Tier 4 visual fidelity, audit 9.1/9.5/9.6/9.8 (t4 lines 411-428):
- 9.1: new _HaloPulsante renders the prototype's 420px amber radial
gradient behind the hero content. Deliberately BOUNDED (one
grow-and-settle cycle), not the prototype's literal `infinite` CSS
animation: this screen's protected dismiss-guard test calls
pumpAndSettle() after every mount/interaction, and a genuinely
infinite AnimationController.repeat() would hang it forever with no
way to fix it. Skips its Animate wrapper under reduced motion, same
as every other entry animation in this app.
- 9.5: the hero time gets a local letterSpacing/height override
(-4/0.95) instead of the shared heroTime token's -2.0/1.0 — the
alarm editor's hour block is the token's other consumer and still
wants height:1, so the shared style is untouched.
- 9.6: ringing-screen art grows from 168/radiusLg(30) to 180/36 (t4
line 421), a local constant like the existing stop-button radius.
- 9.8: the POSPONER eyebrow regains its 19px warmCoral snooze icon.
All 5 protected files (including the dismiss-guard test) remain
empty-diff against main.
Tier 4 visual fidelity, audit 7.5 (t4 line 334): the prototype stacks
alarm cards with a 10px gap; the build used 12. Each gap is now keyed
per alarm id so a guard test can pin the exact value between two known
cards.
Tier 4 visual fidelity, audit 4.5 (t4 line 235): the border and the
label/icon now use the prototype's two DIFFERENT opacities
(rgba(255,255,255,.16) stroke vs rgba(242,247,250,.6) text/icon,
previously one shared 50% colour for both), padding is a uniform 14
instead of symmetric(18,16), and the icon is a plain add glyph at 20px
instead of add_circle_outline_rounded.
Tier 4 visual fidelity, audit 1.11/1.12 (t4 line 80): a pill badge next
to "Tus emisoras" now shows the TOTAL favorite count (not the 8-capped
grid size), and "Ver todas" is plain 12px/w800 brand-teal text instead
of a Material TextButton with its own padding and splash.
Item 1.4 (hero art radius) was already fixed as a side effect of an
earlier commit — no change needed here.
Tier 4 visual fidelity, audit S10/10.6/10.8/10.9 (t4 lines 514-516):
GrupoAjustes' row divider now indents 47px instead of running
full-bleed, FilaAjuste's leading icon drops from Material's 24px
default to 21px, the row title is a local 14px override of cardTitle
(14.5), and the chevron shrinks to 19px at 40% opacity instead of the
24px full-opacity default.
Audit 2.4 and 2.6: the quality row sat right after the subtitle, four
positions before where the prototype puts it (t4:114-140), and the play
button was 72 with a 40 icon instead of 78/42 (t4:127).
Item 24 / audit 5.1-5.3, 5.5 (t4:255-269): Paises' "Tus idiomas" was
a Wrap of non-interactive Chips, and the full list was a plain
ListTile with no ISO column and no onTap. Both now share one tappable
row (ISO code, name, station count, chevron); the first "Tus idiomas"
row gets the prototype's teal-tinted highlight. The one production
call site wires the tap to filter Buscar by that country's code and
pop back -- EstadoBusqueda.buscar(pais: ...) already accepts any ISO
alpha-2 code, not just the ~10 presets in the filter sheet.
Item 25 / audit 13.2 (t4:643-646): a reconnect card (rotating ring,
station name, "Reconectando...", a stop affordance) replaces the
complete absence of any reconnect signal outside a word in the mini
player. Ships WITHOUT the prototype's attempt counter: the only live
ControladorReconexion instance is a private field of
PluriWaveAudioHandler inside servicio_audio.dart, a file this task
requires stay byte-identical to main, and nothing else re-exposes it.
Reconstructing a count from estadoStream's reconectando emissions
would not be faithful (the stream can emit it many times per actual
backoff attempt), so that was deliberately not attempted.
Item 23 / audit 4.3, 12.4 (t4:226-232, 616-619): Favoritos and
Grabaciones rows were full glass cards / ListTiles with two stacked
buttons and no artwork slot. Replace with flat, background-less rows
via a new shared FilaEmisoraPlana widget (square art, name+meta, a
circular play affordance) plus a bespoke Grabaciones row (44x12
placeholder art -- recordings carry no per-station favicon, so this
is a themed fallback, not invented artwork).
Favoritos keeps "Move to list" / "Remove from favorites" behind an
overflow menu (same underlying methods, unchanged) instead of two
always-visible buttons, since dropping either would be a functional
regression the prototype's own row doesn't have to solve for.
Also 12.1 (t4:610): the Grabaciones header action is folder_open, not
a generic gear.
Item 22 / audit 3.6 (t4:184-188): the mini player was a floating
999-radius glass pill with no artwork. Replace it with a 60px opaque
bar (listSurface at .97 alpha), full-bleed edge to edge, showing the
station's square 42x42 artwork instead of the abstract playing-bars
indicator. app.dart no longer wraps the bar in the balloon nav's own
8px side margin, so it now spans the full width.
MiniReproductor.altura is re-measured (72 -> 60) now that the bar's
content height is fixed by construction; PluriLayout.bottomChromeInset
derives from it as before. Both the S3-R3 configurarLocalizaciones
guard and the altura measurement test still pass unmodified.
Item 21 / audit 9b.4-9b.5 (t4:451-462, 469-479): the active vacation
range showed a LinearProgressIndicator the prototype never draws.
Replace it with the screen's real signature element -- a start/end
date pair (day+month, weekday, connector rule) -- reused with a flat
connector for the "programados"/"pasados" rows, replacing their plain
ListTiles with proper cards.
Audit 8.1-8.4: opaque bottom-anchored sheet with a grab handle, a framed
time card and 7 circular day buttons.
The date field, fallback-station picker and sound dropdown all stay
reachable -- the prototype omits them, but presentation changes never
remove capability.
Audit 11.10/11.8 (t4 lines 570-602): the prototype draws the explainer
banner, the band sliders, "Salida activa + Guardar como preset +
Restablecer a plano" and "Emisoras con ajuste propio" as 4 separate
cards. The build nested everything in one outer PluriGlassSurface, so
it read as a single merged block, and had no reset-to-flat action at
all.
Removes the outer wrapper from _CuerpoEcualizador. EcualizadorWidget
already draws its own card (design ADR-5, unchanged); the explainer
banner already has its own tinted background (unchanged); "Emisoras
con ajuste propio" now wraps itself in its own PluriGlassSurface; a
new _TarjetaSalidaYAcciones groups Salida activa with two tappable
rows — "Guardar como preset" (relocated from a floating end-aligned
button) and the new "Restablecer a plano", which applies
PresetEcualizador.flat via the existing EstadoEcualizador.cambiarPreset
— the same call the "Plano" preset chip already makes, not a new
capability.
The 5-slider band count, servicio_ecualizador.dart and
preset_ecualizador.dart are untouched.
Audit 3.2 (t4 lines 162-171): the prototype groups Países, Géneros,
Tendencias and the entirely-missing Novedades into one 2x2 entry-point
grid. The build had them as three unrelated always-visible widgets
(a Países ListTile, a Géneros FilterChip Wrap, a Tendencias ActionChip
strip) and no Novedades entry at all.
Replaces all three with a single grid section, capability-preserving:
Países still pushes PantallaPaises; Géneros and Tendencias now open
their exact existing content in a picker sheet instead of always-on-
screen (Géneros auto-closes on selection, matching this screen's other
single-choice filter sheets; Tendencias stays open, a browse list, not
a filter). Novedades re-triggers the existing discovery refresh, since
no distinct "new stations" feed exists anywhere in the domain.
New ARB key set (exploreByTitle/exploreTrendingTitle/
exploreTrendingSubtitle/exploreNewTitle/exploreNewSubtitle) translated
across all 13 locales with zero anti-copy allowlist entries needed.
Also fixes a pre-existing PluriEmptyState overflow this restructure
exposed (unrelated to the grid itself, confirmed via isolated repro):
wraps its Column in a SingleChildScrollView so a too-tall title/
subtitle scrolls instead of throwing a hard RenderFlex overflow.
Audit 1.10 (t4 lines 82-88): the prototype shows a 2x2+ grid of 8
favorite stations; the build was a 260px-wide horizontal strip capped
at 6.
Raises the cap to 8 and replaces the horizontal ListView with a
GridView.builder (2 columns, gap 10). Adds a dedicated _CeldaTusEmisoras
cell (44px square art radius 11, name + genre) rather than reusing
TarjetaEmisora(esCompacta: true), since that widget always renders a
favorite button and a live badge that this prototype cell never draws.
Audit 1.7/2.5 (t4 lines 66-68, 120-122): the prototype draws 30
discrete bottom-anchored bars (radius 2, vertical gradient); the
build's VisualizadorAudio painted a single continuous oscilloscope
stroke, so its `barras: 30` parameter never produced bars.
Adds an opt-in `barrasDiscretas` mode (default false, byte-identical
continuous rendering preserved for any other caller) plus a
`gradienteFinAlpha` knob for the two screens' differing gradient
end-alpha (.3 vs .45). Wires it into the Escuchar hero and the full
player, correcting the player's bar count/height to match the
prototype (26->30 bars, 46->40px) at the same time.
Audit 14.2 (t4 lines 693/706): content was top-anchored in a plain
SingleChildScrollView; the prototype anchors it to the bottom with two
flexible spacers, so the CTA sits at the screen edge. Restructured as
a scrollable Expanded region (safe on short screens / large text
scale) with the CTA pinned below it, always at the bottom.
Audit 14.1 (t4 lines 687-689): added the full-bleed blurred banner
backdrop, reusing aurora_wave_banner.png (the same asset the deleted
PluriScreenHeader used) since this app has no bundled equivalent of
the prototype's mockup-only banner.jpg.
No monetization content touched (PRO pill, "14 días", pricing card,
free-version link all remain absent per the binding no-monetization
decision).
Audit 13.1 (t4 line 641): the banner rendered wifi_off in
colorScheme.error (red) inside a generic glass card. Rebuilt using
offlineAccent (#E8879A) — a token that existed since Tier 1 with no
consumer — for the tint, border and icon, plus the prototype's title/
detail layout and a restyled Retry chip. New offlineBannerTitle key
translated to all 13 locales.
Audit 1.6 (t4 line 62): the Escuchar hero never rendered "genre ·
country · bitrate kbps" between the station name and the visualizer.
Built from fields Emisora already carries (tags/pais/bitrate) — no
new fields, no service calls — omitting gracefully whatever a station
lacks.
Audit 9.3 (t4 line 415): the ringing screen never showed its schedule
pill. Reuses the alarmScheduleOnce/alarmScheduleWeekdays ARB keys that
already existed with no consumer, plus a new alarmScheduleDaily key
(translated to all 13 locales) for the recurring-alarm case shown in
the prototype.
Audit 9.7 (t4 line 423): the station name was cardTitle (14.5px/w700),
5.5px and 100 weight units under the prototype's 20px/w800.
Audit 9.9 (snooze tiles as number-over-unit) is intentionally NOT
included in this commit: it would require the tile to stop rendering
as a single flat "{minutes} min" Text node, which is exactly what
pantalla_alarma_sonando_dismiss_guard_test.dart taps via
find.text(l10n.alarmSnoozeOptionLabel(N)) in four places. That file
must stay untouched, so this restyle is deferred pending a decision
on how to restructure the tap target safely.
Audit 9.11 (t4 line 434): the Stop button was a FilledButton in
colorScheme.primary (brand cyan), the wrong colour family for the
largest element on the ringing screen. The prototype draws a neutral
rgba(255,255,255,.08) surface with a rgba(255,255,255,.16) border and
radius 24 (none of PluriWaveTokens' three named radii).
Audit 1.3: the hero art shipped at 84 where the prototype draws 132
(t4 line 56), with a 24 corner radius rather than the shared radiusMd.
Adds a dimension guard. Every existing test on this screen asserts
behaviour only, which is why an artwork 36% too small went unnoticed.
Audit 11.4 and 11.6: the band column was 152 instead of 280 (t4 line
581), leaving the sliders 46% short, and the fill read liveGreen -- the
LIVE badge colour -- instead of brand teal (t4 line 585).
Adds two guards. The existing tests only asserted the band COUNT, which
is exactly why both values could drift unnoticed.
PluriScreenHeader was a 38-radius glass hero: an aurora banner at 24%
opacity, a black-to-transparent scrim, two radial orbs, a 120px app-mark
watermark and a 56px tri-gradient glyph badge. None of it is in the
prototype (t4) -- every root's title is plain text in its own 56px
row, which PluriRootHeader (S1) already provides. Delete the class
(and its now-orphaned _Orb helper) and its four call sites (Buscar,
Favoritos x2, Alarmas, Ajustes landed with S8).
Two of those call sites carried the hero's only functional bit besides
the title, so PluriRootHeader gains an `actions` slot (rendered before
the shared bedtime button) to keep them reachable:
- Alarmas' create-alarm button (FilledButton.tonalIcon, unchanged
shape, just relocated)
- Buscar's filters entry point (the same tappable PluriStatusPill,
just relocated)
Every other retired trailing pill (Ajustes' "Secure" status,
Favoritos' collection-count badges, Alarmas' alarm-count badge) was
purely decorative and matches the prototype by simply disappearing.
S2, Tier 1 visual-fidelity pass (audit id 2521).
The prototype runs three horizontal padding tiers (t4): 20px for
section titles/eyebrows (lines 153, 254, 299, 511), 16px for cards
(lines 327, 512, 610), 12px for background-less list rows (lines 174,
226, 301). The build had collapsed all three into a single
PluriLayout.horizontal = 16, used everywhere regardless of context.
Add PluriLayout.titleHorizontal (20) and PluriLayout.rowHorizontal
(12) alongside the existing `horizontal` (16, unchanged -- it already
covers the card tier). horizontal keeps every one of its ~30 existing
call sites unchanged.
Committed ahead of S2 (item 6 in this pass) because that item's
PluriRootHeader edit reuses titleHorizontal/rowHorizontal for the
header's own padding, matching the prototype's own header spec exactly
(e.g. Alarmas padding:0 12px 0 20px) -- a real dependency, not just
numbering.
S5, Tier 1 visual-fidelity pass (audit id 2521).
The prototype puts a trailing current-value string on nearly every
settings row (t4 lines 512-539, 625 -- "3 guardados", "Alfabetico",
"Espanol", "7 . 84 MB"). FilaAjuste only accepted icon/titulo/onTap,
so every row was value-blind.
Add an optional `valor` slot to FilaAjuste (13px, rgba(242,247,250,.55),
rendered before the chevron). Wire 8 of the 12 built rows to state
already available at the settings root: equalizer on/off, sleep-timer
active, favourite-group count, preferred station name, custom-station
count, sort order, recordings count-and-size (FutureBuilder over
EstadoGrabacion.listarGrabaciones), and the current language (hoisted
pantalla_ajustes_idioma.dart's native-name list to module level so the
root can read it without duplicating it). Salida de audio, Musica
local, Backup and Info's version are left without a value -- each
lacks a low-risk, deterministically-testable data source (see the
apply-progress note for the reason per row).
Reading EstadoRadio for these values through a root `context.watch`
would rebuild the whole settings list -- including the Grabaciones
FutureBuilder's disk read -- on every unrelated audio notification;
this follows the codebase's existing S4-R5 convention of narrow
`context.select` per field instead.
Ajustes' own PluriRootHeader/PluriScreenHeader edit (S2 in this same
pass) landed in this commit too, since both touched the same header
block in pantalla_ajustes.dart at the same time.
S8, Tier 1 visual-fidelity pass (audit id 2521).
The prototype's eyebrows are always rgba(242,247,250,.42) (t4 lines
254, 299, 381, 450, 511, 660). eyebrowLabel set size/weight/letter-
spacing only, no colour, so every call site (settings group headers,
vacation section titles, the ringing screen's snooze label) rendered
at whatever full-opacity default the ambient text theme resolved to.
One factory-level change fixes every current and future consumer of
PluriWaveTypography.eyebrowLabel — no call site needs editing.
S9, Tier 1 visual-fidelity pass (audit id 2521).
The prototype's station thumbnail is a plain 44-48px square, radius
11/12, with no ring or glow (t4 lines 84, 175, 227, 302, 614).
TarjetaEmisora's compact row variant instead painted a 58x58 circle
wrapped in a SweepGradient ring (magenta/cyan/coral) and a 22-blur
glow behind a 50x50 ClipRRect(18).
Replace it with a 48x48 ClipRRect(12) square. Update the loading
shimmer placeholder to match (was also a 58x58 circle block).
S6, Tier 1 visual-fidelity pass (audit id 2521).
The prototype states an explicit system rule (t4 line 40): "opaque list
surface #102532, glass only in the chrome and in the active card."
PluriGlassSurface backs nearly every card/row in the app, but always
rendered translucent + blurred (glassSurface, blur 18) — the listSurface
token (#102532) existed since WU1 but was only ever used at reduced
alpha, never opaquely.
Add a `glass` flag to PluriGlassSurface, defaulting to false: an opaque
listSurface fill with no BackdropFilter. Chrome (MiniReproductor) and
the active/now-playing card (Escuchar's hero, when a station is
playing) opt into the old translucent look via `glass: true` — every
other of the ~26 call sites needs no change and now renders opaquely.
Also fix the card radius: the prototype's dominant card radius is 18
(t4 lines 512, 613, 715, 133); radiusMd was 22, a systematic +4px drift
across every card that defaults to it. Retune TarjetaEmisora's own
radius ternary so its compact (row) variant uses the dominant row
radius, 14, instead of inheriting the card radius.
S3+S4, Tier 1 visual-fidelity pass (audit id 2521).
The prototype (t4) draws no global app bar anywhere: every root paints
a plain ~56px title row inside its own content instead (Alarmas line
325, Ajustes line 511, Explorar line 641). app.dart wrapped every tab
in PluriWaveScaffold(appBar: AppBar(title: Text(appTitle), ...)),
adding 56dp of chrome and a "PluriWave" title the prototype never
shows.
Add PluriRootHeader, a shared 56px title-row widget reused by all 5
roots. Extract app.dart's old _mostrarTimerDialog (only reachable from
the removed AppBar action) into a free function,
showPluriSleepTimerSheet, so every root's header can open the same
sheet directly and the sleep-timer feature stays reachable from every
tab with no behaviour change.
S1, Tier 1 visual-fidelity pass (audit id 2521).
The functional redesign never touched pluri_bottom_navigation.dart, so the
bar kept the old glass + magenta/coral language while every other surface
moved on. Rebuilds it to prototype t4/4a: a 52px opaque #0A1B24 pill with a
110x74 balloon raised over the active tab, a teal radial glow behind it, the
active item lifted 15px with its label, and inactive items reduced to a 23px
icon at 46% opacity.
The five tabs and their PluriIconGlyph icons are unchanged, per the standing
decision — only the bar's shape, colour and behaviour move.
altura and PluriLayout.bottomChromeInset are both asserted against the real
laid-out height rather than hardcoded guesses.
The CI Flutter SDK predates v3.41 and only exposes ReorderableListView's
onReorder; the newer onReorderItem broke the build with three analyzer
errors. onReorder exists in both SDKs, so use it and compensate for its
pre-removal newIndex internally.
Fixing the call site surfaced a real ordering bug: _onReorder located the
target neighbour in the untrimmed global list, while ServicioFavoritos
.reordenar inserts into the list after the station is removed. Dragging a
station downwards past its neighbours therefore landed it one slot too far.
Adds a mid-list downward-drag test, the only case that separates the two
coordinate spaces.
Adopts the approved Claude Design handoff across every screen while keeping
the existing five-tab navigation and icon set.
- Design tokens, named type scale and a shared push-chrome scaffold
- Settings split from 1897 lines into a 198-line root plus 12 detail screens
- Escuchar gains an embedded player and a favorites grid
- Buscar gains the discovery landing state, filter pills and client-side sort
- New screens: countries browser, vacation manager, recordings library, welcome
- Equalizer restyled at 5 bands with custom presets; full player restructured
- Alarms, ringing screen and connectivity banners restyled
- All 466 message keys translated across 13 locales
Tests: 530 -> 747. Analyzer unchanged from baseline.
Pre-existing bug, reproduces identically before this branch's changes:
_editarGrupo (pantalla_ajustes_grupos_favoritos.dart) and
_editarTamanoMaximo (pantalla_ajustes_grabaciones.dart) each created a
TextEditingController, awaited showModalBottomSheet, then disposed the
controller immediately on resolve - racing the sheet's own close
animation, which still holds a bound TextField for a couple more
frames. Manifests as "A TextEditingController was used after being
disposed" plus a couple of cascading framework-internal symptoms.
Fix: extract each sheet's content into its own StatefulWidget
(_HojaEditarGrupo, _HojaTamanoMaximo) that owns the controller in its
own State. Flutter only calls State.dispose() once the widget is
actually removed from the tree, i.e. after the close animation
finishes, so there is no dispose-timing decision left for the caller
to get wrong.
_editarGrupo's own test previously suppressed the crash via a
FlutterError.onError override instead of fixing it; that suppression
is removed here. _editarTamanoMaximo had no coverage at all for this
interaction; added it. Strict TDD: confirmed both sites fail without
the fix (RED) before applying it (GREEN).
Build the first-run welcome surface from mockup screen 14, stripped of
its entire monetization block: no PRO pill, no "14 dias PRO gratis"
trial line, no pricing card, no secondary "free version" link. Ships
only logo, headline, body copy, exactly 3 feature bullets, and the
single "Empezar a escuchar" CTA, as a full-screen route (not a modal).
Spanish copy is re-cast from the mockup's "tu" form to the app's
established voseo register (matching ~750 existing app_es.arb lines),
using "auto" instead of "coche" per the one existing precedent. The 3
bullet icons reuse existing tokens (electricMagenta/liveGreen/
warmCoral) that already match the mockup's own hex values for them.
CTA switches to the Escuchar tab and pops the route. Wiring this
screen into the real first-launch flow (main.dart/app.dart) is left
for a follow-up unit, same shape as WU15/WU15b - this WU only covers
the isolated, tested screen per its own task list and verify command.
WU17.
Tint the mini player's reconnecting/error sub-states with the
offlineAccent token (added in WU1, unused until now): the status
label, the reconnect spinner, and the error retry icon now read as
visually distinct "connectivity trouble" states instead of blending
into the ordinary loading/paused look. Plain buffering keeps the
default colour, confirmed by a dedicated regression test.
Verify-first gate (task 16.1): ControladorReconexion.intentos exists,
but ServicioAudio never surfaces it past a debug log line, and its
estadoStream only carries the EstadoReproduccion enum. Adding an
attempt-count label would require a getter/stream on
servicio_audio.dart, one of the files this change must keep at an
empty diff against main. Ship the restyle without the counter, per
the risk register's own fallback.
WU16.
Restructure pantalla_reproductor.dart onto PluriPushScaffold (design
ADR-2 - this screen is the documented single consumer of titleOverride,
a centered live/not-playing status pill, and the non-default
keyboard_arrow_down leadingIcon). Square art replaces the old circular
hero, favorite moves from the AppBar into the transport row (the
redundant live-indicator dot is dropped - the AppBar pill already covers
that signal), the old separate info chips collapse into a single
subtitle line, and a new quality row surfaces codec/bitrate with a
"Cambiar" action that reconnects the current stream (this app has no
per-station alternate-quality capability to invoke, so this reuses the
same reproducir() call the existing error-state retry button already
uses, rather than a dead button or an invented picker).
The always-expanded recording panel and the standalone sleep-timer
button both become tool-tray tiles (EQ propio / Grabar / sleep timer /
Compartir), each opening its own bottom sheet. "EQ propio" opens a sheet
hosting EcualizadorWidget - the exact same component WU13 restyled for
Settings, bound via the existing presetParaEmisora/guardarPresetPorEmisora
per-station persistence path. No second editor was created; the
multi-device-eq resolution hierarchy is untouched.
pantalla_reproductor.dart had zero test coverage before this commit (907
lines) - writing it first surfaced two pre-existing bugs blocking any
coverage at all, both fixed: initState called estado.reproducir()
directly, which notifies listeners synchronously before its first await
and threw "setState() during build" the instant the screen mounted
against a fresh Provider tree (fixed via addPostFrameCallback); and the
body Column had no scrollable ancestor and overflowed even a generously
tall viewport (fixed by wrapping it in a SingleChildScrollView, a real
UX improvement and not just a test workaround).
The three protected EQ test files (servicio_ecualizador_test.dart,
estado_ecualizador_test.dart, servicio_audio_eq_reapply_test.dart) stay
unmodified. Full suite: 730/730 green (2 skipped, unchanged), up from 713.
size:exception - realized 1,410 changed lines (25 files including this
docs update) against the 450-600 forecast: the restructured screen file
alone is 658 lines (a near-total rewrite of a 907-line file, not a
patch), its new test file (first-ever coverage) is 519 lines, and a new
test fake plus a togglePlay() override account for the rest. Not
splittable: the restructure, the tool tray, and the EQ-sheet wiring are
one cohesive change to one screen.
Restyle the Ecualizador settings screen to the new visual language while
keeping the equalizer at 5 bands (spike-resolved, Engram id 2498 - band
count is device-reported via just_audio's AndroidEqualizer, not app-chosen;
the approved mockup's 7 sliders would silently no-op on typical hardware).
- Restyle EcualizadorWidget in place: strip its internal title + preset
chip row (the pushed screen's header now carries the title), add a
habilitado parameter that greys/disables every slider when EQ is off.
Widen PresetsEcualizadorWidget additively (personalizados param) so
custom presets can join the chip row without a second implementation.
- Add servicio_presets_personalizados.dart (new file, own SharedPreferences
key eq_custom_presets_v1) for custom EQ preset persistence - kept out of
servicio_ecualizador.dart, which has an empty-git-diff success criterion
for this change. preset_ecualizador.dart is unchanged: a custom preset is
just a PresetEcualizador with a user-supplied name.
- Extend EstadoEcualizador with presetsPersonalizados,
guardarPresetPersonalizado (validates non-empty name),
eliminarPresetPersonalizado. The load is a new explicit
cargarPresetsPersonalizados(), deliberately NOT folded into
cargarPersistido(): that method is exercised ~30 times by
estado_ecualizador_test.dart (protected, must stay unmodified) via Fakes
only, with no SharedPreferences awareness in that file.
- Build out the Ecualizador screen body: base-vs-per-station explainer
banner, a "Salida activa" row surfaced on the main screen (previously
Advanced-only), an "Emisoras con ajuste propio" drill-down sourced from
the existing presetsPorEmisora map, and a "Guardar como preset" action.
New coverage lives in new files rather than touching the three protected
EQ test files: ecualizador_widget_test.dart (component-level, did not
exist before this commit), servicio_presets_personalizados_test.dart, and
estado_ecualizador_presets_personalizados_test.dart. servicio_ecualizador.dart,
servicio_audio.dart and the three protected EQ test files keep an empty
git diff. Full suite: 713/713 green (2 skipped, unchanged), up from 682.
size:exception - realized 1,954 changed lines (25 files, plus this docs
update) against the 400-550 forecast: lib/ + ARB alone is ~650 lines, near
the top of the forecast band by itself since this WU also had to build out
a screen body WU3a only stubbed; the rest is 4 test files (675 lines) and
11 new ARB keys regenerating 13 lib/l10n/gen files (~546 lines) - the same
pattern every prior work unit in this branch has hit. Not splittable: WU14
reuses this unit's editor component by exact runtime type and cannot begin
until this lands as a whole.
Replace the glass-card container with a full-bleed blurred-art
background, giant heroTime display, 3 fixed snooze tiles (3/5/10 min,
one highlighted), and a full-width stop pill. The status label now
also renders a static "Subiendo volumen" line - no seconds counter -
when the alarm has a configured fade-in, per resolution 4: the
native-to-Flutter progress channel a live counter would need is
deliberately absent from this architecture.
The dismiss guard and force-stop retry banner are untouched: the
banner is byte-identical to its pre-restyle form, only repositioned,
and the guard test's own diff against main stays empty.
size:exception: 530 changed lines (440+/90-) against the 200-300
forecast - lib/ alone is 302 lines, at the edge of the band; the two
touched test files account for the rest. Not split further: this is
one cohesive restyle to the single screen in this branch where an
inconsistent intermediate state is least acceptable.
Replace the native showTimePicker dialog in the alarm editor sheet with
a giant inline HH:MM editor (drag/tap to adjust, wraps at 23:59-00:00).
Weekday circles now render unconditionally (disabled outside weekly
mode) instead of being gated behind an `if`.
The date field, fallback-station picker, and sound dropdown are not
dropped: per resolution 3 they move into a collapsed "Advanced" section
so the mockup's weekday-circles-only layout does not lose capability.
Volume/fade-in sliders get a cosmetic type-scale restyle only.
size:exception: 993 changed lines (891+/102-) against the 500-650
forecast - lib/ production code alone is 429 lines, within band; new
test files and 13 regenerated l10n/gen files account for the rest, the
same pattern every prior work unit in this branch has hit.
Same estimating lesson as every prior WU: the 350-450 forecast
covered the state/model/UI work, not the matching test files, the 14
regenerated lib/l10n/gen/ files, or dart format correcting pre-existing
drift in a touched test file (Dart SDK 3.12.0 skew, not new logic).
Realized 1,822 changed lines; recorded as an accepted size:exception
with a breakdown, not a scope-creep surprise.
Add the Vacaciones manager screen per design ADR-6: an active-range
hero (name, days-remaining countdown, determinate progress bar, and a
per-alarm pause-impact line), a "PROGRAMADOS" upcoming-ranges list, an
"Add range" CTA, and a "Rangos pasados" history section. This is the
real destination WU8's Alarmas-root summary row pushes to, replacing
WU8's own temporary placeholder (_PantallaVacacionesTemporal, now
deleted).
EstadoAlarmas gains 4 pure query methods (rangoVacacionesActivo,
vacacionesProximas, vacacionesPasadas, impactoDeRango) -- read-only
over _alarmas/_vacaciones, no writes, no rescheduling, no native
bridge calls. impactoDeRango mirrors ServicioProgramacionAlarmas's own
pause predicate exactly, so the screen never disagrees with the
scheduler about which alarms are paused. ImpactoVacaciones joins
RangoVacaciones in alarma_musical.dart.
The add-range form (_EditorVacacionesSheet, _PickerButton) moved
verbatim from pantalla_alarmas.dart to its one remaining consumer.
estado_alarmas.dart's scheduling/snooze paths and the ringing screen's
dismiss guard are untouched; both test files pass unmodified.
New ARB keys (en/es only; other 11 locales are WU18's job):
vacationImpact{Paused,Continues}Label, vacationUpcomingSectionTitle,
vacationPastSectionTitle, addVacationRangeCta,
vacationNoActiveRangeHint.
Same estimating lesson as WU3a/WU3b/WU15: the 400-550 forecast covered
only the screen restyle itself, not the matching test-file additions
and the 13 regenerated lib/l10n/gen/ files an ARB touch always drags
in. Realized 1,113 changed lines; recorded as an accepted
size:exception with justification, not a scope-creep surprise.
Restyle the Alarmas root per the functional redesign: alarm cards drop
the always-visible edit/skip/delete button row for a minimal giant
time + station + switch layout. Tap opens the editor, swipe deletes
(with an AlertDialog confirmation), and the hero banner gains an
inline "Saltar" pill for the featured (soonest-firing) alarm's skip
action. No capability from the old button row is lost, only the
trigger location moved; estado_alarmas.dart and its scheduling/
snooze/dismiss-guard tests are untouched.
The vacation inline panel becomes a tappable summary row (range count
+ next-range countdown, computed over the existing estado.vacaciones)
that pushes a Vacaciones manager screen. That destination is a
placeholder for now (_PantallaVacacionesTemporal, holding the old
panel's body verbatim so add/delete-range capability is preserved) --
WU9 replaces it with the real PantallaVacaciones per design ADR-6.
New ARB keys (en/es only; other 11 locales are WU18's job):
alarmHeroSkipAction, alarmDeleteConfirmTitle/Message,
vacationRangesCount, vacationSummary{Active,Upcoming}Countdown.
Extracts ServicioRadio's transport loop (server discovery, host rotation,
bounded retries, User-Agent, timeout, status check, json.decode, sticky-host
bookkeeping) out of `_get` into a new `_getJson(path, params) ->
Future<List<dynamic>>` helper, moved as one block with no logic edits. `_get`
is reimplemented on top, still owning every station-specific concern:
`lastcheckok: '1'`, `Emisora.fromApi` + the empty-uuid/url filter, and the
`_compararCalidad` quality sort. `_getJson` is deliberately sort-agnostic and
filter-agnostic so a non-station endpoint can reuse the resilience behaviour
without inheriting station-only semantics.
Non-negotiable ordering followed per design ADR-4: new
test/servicios/servicio_radio_transporte_test.dart characterises all 8
existing station calls (7 via `_get` plus `registrarClick`, which builds its
own URI) against the UNMODIFIED `_get` first - green by construction -
pinning path, lastcheckok=1, hidebroken=true, a non-empty User-Agent, exact
order/reverse/limit/offset, and the exact returned UUID sequence from a
fixture with deliberately shuffled bitrate/clickcount/votes. That last
assertion is what makes the extraction safe: a sort that silently sank into
transport would pass every other check. Re-running the same file after the
extraction is byte-identical green. test/servicios/servicio_radio_test.dart
is untouched by this work unit - its passing unmodified is itself a signal
that transport wasn't disturbed.
The 6 pre-existing `order: bitrate` occurrences (obtenerPopulares,
buscarPorNombre, buscarPorPais, buscarPorIdioma, buscarPorTag, buscar) are
untouched - a deliberate server-side quality bias deciding which stations
return within `limit`, unrelated to and never to be confused with the
user-facing "Ordenar" control, which stays entirely client-side via the
existing OrdenEmisoras (Engram reference/radio-browser-sort-order).
Behaviour delta, accepted per ADR-4, not a regression: moving
`_servidorActual` bookkeeping into `_getJson` means a successful
`/json/countries` call now warms the sticky host for subsequent station
calls too - one shared warm mirror per instance, desirable, not per-call-type
state.
Adds the Paises browser over the verified `/json/countries` contract (Engram
reference/radio-browser-countries-endpoint): new lib/modelos/pais_radio.dart
(`PaisRadio.fromApi` parses `stationcount` via `int.tryParse` since the API
returns it as a JSON string, not an int - an `as int` cast would throw),
`obtenerPaises()` sends neither `lastcheckok` nor `order` (the screen sorts
client-side by name; the API's raw byte order isn't proper collation for any
locale this app ships), and inherits `hidebroken=true` from the unchanged
`_uri` (desirable here too, since the endpoint's own default is false).
`EstadoBusqueda` gains `paises`/`cargandoPaises`/`cargarPaises()` with an
in-memory cache guard so re-entering the screen never refetches.
New PantallaPaises (lib/pantallas/pantalla_paises.dart): a "Tus idiomas"
shortlist (one representative country per the app's 13 supported locales,
matched against the fetched list - the proposal/spec name this section but
don't specify its derivation) above the full alphabetical list, each entry
showing its parsed station count. Reachable from Buscar's discovery landing
state via a new entry row, added now rather than left dangling per this
file's own forward-reference comment (and the WU15/WU15b lesson: a
fully-tested but unreachable screen is a real defect, not a follow-up).
New ARB keys (en/es only, matching this change's established precedent):
countriesScreenTitle, countriesYourLanguagesTitle, countriesAllTitle,
radioCountriesError.
Tests: 631 -> 649 (2 skipped, unchanged). flutter analyze unchanged at 1
pre-existing info. grep confirms `countrycodes` appears nowhere in lib/.
Restructures PantallaInicio's top of screen: a new _EscucharHero
(square art, live/offline pill, VisualizadorAudio at barras: 30 /
altura: 26 / color: liveGreen, a 5-action transport row - favorite,
EQ toggle, stop, play/pause, sleep - plus a tool-tray entry chip
opening the full player) replaces the old PluriScreenHeader hero, and
a new "Tus emisoras" section (favorites preview, capped, "Ver todas")
follows it. Per design ADR-7, EstadoRadio stays the single source of
truth: the hero is a StatelessWidget with no cached fields, reading
emisoraActual via context.select (uuid-based equality scopes rebuilds
to real station changes) and the fast-changing playback status via
StreamBuilder, the same pattern _Controles/MiniReproductor already
use. The still-present discovery sections (_seccionCercanas onward,
including the old grid) are deliberately left in place - WU6
relocates them to Buscar and deletes them from here; removing them
now would leave that content nowhere until WU6 lands.
MiniReproductor gains a `visible` parameter (default true) and a
measured `static const double altura`. app.dart passes
`visible: indice != RaizPluriWave.escuchar.index`, hiding it visually
only (SizedBox.shrink()) while Escuchar is active, since the hero
already shows the same station - the State stays mounted so its
didChangeDependencies side effect (configurarLocalizaciones, S3-R3)
keeps running regardless of tab. altura was measured empirically
(72.0, via tester.getSize) rather than guessed, backing a new derived
PluriLayout.escucharBottomChromeInset constant now wired into
PantallaInicio's own bottom padding.
"Ver todas" switches roots via EstadoNavegacionRaiz.irA(favoritos),
verified via a NavigatorObserver asserting the push count is
unchanged (switches tabs, does not push).
Fixed a pre-existing test-infrastructure gap while writing the
anti-cache test: no test in this codebase had ever exercised
ServicioAudio.androidAudioSessionIdStream against a bare
FakeServicioAudio (pantalla_reproductor.dart has always read it but
has no test file at all) - the real getter needs registrarHandler()
(main.dart, production only) and threw otherwise. Added an empty
stream override to FakeServicioAudio, matching VisualizadorAudio's
own documented no-native-session fallback.
Tests: 614 -> 618 (2 skipped, unchanged). flutter analyze unchanged
at 1 pre-existing info. git diff empty for visualizador_audio.dart
and estado_radio.dart - this WU touches neither.
Replaces the stacked per-group panel layout with a single
chip-filtered flat list. Chips read "{name} · {count}" (new ARB keys
favoriteGroupsChipLabel/favoritesFilterAllLabel), one per group plus
an "All" chip. Rows drag-reorder via a leading handle
(ReorderableDragStartListener, buildDefaultDragHandles: false) using
the modern onReorderItem callback rather than the now-@Deprecated
onReorder (Flutter 3.44 marks it obsolete).
EstadoRadio additions: listaFavoritosManual (a new memoized getter
returning the stored order untouched by the global ordenListas
setting - listaFavoritos itself always re-sorts by
name/quality on every read, which would silently discard any
drag-to-reorder), reordenarFavorito (thin wrapper over the
already-existing ServicioFavoritos.reordenar, previously unused
outside its own service test), and ordenarFavoritos (applies an
existing OrdenEmisoras criterion via ordenarEmisoras() and persists
the result as the new manual order, so the swap_vert sort action's
result also survives a restart). listaFavoritos itself is untouched,
so Android Auto's tree and the future Escuchar grid (WU5) are
unaffected by Favoritos' own manual order.
Group management: an "Manage lists" action chip pushes the existing
PantallaAjustesGruposFavoritos screen (Settings' own screen, reused
rather than duplicated) - a second entry point to the same screen.
Custom-station CTA: a new dashed-bordered card opens the add-station
form directly; that form was renamed from private _FormularioEmisora
to public FormularioEmisoraPersonalizada in
pantalla_ajustes_emisoras_personalizadas.dart so both screens share
one implementation. New ARB keys: favoriteGroupsManage,
customStationsAddCta.
Tests: pantalla_favoritos_plural_test.dart (the file tasks.md named)
never imported PantallaFavoritos - it only covers stationCount's ARB
plural formatting, unrelated to this screen. Left it untouched and
added test/pantallas/pantalla_favoritos_test.dart instead: 3
state-layer tests for the new EstadoRadio surface plus 6 widget
scenarios (empty-state CTA, chip filter, drag-reorder persistence,
sort action, group management + chip reactivity, custom-station
CTA). 604 -> 614 tests (2 skipped, unchanged). flutter analyze
unchanged at 1 pre-existing info.
Recorded in tasks.md with the test-file correction and the
design decisions this WU had to make on its own (no ADR covers
Favoritos' manual-order persistence).
WU15 shipped PantallaGrabaciones (the recordings library: storage bar,
recording rows, the "..." Rename/Share/Delete menu) fully tested but
reachable from nowhere in the app - a gap flagged in WU15's own
apply-progress notes, not fixed there since it needed a design
decision rather than a guess.
Coordinator ruling applied: the approved mockup's "Ajustes > Grabaciones"
screen depicts the library, not the folder/size settings form. So the
GRABACIONES Y MUSICA group's "Grabaciones" row in pantalla_ajustes.dart
now opens PantallaGrabaciones instead of PantallaAjustesGrabaciones.
The settings form is not dropped - it stays reachable, now from within
the library via a settings icon in its PluriPushScaffold actions,
matching the existing pantalla_ajustes_timer_sueno.dart "Add" action
precedent for a real capability living in the header.
One new ARB key (en/es only, per precedent): recordingsLibrarySettingsTooltip.
Tests: 604 -> 605 (one scenario re-targeted in pantalla_ajustes_test.dart,
one new scenario in pantalla_grabaciones_test.dart, which needed the
same ListTile-ink-assertion suppression helper WU3a/WU3b established
since it now pushes a ListTile-bearing settings screen). flutter
analyze unchanged at 1 pre-existing info.
Recorded in tasks.md as WU15b - not part of the original 18-unit plan,
added here to close the gap WU15 flagged.
Moves the AUDIO group (Ecualizador, Salida de audio, Temporizador de
sueno) and the EMISORAS group (Grupos de favoritos, Emisora preferida,
Emisoras personalizadas, Orden de listas) out of pantalla_ajustes.dart
into 7 new lib/pantallas/ajustes/*.dart screens, each wrapped in
PluriPushScaffold. The root now reaches them through FilaAjuste rows
under two new GrupoAjustes cards (lib/pantallas/ajustes/widgets/
fila_ajuste.dart), per design ADR-3.
Verbatim-move rule applied throughout: only each section's panel header
(icon + title, sometimes a status chip) was removed, since the pushed
screen's own 56px header now carries the title. Two sections whose
header row carried a real action (Temporizador de sueno's "Add",
Grupos de favoritos' "Add list", Emisoras personalizadas' "Add") kept
that action in the body instead of dropping it.
size:exception (move-only diff, pre-recorded at design/tasks time):
34 files, ~4250 changed lines excluding the 13 auto-regenerated l10n
files (~90 more lines there) - higher than the 800-1000 estimate
because that estimate covered the 7 production screens but not the
matching 7 new test files (task 3a.2), one of which relocates ~10
pre-existing device-management test cases verbatim. Business logic is
untouched; app.dart's import of pantalla_ajustes.dart is unchanged.
Correction to tasks.md 3a.1/3a.8: those two lines describe the combined
WU3a+WU3b end state ("4 grouped nav lists", "<400 lines"), matching
design ADR-3's own aggregate blast-radius note - not a WU3a-only claim.
This commit converts only the 2 groups that are WU3a's job; the root
is 788 lines with 5 sections (Grabaciones, Musica local, Idioma,
Backup, Info) still inline, reachable, and unchanged, pending WU3b.
Two new ARB keys (settingsGroupAudioTitle, settingsGroupStationsTitle),
en/es only per the WU1 precedent - all 7 detail-screen titles reuse
existing keys. Discovered and worked around, without touching app
code: Directory.systemTemp hangs real dart:io writes in this sandbox,
and pumpAndSettle() cannot settle while a screen shows an indeterminate
CircularProgressIndicator - both are test-only concerns, documented
inline where hit.
Tests: 560 -> 579 (32 in this commit's scope, net +19 after retiring
13 relocated cases from the old combined pantalla_ajustes_test.dart).
flutter analyze: unchanged at 1 pre-existing info. git diff is empty
for navegacion_auto.dart, servicio_ecualizador.dart and
servicio_audio.dart; pantalla_alarma_sonando_dismiss_guard_test.dart
untouched.
Re-ran test/servicios/navegacion_auto_test.dart (129 tests) unmodified —
all pass. git diff for lib/servicios/navegacion_auto.dart is empty; this
commit makes zero code changes.
Formally closes A1-A5 from the proposal's Android Auto surface review. The
app uses the classic MediaBrowserService model (automotive_app_desc.xml,
no androidx.car.app); Android Auto's system templates render everything,
and the only controllable surface is MediaItem metadata, which already
matches the redesign's intent. A4's waveform visualization is confirmed
non-implementable on this platform — MediaBrowserService cannot render
custom widgets, only system-templated browse/playback UI.
WU1's commit invented an arbitrary value for offlineAccent instead of using
the exact hex the proposal's WU1 scope line already specifies. listSurface
and liveGreen were correctly de-literalised from existing theme.dart
literals; offlineAccent has no prior literal, but its value is still not
this file's discretion — the proposal states #E8879A explicitly.
cambiarActivo persisted BEFORE telling the audio engine, so two quick taps
raced on a SharedPreferences write. When the first write resolved last, the
engine received the FIRST tap's value after the second one: the checkbox read
enabled while the sound stayed flat, and toggling again could invert it the
other way. Reported as the equalizer connecting and disconnecting at random
and the checkbox disagreeing with what is audible.
Reorder to engine first, disk last. The engine call is now issued before any
await, so overlapping taps reach it in tap order and the last tap wins. Each
subsequent step re-checks _activo, so a call that a newer tap superseded
mid-flight neither applies a preset nor persists a value the user has already
changed their mind about. Persisting last also puts what the user HEARS ahead
of what is merely stored.
The regression test drives two opposite taps through a persistence fake whose
FIRST write is the slow one — the exact ordering hazard — and asserts the
engine ends matching the state the UI shows. It fails on the previous
ordering and passes on this one.
An earlier attempt serialized every engine mutation through a shared Future
lane. It fixed this case and deadlocked four widget tests: the lane field
outlived a tester.runAsync block, so a future created in the real async zone
was later chained from the fake-async zone that never advances it. Reverted
in favour of the ordering fix, which needs no cross-zone state.
Only the enable toggle is addressed here. The other reported symptom —
equalization seeming to come and go while playing — is not explained by this
race and is still open; the handler rebuilds the whole AndroidEqualizer on
every player recreation, which is the next place to look.
The car tree carried a whole equalizer feature: an `Ecualizador` root folder
with the six factory presets, a browsable `Personalizado` folder, five band
folders and thirteen gain leaves each, plus the media-id namespaces, routing
predicates, persistence-targeting and children-changed plumbing that fed
them. Equalization is a phone task; the driver should not be tuning 5 bands
from a car screen.
Removed: the `eq_preset:`/`eq_banda:`/`eq_gain:` namespaces and their
predicates and parsers, the `ecualizador`/`eq_custom` folder ids and their
getChildren branches, itemPresetEq/presetsEq/itemEqPersonalizado/bandasEq/
gananciasBandaEq, resolverPresetEq, presetPersonalizadoEfectivo,
dispositivoDestinoEq, debeAplicarPrincipalAhora/debeAplicarSeleccionAhora,
aplicarPresetPorMediaId, aplicarGananciaPorMediaId, and in the handler the
playFromMediaId branches, _presetPersonalizadoAuto, _dispositivoActivoAuto,
_dispositivoDestinoEqAuto and the subscribeToChildren/_hijosSubjects
notification machinery that existed only to refresh band titles after a gain
tap.
Deliberately KEPT: automatic per-device EQ. Reaching the car still applies
that device's preset, because that lives in EstadoEcualizador and the
output-device detection, not in this tree — it works with Android Auto or
without it. Configuring is what moves to the phone; applying stays automatic.
Also kept: the `eq_preset_*_v1` SharedPreferences keys in
ServicioEcualizador, which share a name with the deleted media-id prefix by
coincidence only and hold the phone's own presets.
The root folder set goes from five entries to four (three without local
music); its test now asserts no equalizer folder is offered at all, so a
reintroduction has to be deliberate.
Two of the three Radio Browser hosts this client shipped no longer resolve.
The retry loop rotates de1 -> nl1 -> at1, so once the first attempt failed
for any transient reason the remaining two were guaranteed to fail as well:
the retries meant to add resilience had become a dead end, and a single blip
surfaced as "No connection to the radio API" with a healthy API and a healthy
network. The live mirror list confirms only one server remains:
[{"ip":"91.98.4.78","name":"de1.api.radio-browser.info"},
{"ip":"2a01:4f8:1c1d:699::1","name":"de1.api.radio-browser.info"}]
The API docs say exactly what this code was doing wrong: "Never use a direct
link to a single new server. It is much better to get a list of the servers",
pointing clients at all.api.radio-browser.info to enumerate what exists.
Seed with that round-robin host plus de1, then resolve the real list from
/json/servers once per instance and rotate over that. Discovery shares one
in-flight request across concurrent callers, because the home screen loads
two lists at once through Future.wait, and any failure silently leaves the
seed list in place — it still contains a working host, so a failed discovery
must never be worse than not trying. Explicitly injected servers disable
discovery so callers can still pin a mirror.
Build the User-Agent from the running package too. The API asks clients to
identify themselves, and this header claimed PluriWave/0.1.0 while the app
shipped 1.1.x. A literal cannot stay correct here — CI bumps the version on
every single release — so read it via package_info_plus, already a dependency
used in three other places. If package info is unavailable the product name
goes out alone rather than a made-up version, and resolution never throws: a
header must not be able to fail a request.
Regression from the previous commit. Ranking every AudioDeviceInfo type this
build does not name individually ABOVE the built-in speaker was meant to let
a car stereo on LE Audio or an automotive bus win. It also promoted the
internal sinks a phone exposes permanently: on the Xiaomi test device
AudioManager reports TYPE_FM (14) as an output, so getActiveAudioDevice
picked it over the real speaker with nothing connected at all. Confirmed on
device:
audio_devices.onListen -> {id=other:14:4, type=14, name=2412DPC0AG}
It then reached Dart under an `other:14:4` id whose type is neither the base
speaker nor a known one, slipped past the collision guard and had a preset
row persisted for it -- reinstating the exact symptom this series set out to
kill: a permanent green active-output dot on a device that was not connected.
Replace the deny-by-omission ranking with an explicit allow list of outputs a
user actually connects. The built-in speaker sits below all of them and above
everything else, so any sink that physically exists but is never where media
plays (TYPE_FM, TYPE_BUILTIN_SPEAKER_SAFE, telephony, remote submix) can no
longer be selected. A one-time purge clears the `other:` rows the bad build
persisted; genuine ones re-register on their next connection.
Fix the USB type constant while here: TYPE_USB_HEADSET is 22, not 14, and 14
is TYPE_FM. The Kotlin USB branch hardcoded 14 and the Dart type table
mirrored the same mistake, so the two cancelled out for real USB headsets
while making a phone's own FM sink decode as USB audio. Both now use 22.
Verified with javap against android.jar (android-36) rather than trusting the
comment that introduced the error.
A Bluetooth device only reports its own name through
AudioDeviceInfo.productName while it is enumerated as an active output, i.e.
while it is connected. Paired-but-switched-off devices therefore had no name
to fall back on, and the platform-name cache is in-memory only by design
(bt-device-identity ADR-4), so it self-heals per session ONLY for whatever
happens to be connected. Every other device showed its raw id.
Android already knows those names: BluetoothAdapter.getBondedDevices() lists
every pairing with its name and MAC, connected or not, and nothing in this
app was asking. Read it and seed the platform-name cache from it, keyed
bt_a2dp:<uppercase MAC> to match the ids the audio layer emits.
Seeded BEFORE the active-device query so a live enumeration name, being the
fresher of the two, still wins; a user's custom name outranks both. Re-read
on refrescarDispositivoActual so pairing or renaming a device in system
settings shows up as soon as the list becomes visible.
Reading the bond list is gated by BLUETOOTH_CONNECT from API 31 and by the
legacy BLUETOOTH permission below it, so declare the latter with
maxSdkVersion 30. It is a normal permission: granted at install, no runtime
prompt, no new friction. When the answer is unavailable — permission denied,
no adapter, Bluetooth off — both layers return an empty map rather than
throwing, and the row degrades to the id exactly as before.
Does not help rows persisted under a bt_a2dp:name: placeholder id: those
never had a MAC to match against.
deviceToMap handed the builtin_speaker id to EVERY output type its `when`
did not name. A car stereo on LE Audio (TYPE_BLE_HEADSET) or an automotive
bus (TYPE_BUS) therefore arrived in Dart under the phone speaker's own id,
carrying a type that maps to `desconocido` -- which slipped past the
type-only esBase guard and persisted a device entry keyed builtin_speaker.
From that moment on, every playback through the phone's own speaker matched
that entry, so the green active-output dot stayed pinned to whatever the user
had renamed it to (a car, in the reported case) whether or not anything was
connected. The dot was never wrong; the row was poisoned.
Give unnamed output types their own `other:<type>:<address>` id namespace,
and match esBase by id as well as by type so no future native regression can
re-create the collision. A guarded one-time migration purges what the
collision already persisted from all three device-keyed maps.
Fix the ranking too: builtin_speaker sat inside the priority list as a peer,
so any type absent from that list sorted BELOW the always-present speaker
and could never win. The speaker is now the explicit last resort, externally
connected outputs outrank it, and virtual or call-only sinks (earpiece,
telephony, remote submix, SCO) are ranked below it so they can never be
reported as where music is playing.
Route every AudioDeviceInfo.getAddress read through a version-guarded
helper. It is API 28 with minSdk 24, and two pre-existing unguarded calls in
this same method were latent NoSuchMethodError crashes on Android 7-8.1.
Android lint for :app goes from 8 errors to 6.
Also lets the user manage the list, which is how they recover from a bad
entry without waiting for a release: a remove action clears a device's
preset, name and matrix entries, unnamed rows show their transport and
address tail instead of a raw bt_a2dp:AA:BB:... id, and the green dot
finally carries a tooltip and a semantics label saying what it means.
Device QA pending for wired and USB outputs: no jack or adapter available to
exercise those paths. Their detection is unchanged by this commit.
The recordings live in app-private storage (<data>/app_flutter/grabaciones),
which the Android sandbox forbids any other app from reading, so no
ACTION_VIEW on a file:// or FileProvider URI could ever open it. On top of
that, viewDirectory built an EMPTY candidate list for that path:
directoryDocumentUri returned null (path outside external storage) and
FileProvider.getUriForFile threw because pluriwave_file_paths.xml never
covered app_flutter. The loop never ran, so both entry points -- the radio
recorder and Settings -- always showed "could not open the folder".
Publish the folder as a browsable storage root via
RecordingsDocumentsProvider instead. The files never leave private storage;
the document framework asks us for them one document at a time, and the user
can browse, copy out, rename and delete straight from the file manager. The
root follows a user-configured path and falls back to the default recordings
directory. Its title reuses the already-translated recordingsFolderTitle, so
no new literal is introduced in any of the 13 locales.
Also fixes "open last recording", broken by the same missing FileProvider
root, and replaces Intent.createChooser with a bare startActivity in the
candidate loop: a chooser never throws when nothing can handle the intent, so
the first candidate always "succeeded" and the fallback chain never ran.
Device QA pending -- the provider is driven entirely by the platform's
document framework, so no unit test covers it. Each candidate logs its own
name under file_actions.viewDirectory for logcat triage.
AudioService.init has no internal timeout and an unhandled
onConnectionSuspended case in its MediaBrowser self-bind; under bind
contention with the car's connection it can hang forever, so runApp
never ran (black car screen, white phone UI until process kill).
Race init against an 8s timeout without ever re-calling it: on timeout
run a bootstrap app that waits on the same future, wires the handler
exactly once when it resolves, reports errors via FlutterError, and
swaps to the real app. Auto browse sources now register before the
init await since they take no handler dependency.
Merges its delta requirements into the android-auto-media base spec.
Completes the 3-phase local-music-in-Android-Auto feature. Standing
pre-release gate: on-device/DHU validation of every native surface
built across all 4 phases (SAF picking, metadata extraction, art
cache, queue/shuffle handler wiring) is still outstanding.
Adds "Reproducir carpeta" (sequential) and "Reproducir aleatorio"
(Fisher-Yates over the name-sorted order) as folder-scoped playable
actions, with auto-advance on track completion and skip next/prev.
Isolation from live radio is structural, not disciplinary: the
public playMediaItem always clears the local queue on any call, and
a new private _encolarCambioFuente is the only path that can advance
within it. _cambiarFuente, ControladorReconexion, and the reconnect
error path are untouched -- confirmed by a byte-for-byte empty diff
on all 4 pre-existing radio/reconnect regression suites, independently
re-run before and after (21/21 both times).
Handler wiring itself is static-review-only (PluriWaveAudioHandler
can't be unit-instantiated); the isolation/advance/race-guard
decision logic is extracted into cola_local.dart's pure functions,
which are fully unit-tested.
Merges its delta requirements into the android-auto-media base spec.
Completes Phase 2; only Phase 3 (subfolder scoping, shuffle) remains.
On-device validation of the native metadata path is still an
outstanding pre-release gate across all local-music phases shipped
today.
Local tracks now show embedded title/artist/album art (via native
MediaMetadataRetriever, cached through the existing FileProvider)
instead of the raw filename, falling back gracefully when a file
has no usable tags. Adds two navigable entry points per folder: sort
by audio quality (bitrate, capped at 150 tracks per folder to bound
worst-case latency) and alphabetical name buckets -- the closest
realistic form of "filtering" given Android Auto has no text-search
UI in this integration.
Metadata resolves only for the page actually being browsed (same
slice-cheap-then-map discipline as the paging change), backed by a
flat 256-entry LRU session cache that survives across pages. No new
permission, no new pub dependency, no l10n changes (car-tree labels
stay hardcoded Spanish, matching every existing label in the tree).
Merges its delta requirements into the android-auto-media base spec.
This closes out Phase-1 polish for local music; Phase 2 (metadata,
sort/filter, real art) and Phase 3 (subfolder scoping, shuffle)
remain the only planned future work for this feature.
Folders over the 50-item cap now show a "Mas..." item that reveals
the next page on tap, instead of silently dropping the rest. Paging
slices the cheap raw list before building any MediaItem, so items
beyond the requested page are never resolved (art, title) -- proven
by a call-count test. Also swaps the raw SAF content:// URI shown in
settings for a parsed, human-readable folder name with a localized
fallback across all 13 locales.
servicio_audio.dart is untouched; this stays entirely within the
local-music tree/dispatch layer.
Merges its delta requirements into the android-auto-media base spec.
Phases 2 (metadata/sort/filter/art) and 3 (subfolder scoping/shuffle)
remain planned future work.
Phase 1: pick a device folder via SAF (persisted grant, no new
permission), browse its nested subfolders/tracks as a 5th Android
Auto root folder (hidden until configured), and play tracks through
the existing pipeline (EQ, art rotation, cold-start-safe source).
No metadata/sort/filter/shuffle yet -- filename is the title, generic
rotating art is the placeholder; deferred to a follow-up phase.
Adds a new pluriwave/file_actions native method (listAudioChildren)
and an onActivityResult override in MainActivity for the SAF folder
picker -- both static-review-only, no Android build available here.
Adds an Ecualizador folder listing the 6 fixed presets; selecting one
applies and persists it through the existing headless-safe seam
without touching playback or the now-playing media item.
Favoritos now renders non-empty custom groups as grupo:<id>
sub-folders (hidden when empty) with ungrouped stations left as
direct leaves, reusing the existing hijos() path so the zero-groups
case stays byte-identical to today's flat list.
Android Auto no longer copies the launcher icon as placeholder art; it
rotates through the same 4 on-brand station_art assets the phone UI
already uses, keyed by the same per-station hash for visual parity.
Malformed or unusable favicon URLs (including a Dart Uri quirk where
'http://' reports hasAuthority=true with an empty host) now fail the
validity gate instead of being handed to the OS media browser as-is.
Browsable items also show codec/bitrate as a subtitle when known.
Full time-domain audit (three shipped date bugs prompted it) found one
root cause and two latent travel defects, all now fixed:
Day-stepping used add(Duration(days: 1)), which shifts the absolute
instant by exactly 86400s — documented Dart behavior (sdk#47666), so
crossing a DST transition drifted the wall hour by +-1h permanently
for the rest of the candidate scan (verified: 2026-03-28 07:30
Europe/Madrid + "1 day" = 08:30). The native Calendar engine preserves
wall time, and the single-authority fix made the drifted Dart verdict
win. Candidates now advance by calendar reconstruction (_siguienteDia:
DateTime(y, m, d+1, hora, minuto)), the same wall-clock-preserving
semantics as Calendar.add(DAY_OF_YEAR, 1) plus AOSP DeskClock's
defensive hour/minute re-assertion, keeping both engines in agreement
through any transition.
Instant-valued fields (snoozeHasta/snoozeOrigen/proximaEjecucion/
ultimaEjecucionGestionada/creadaEn/actualizadaEn) serialized as
offset-less local ISO, so re-parsing after a device timezone change
reinterpreted the same wall fields as a different instant. They now
serialize as UTC ("Z"); reads normalize to local, and legacy
offset-less payloads parse identically — no migration. fechaUnica
stays local on purpose: it is a wall-clock date.
One-shot alarms sent fechaUnica's midnight epoch to the native side,
whose boot/travel re-arm derives the calendar day back from it in the
CURRENT zone — a westward shift rolled the date to the previous day.
The channel now anchors the date at local noon, keeping it stable
across real-world zone shifts.
Property tests lock the no-drift guarantee (400 daily / 200 weekday
iterations must all land exactly at hora:minuto — on DST-observing
dev machines this crosses real transitions), plus UTC round-trip,
legacy-payload compatibility, and wall-date preservation tests.
posponerAlarma anchored the snooze to snoozeOrigen ?? proximaEjecucion,
but once the native fire path works, the fire-time sync records the
handled occurrence and recalculation advances proximaEjecucion to the
NEXT day before the user can even tap snooze on the still-ringing
screen. "Posponer 3" therefore armed the snooze a full day out
(captured on-device: snoozeCountdown remaining=1443 minutes). The bug
was invisible before because the broken delivery path never advanced
proximaEjecucion while ringing — each fix unmasked the next.
The anchor is now the newest occurrence that is not meaningfully in
the future (shared 90s imminence window): snoozeOrigen for re-snoozes,
proximaEjecucion on the watchdog path where it is still today's
just-due occurrence, ultimaEjecucionGestionada on the native-fire path
where the sync recorded the ringing occurrence, then now. ServicioAlarmas
exposes ahora() so the anchor uses the same injectable clock as the
rest of the scheduling math. Test fixtures that snoozed half an hour
before the ring — a state the ringing screen can never be in, since it
is posponerAlarma's only production caller — now move the clock to
ring time, preserving their original expectations.
The system ran two independent next-occurrence engines: Dart computes
proximaProgramable (what the UI shows) and sends it as triggerAtMillis,
but the native scheduleAlarm discarded it and recomputed from
hour/minute/weekdays. Two engines over the same data WILL diverge —
observed on-device: Dart said "today 22:48", the native weekday scan
armed next Friday, and the alarm silently never rang at its hour while
snooze (which bypasses recomputation and obeys a timestamp) always
worked. That asymmetry was the user-visible "saving an alarm breaks,
snoozing works" split.
Fresh channel calls now arm exactly the trigger Dart sent whenever it
is in the future or within the shared 90s imminence window; the native
recompute remains as the fallback for stale triggers and for
autonomous re-arms with no fresh Dart data (onAlarmFired's next
occurrence, boot/persisted reschedules). Snooze preservation is
untouched: a live native snooze still short-circuits through the
compute path. Also logs the weekdays/trigger/lastHandled payload on
every schedule call so day-convention divergences are diagnosable
from logcat.
Close the SDD cycle for the ring architecture replacement: verified
with one critical (channel silence by omission) fixed and re-checked
before archive, delta merged into the main native-alarms spec (2
requirements removed, 5 added), artifacts archived byte-for-byte.
Phase 3 on-device QA (9 items) remains the mandatory human gate.
Omitting setSound on a NotificationChannel leaves the platform DEFAULT
notification sound active — omission is not silence. The v3 channel
now calls setSound(null, null) exactly like the pre-notice channel
does, so the native STREAM_ALARM player stays the ring's only audible
source. Caught by verification against design D4 before any build.
PluriWaveAlarmService becomes the sole ring-audio owner for the whole
ring (WU2 of 2, completes the split started at bd7f883): the linear
step ramp (initialVolume/startFadeIn, 250ms steps) is replaced by a
DeskClock-style exponential dB curve (computeFadeVolume: gainDb =
fraction*40-40, curve = 10^(gainDb/20)) on a 50ms Handler loop anchored
at ring start, not audio start, so all three fallback sources (station,
fallback station, bundled WAV) share one clock and a source that joins
mid-fade enters at the elapsed level instead of restarting from
silence. Each source also recomputes and applies the curve immediately
before start() to stay pop-free through prepareAsync's variable
buffering delay.
The STREAM_MUSIC device-volume override/restore is replaced by manual
AUDIOFOCUS_GAIN_TRANSIENT request/abandon on STREAM_ALARM (no-op focus
listener, requested once per ring in startAudio, abandoned
unconditionally in stopAlarm's full-teardown branch): the service never
calls setStreamVolume on any stream. The fire notification channel
migrates pluriwave_alarm_fire_v2 -> pluriwave_alarm_fire_v3, now silent
(no setSound; native MediaPlayer is the only audible source) while
keeping vibration and IMPORTANCE_HIGH for the full-screen intent; the
migration guard folds in a third delete for the v2 id alongside the two
pre-existing legacy ids, guarded by a renamed channels_migrated_v3 flag
so it still runs exactly once.
Since the Dart ringing screen (WU1) no longer calls confirmFlutterAudio,
overrideMediaVolumeForRing or restoreMediaVolume, their native surface
is now dead: deletes the flutterOwnsRing handoff flag and both its
backstop call sites in PluriWaveAlarmService, and the three
MethodChannel handlers plus their backing methods and companion state
in MainActivity.
PantallaAlarmaSonando no longer owns any audio orchestration (fallback
player, dB ramp, native handoff confirm, media-volume override/restore):
it only calls EstadoAlarmas.finalizarEjecucion/posponerAlarma from
Stop/Snooze/back, keeping the single-exit guard, PopScope back=Stop and
dismiss semantics intact. The status line now reads directly from the
alarm's static config (station name or a neutral label) instead of a
live playback/handoff state.
PuertoAlarmasAndroid drops confirmarAudioFlutter,
forzarVolumenMediaParaAlarma and restaurarVolumenMedia, and app.dart no
longer pre-starts a station before pushing the ring screen. This is the
Dart half of moving to a single native ring-audio owner (WU1 of 2); the
Kotlin service rebuild lands next and keeps this intermediate state
shippable with no double audio.
On-device logcat from the latest test showed two defects the previous
design created. The fade-in was gated on the station reaching
`reproduciendo`, and the stream took 18.7 seconds to buffer: the ring
sat frozen at 5% the whole time and the configured fade seconds only
started counting afterwards. And the stream override was raised during
pre-start, so the ExoPlayer AudioTrack spin-up — which runs at gain 1.0
for an instant before the player gain lands — blasted at the configured
ring level, heard as "starts directly at the alarm volume".
The ramp is now anchored at alarm time: it starts when the screen
starts, buffering just joins it at the elapsed level, and the fade
duration means seconds-from-alarm. _iniciarFadeIn is single-start so
the handoff confirmation and fallback paths can no longer restart an
in-progress ramp from 5%. The stream override moved from the app-side
pre-start into the screen and is raised only when audio is actually
about to flow (first `reproduciendo`, the already-playing branch, or
right before the fallback WAV plays), so track spin-up happens under
the user's original low volume and the blast is physically impossible.
Exit teardown restores the device stream before resetting the player
gain, removing the brief exit blip seen in the capture.
Two exit-path holes found by adversarial review before the next build:
PluriWaveAlarmService.stopAlarm() never compared the requested id to
activeAlarmId, so any stop request for a DIFFERENT alarm tore down
whichever ring was active: with two alarms firing close together, the
second one's routine hide-notification call (via dismissAlarmNotification
-> ACTION_STOP) killed the first alarm mid-ring and prematurely restored
the device volume override. A mismatched id now only cancels that id's
notification and returns; null keeps full-teardown semantics for
internal/onDestroy callers.
The ringing screen never intercepted the system back gesture: a plain
route pop ran only dispose(), leaving the shared radio player ringing
with no alarm UI left anywhere to stop it. Back now routes through
PopScope into the same _detener() flow as the Stop button, guarded by a
single-exit flag so a back-press racing a button tap cannot run the
teardown twice and pop the route underneath.
Also resets the shared handler gain to 1.0 on ring exit: the fade-in
mutates the radio player's persistent volume, and exiting mid-ramp used
to leave every later radio play at the partial ramp level.
overrideMediaVolumeForRing parsed and logged the fraction argument but
called the no-arg override, which always forced STREAM_MUSIC to the
device maximum. On-device logcat confirmed it: the Dart side sent
fraction=0.5 yet the stream was set to index=30 of 30. Combined with
the player now ramping to its full range, the ring peaked at 100% of
the device maximum instead of the configured 50%.
The override now sets the stream to round(max * fraction), clamped to
at least 1 so rounding can never mute the ring. With fraction=0.5 the
stream caps at half the device maximum and the player ramps up to that
cap, so the ring peaks at the configured level and the opening buffer
click drops to the configured fraction rather than full scale.
The native next-occurrence recompute required the trigger to be
strictly in the future, while the Dart side keeps an occurrence whose
trigger passed within a 90s tolerance. When the periodic resync
re-armed an alarm microseconds after its trigger (app foregrounded,
the 60s tick straddling the trigger instant), computeNextTriggerMillis
recomputed the next weekday/daily occurrence as tomorrow and, through
the shared FLAG_UPDATE_CURRENT fire PendingIntent, replaced the
in-flight fire before AlarmManager delivered it. The alarm never rang
until the screen was turned on and the Dart watchdog caught it late.
computeNextTriggerMillis now mirrors Dart's toleranciaDisparoInminente:
base is lowered by a 90s grace window so a just-passed occurrence is
armed (and delivered ~immediately) rather than pushed to the next day.
The handledFloor (lastHandledAtMillis + 60s) stays a hard lower bound,
so an already-fired occurrence can never be re-selected — no
double-fire. Dart contract tests lock the boundary the native constant
must track. Native verification is on-device (no JVM test harness).
FreeTLab