favouriteFuel), and that
resolution fails on-device — the whole default config fails and the
small widget strands on the skeleton. Favourite picker now lists ALL
favourites, each row labelled with its fuel (the fuel-scoped filter was
the dependency's only purpose). Configured Favourites-mode widgets are
unaffected. Per-intent beacons remain in place to verify the outcome.
The single-slot beacon couldn't distinguish small vs medium (last writer
wins). Beacon keychain service now embeds the intent type name so small
and medium never overwrite each other; the relay GET now carries the
intent type + entry state as query params so the access log shows exactly
which widget kind ran makeEntry, when, and with what. Diagnostics screen
shows SMALL and MEDIUM beacons side by side — a missing SMALL beacon
after adding the small widget proves its timeline never reaches the
provider (system-level config resolution failure).
The small-widget skeleton survived four resolution fixes; instrument the
widget instead of guessing. Every makeEntry now writes its entry state
(intent type, location source, station count, fuel, sort, favourites flag,
first station) to a keychain beacon (app-readable) and fire-and-forgets a
GET to the relay's /api/v1/widget-diag so its access log records whether a
widget timeline actually runs in the extension and what it produced.
Settings → Debug gains a Widget Diagnostics section: the keychain beacon +
the installed-widget inventory from WidgetCenter. Relay gains the
/api/v1/widget-diag route (204, log-only).
The small-widget skeleton (fresh widget stranded on the placeholder while
the medium populated in the gallery) traced to the small intent's default
configuration resolution: FavouriteFuelQuery/WidgetFavouriteQuery
defaultResult() read keychain in the gallery-preview/fresh-add path,
failing the whole default config and leaving no timeline. Configured
widgets bypassed defaultResult (stored values → lenient entities) — the
exact favourites-works/cheapest-closest-doesn't split.
- Remove defaultResult() from both queries: fresh configs now resolve
from the static @Parameter(default:) values through lenient
entities(for:) with ZERO storage reads.
- WidgetFavouriteQuery.entities(for:) skips the keychain name lookup for
id-only (empty station) values.
- Auto-populate moves to makeEntry: empty pinned favourite → first
favourite of the fuel (same rendered result as the old sheet default).
- Sheet keeps the dependent picker via suggestedEntities() + dependency;
fresh widgets show static defaults until picked (approved tradeoff).
