A same-route remount that cannot edit yet - hass arrives after the
element is inserted (the demo's own order), or a non-admin waits for the
server's can_write - keeps the editor in _pendingNavMode and enters it
later through _resumePendingNavMode -> _setMode. The draft revival was
wired only into the immediate warm adoption (_requestMode(..., adopt)):
on the pending path the draft was lost, _warmRevivePending stayed up for
the instance's life, _warmSnapshot stopped writing dlg, and the next
remount brought back the predecessor's stale draft.
The adoption tail (draft revival once under a held refit, then the
settled stage as the refit baseline) is shared by both paths: the
sequencing lives in src/warm-mode-adoption.ts, the refit bookkeeping in
the card's _holdWarmRefit/_releaseWarmRefit. The pending mode still enters through _setMode, the transition
authority smoke_nav_persist holds it to; resumeWarmMode then settles the
revival. _setMode ends the passive boot grace and refits the camera to
a header measured before the editor chrome rendered, so a camera the
pending window left untouched is put back and held exactly as an
immediate adoption holds it; a camera that has already moved on (View
refit, the user's pan, another space) is left to the ordinary refit. A
mode that did not commit, an explicit mode command in the pending window
and a route departure settle the revival too: no outcome leaves
_warmRevivePending up. The core file gives back 4 lines.
smoke_warm_dialogs gains section H through the UI: the three late-write
orders keep mode, draft, dirty baseline and a frame-by-frame identical
viewport; the chain carries this instance's draft, not the predecessor's;
a space switch in the pending window eats the draft. 14 checks are red
on dev. The mutant warm-pending-mode-leaves-revive-waiting is guarded by
the smoke; docs/WARM-REMOUNT.md §2 describes the pending editor.
Issue: #756
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The space card drew its rooms with a bare map(), so Lit reused room nodes by
position, and `.room { transition: 0.12s }` (planStyles is part of this card's
styles too) drew a node's fill and stroke in from whichever room held it
before. Two paths change the room set in the same DOM: a new `space` in
setConfig of the same element (the card editor's preview), and a config event
from any device that inserts, removes, reorders or re-zones a room of the
shown space. Filled rooms faded out and back in for ~0.12 s, unfilled ones
briefly darkened in a filled room's place.
The list is now keyed(space.id, repeat(rooms, (r, i) => r.id || i, ...)),
the same shape as the full card after #742: the outer key handles the space
change, the inner one keeps a node bound to its room inside a space. An
id-less room keys by its numeric index, which never equals a string id. The
transition itself stays: it smooths a real fill change on the same room. The
#742 note in plan.styles.ts now names the space card as well.
Witness: a new section of smoke_space_card. The config is delivered by a
server push (__hpTest.setServerConfig), the event the card subscribes to.
Red on dev: node r1 reused for g1 with fill/fill-opacity transitions and a
first-frame fill of rgba(0, 0, 0, 0) / 0; a room inserted first shifts all
four nodes and replays fill transitions. A real custom_fill change still runs
a fill transition on the same node (catches `transition: none`). One mutant:
inner key replaced by map(), guarded by AC2 (checked by hand: red).
Issue: #745
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The lazy-runtime contract (#353) says a non-terminal failure waits for
the next explicit intent and that there are no background retries. But
_renderBody calls ensure() on every repaint while a surface waits for
the editor or onboarding runtime, and to the loader that call was
indistinguishable from an intent. Surfaces the core opens without the
runtime - the kiosk size dialog after a 3 s hold, the floor import
wizard on an empty plan, a dialog a warm remount revives - therefore
turned one failure into a loop: the loader's own state change, the
toast and its expiry, every hass tick repainted, started a new cycle
and showed a new toast every ~3.5 s. A wall tablet whose old hashed
chunks answer 404 after an integration update sat in that loop forever.
EditorRuntimeLoader.ensure takes an intent: the render calls it as
'reconcile'. A reconcile starts the first cycle a surface needs, but
after a non-terminal failure it returns false without loading until an
explicit ensure() - a tab, an opener, _requestMode, "Add space" - has
started a new cycle. Explicit calls, the terminal fingerprint failure,
ready and an in-flight cycle behave as before, for every loader
instance. The card's render lines stay line-neutral.
smoke_lazy_editor_chunk gains the three surfaces offline through their
real paths (a 3 s touch hold on a kiosk card, an empty plan pushed by
the server, General settings revived by a remount): one cycle, one
notice and an idle loader over 8 s, then the Plan tab and "Add space"
heal. On dev: 4 requests / 3 notices, 6 / 2 and 3 / 2. The loader unit
test pins reconcile versus intent; the mutant
render-reconcile-restarts-editor-runtime-cycle is guarded by the smoke.
Issue: #757
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The flat room list was a bare map(), so Lit reused room nodes by position.
On a floor switch the previous floor's room node became the new floor's room
and `.room { transition: 0.12s }` drew its fill and stroke in from the old
computed values: one paper-white frame, then two or three frames darker than
the final fill (alpha rises while fill-opacity falls), then the fill. Between
two filled floors the fill bled in from the other floor's colour.
The list is now keyed(space.id, repeat(rooms, (r, i) => r.id || i, ...)), the
shape #534 settled on for openings and markers. The outer key handles the
floor switch, including room ids repeated on two floors (ids are unique only
within a space); the inner key keeps a node bound to its room inside a space,
where inserts, merges, splits and the editor filter shift positions. An
id-less room keys by its numeric index, which never equals a string id. The
transition itself stays: it smooths hover and a real fill change on the same
floor.
Witness: a new section of smoke_space_switch_transitions, before physicalize
(after it the first room changes template branch and the node is recreated
anyway). Red on dev: five transitions on g1, node r1 reused for g1, computed
fill rgba(0, 0, 0, 0) / 1; room nodes swapped by an insert; same-id case
reuses r1. A real custom_fill change still runs a fill transition (catches
`transition: none`). Two mutants: inner key removed, outer key removed.
Issue: #742
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
#718 K7 takes the moon status once per opening of General settings,
outside the draft. A warm remount revives the open dialog on a new card
instance, but `_warmReviveDialog` restored only the draft: the new
instance had no opening of its own, so the "Now: ..." line never came
back.
A revive is an opening too. The `settings` branch now asks for the
status the way `_openSettingsDialog` does - through the lazy editor
runtime (`_openMoonStatus` -> `openMoonStatus`): at once when the
runtime is there (an editor revives after `_requestMode(..., adopt)`
has installed it), after it loads in View; once per revive and only
while that revived dialog is still open. The snapshot of now,
`hass.config` and `sun.sun` is the revive's own, nothing of the dead
instance's opening is carried over, the draft key and the dirty flag do
not change. The View graph gets no static moon-status import; other
dialog kinds never ask for the moon chunk.
demo/smoke_moon_status.mjs gains the revive scenarios - View, the plan
editor, a revive while the chunk is still loading, a space-dialog
revive that must not load the chunk; the first three are red on dev.
test/moon-settings.test.mjs executes the revive as a new opening; the
wiring itself is proven by the smoke, not by reading the monolith as
text (#624). docs/SUN.md and docs/WARM-REMOUNT.md say a revive is an
opening; scripts/smoke-links.mjs links the two new symbols to the smoke.
Issue: #731
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The owner decided on 30.09 that the moon is not part of the "Follow the Sun"
environment but a switch of its own: with a static background (global or a
space's own) the card showed no moon even with the switch on, and the switch
said nothing about why the moon was missing right now.
With a static background there is no environment, so the moon stands in its
own layer, `.hp-moon-sky`: the first child of `.stage` / `.hp-static-stage`,
the whole scene, no z-index, filter or will-change, under the plan by DOM
order, fading with the #101 View weight. Inside is the very #661 element, so
place, size, art and fades are unchanged, and a background switch moves it to
its new parent in the same render without a flicker. The phase comes from the
same `resolveDayCycle`, computed only while the moon is on and on View; without
`sun.sun` both cards keep their 30 s clock ticker and re-render only when the
phase changes (the environment is still compared by its whole fingerprint).
General settings get a second caption line under the moon switch
(`data-moon-status`): one snapshot per opening, judged by the lazy chunk as if
the switch were on, first reason wins (no home, day, below 3°, under 3 %),
numbers rounded and clamped below the threshold they missed. `moonStatus`
decides "shown" with the same `moonShownAt` as the element. It lives in a
WeakMap beside the draft, so it never makes the dialog dirty; a closed
opening's result is dropped. The dialog loads the chunk through the gate's
loader (`withMoon`), now shared by every caller while a load is in flight, so
there is still one fingerprint check and one retry token.
Bundle (same build, against origin/dev): initial View 300 072 -> 300 248 B gzip
(+176 B, under the 500 B of the spec; budget and ceiling not raised); lazy
editor 238 558 -> 238 991 B (+433 B, the line and English strings); lazy moon
11 385 -> 11 712 B (+327 B, layer CSS and status). `src/moon.ts` stays out of
the initial and the editor graph; bundle-budget now refuses an editor/moon
overlap. Monolith metrics: hostRefs 4 885 -> 4 888 — the three `host.` reads of
`src/editors/moon-status.ts` (hass, `_settingsDialog`, requestUpdate) through
its own three-member interface, not the editor port; the other five metrics
are unchanged. houseplan-editor-runtime.ts grows by two lines (import, call).
Tests: AC9/AC10/AC15 and the sky layer in test/moon.test.mjs (the #661
"static -> nothing" check inverted), AC14 and the opening lifecycle in
test/moon-settings.test.mjs, smokes demo/smoke_moon_static.mjs (AC1-AC6; AC1
and AC3 were red on dev) and demo/smoke_moon_status.mjs (AC11/AC12), AC7 in
smoke_daycycle_layer_budget. Golden: two new scenes
(static-bg-moon-gibbous-white-light, static-bg-moon-crescent-south-dark,
matrix v70), the harness checks the moon's parent by background and waits for
the status line in the General settings frames. Four new mutants; the clock
ticker one is a browser guard (201 at the guideline of 200).
Issue: #718
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
- Vertical oblique projection: the floor matrix is the identity and a height
rises straight up by z·sin 20°, so the on-screen wall height is unchanged
and the cos 20° foreshortening of the plan, decor and anchors is gone.
- Device tiles and lock badges stand on the wall-top plane with one common
shift; the #651 placement search no longer runs in the live scene (its
removal is #714). Room names keep their Flat floor point.
- The 2.5D fit no longer reserves the 48 CSS px nudge budget.
- Switching projection keeps the camera when the previous projection was on
screen: saving the setting, entering an editor from 2.5D and adopting a warm
memo from the other projection re-read only the scalar zoom. A cold 2.5D
start still opens the 2.5D home.
- Opening faces are ordered along the oblique projector (s·y + z).
Witness: demo/smoke_iso_flat_parity.mjs (AC2–AC5, AC11) is red on the old code.
Issue: #713
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
At dawn, dusk and night the environment shows the moon in the top-left
corner of the scene, behind the plan: computed in the card from the home
coordinates and the browser clock (short Meeus series + topocentric
parallax, within 1.4° / 1.8 pp of JPL Horizons), one designer image under
a continuous phase mask with the lit side always on the left (owner
2026-09-29), a feathered terminator, 3°/3 % thresholds and the 2 s fade of
the window rays. Everything but a small gate lives in the lazy
moon-runtime chunk, with its own 30 s ticker. General settings: "Sun"
becomes "Sun and Moon" with one switch, on for new installations
(DEFAULT_CONFIG), off for existing ones.
The initial View graph sat 728 B under its budget: the gate is paid for by
moving fifteen dialog-only strings of General settings into the lazy
settings dictionary (#459) and by one build fingerprint literal instead of
three, so the graph ends 4 B above dev. Golden: two new moon scenes, and
the two General settings help frames show «Sun and Moon»; the WSL artifact
test fixture now models scenes whose first capture awaits acceptance.
Issue: #661
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
Owner's decision in #694: icons must not change position with state.
Since #651 the rigid overlay layout sized a device by its value text and
badges, which change with HA state: a light toggling on changed its width
from 37.2 to 26.7 CSS px and re-laid out all 62 overlays of the dense
scene (~385 of 495 ms of stateUpdate; v1.77.0 had 208 ms).
- iso-scene-render.ts: the layout sees the state-free tile (icon at its
configured size, no value text, badge or supplemental metrics). An
HA-only change keeps the layout and refreshes only the visual extent
that scene bounds read, without a collision search.
- iso-overlays.ts: the rigid-group search skips candidates that already
lose to the fallback on room, then wall violations (lexicographic
bound). The result is unchanged — identical placement hash on the dense
scene — at about 35 % less work.
- docs/ISOMETRIC.md, changelogs; test #711; mutant
iso-device-layout-follows-state-again (written, not run — #709).
Local isometric-stage3-dense-v1, 3 samples: stateUpdate 522 → 89 ms
(v1.77.0: 208), modelReady 3665 → 3129, switchCycle 5544 → 4700.
Issue: #711
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The Plan editor rule `.hp-stair.input-enabled .hp-stair-hit { cursor: move }`
also matched the View layer, which sets input-enabled only to receive
clicks. The hit area sits over the outline, so the link's pointer on the
group was never visible and every stair in View showed a drag it cannot do.
- src/stairs-view.ts: View stairs carry `hp-stair-view`.
- plan.styles.ts: `move` applies only without it; in View the hit area
keeps the group's cursor — pointer on a link, the stage's otherwise.
- demo/smoke_stairs.mjs: computed cursors in View (link, no target) and
in the Plan editor; the two View checks are red on the old code.
- test/stairs.test.mjs: the cascade without Chromium; mutant
view-stair-cursor-move-again.
- docs/STAIRS.md, changelogs.
Issue: #693
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
With the day/night background the plan went blurry after zooming from 100 %
(sharp when the page was opened at 800 %) and navigating a strongly zoomed
plan flashed the page white. Both came from #582's composition, not from the
wall hatch that #685 replaced:
- `.stage.daycycle.hp-safe-daycycle-outline .plan-svg` promoted the scene
with `will-change: transform`; Chromium freezes the raster scale of such a
layer, so the 100 % raster was shown stretched. The explicit layer #582
needs is now `will-change: opacity` (re-rasters at the current scale).
- The filtered outline had `overflow: visible` and a gesture exposed every
scene (#544) without bound, so the promoted layers grew with zoom squared
(CDP LayerTree, ~460 %: plan-svg 15.9x, outline 13.2x the stage; 39x after
navigating at 800 %). The full card clips its outline to its box and marks
it data-hp-live-overflow="clip" (never exposed); the live viewport bounds
every other exposure with an inline clip-path: inset(-25%) that leaves
with it, so idle DOM stays byte-identical (#531).
Owner-verified in Chrome 152 (built-in browser, DPR 2): sharp after 100 ->
800 %, no white flashes after reloading at 800 %.
Owner decision: #685's analytic gradient is reverted (13af1d5e), the single
<pattern> is back at every zoom; its close-up golden scenes stay and check
the pattern, its terminal-frame smoke checks the pattern.
Witnesses: demo/smoke_daycycle_zoom_layers.mjs (800 % x DPR 2: layers vs
stage, the hint, reload path, button/wheel/pinch); #582/#532 smokes now pin
the opacity hint; test/live-viewport.test.mjs (bounded exposure, clipped
scene); test/daycycle-layers.test.mjs (cascade). Four Node-guarded mutants.
Issue: #689
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
The Plan editor hid every device marker (display: none), so walls,
openings and stairs were placed against devices from memory. It now
shows them exactly as the Background editor does (#362): the same
markers as View, at the Background editor's effective opacity, fully
pointer-inert. The fade is filter: opacity(0.35) on each marker, not
on .devlayer, because room labels and room settings buttons share that
layer and stay opaque and interactive; filter multiplies the marker's
own opacity (.unavail -> 0.35 x 0.35), as the Background layer does.
Marker handlers already fail closed outside View/Devices.
demo/smoke_plan_device_landmarks.mjs proves AC1/AC2: same markers as
View, per-marker alpha equal to Background, core and capsule fall
through, a real click on a marker reaches the Walls tool and a room
settings button below, cursor and target unchanged, handlers inert, no
tabindex. Three mutants guard hiding, opacity override and the pointer
boundary. UX-MODES, DECOR-EDITOR, USER-GUIDE.ru mode table, changelogs.
Issue: #687
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
Решение владельца 2026-09-28: «убрать рамку фокуса полностью». Навигационная
лестница View — фокусируемый `<g role="link" tabindex="0">` (#676), а своих
стилей фокуса у `.hp-stair` не было, поэтому браузер рисовал чёрно-белое
кольцо по прямоугольнику символа. Теперь `.hp-stair:focus` и
`:focus-visible` — `outline: none`: ни кольца браузера, ни замены. Лестница
остаётся в порядке Tab, Enter/Пробел по-прежнему переходят; hover и
выделение в редакторе не тронуты.
Свидетель — `demo/smoke_stairs.mjs`: снимок области лестницы с запасом на
кольцо без фокуса и с клавиатурным фокусом (`:focus-visible`) обязан совпасть
байт в байт; Enter на сфокусированной лестнице переводит на целевой этаж.
Без правила смок красный (`focusedStairPaintsNoFrame`). STAIRS.md и оба
changelog.
Issue: #686
User-Visible: yes
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
Лестница рисуется протяжкой, как прямоугольник или эллипс декора: длинная
ось протяжки — ось подъёма, стрелка смотрит от нажатия к отпусканию, клик
ставит размер по умолчанию. Выделенная лестница получает рамку декора в
верхнем оверлее карточки — выше тел стен, с узлами постоянного экранного
размера и курсорами по мировому направлению узла; винтовая — четыре узла на
касательных. Ресайз идёт от противоположной стороны, не зеркалит и не
поворачивает; магнит работает по сторонам: в пределах 6 клеток и 5° сторона
ложится на видимую грань стены, при перемещении лестница доворачивается не
больше чем на 5°, при ресайзе и черновике — никогда. Грани перегородок и
колонн получают внешнюю сторону по обходу многоугольника.
Клик, синтезированный после жеста, глушится: под «Лестницей» больше не
появляется копия, под «Выбором» не пропадает выделение. Диалог свойств
считает в сантиметрах лестницы (30…10000), а не в конвертере толщины стены;
нетронутое поле хранит число бит в бит, «Сохранить» без изменений ничего не
пишет. В Просмотре наведение на лестницу с корректной целью показывает
«Переход на этаж …» — ровно при том же условии, что делает её ссылкой.
Чистые функции слоя — в stairs-editor-model.ts с юнитами; смок лестниц
покрывает протяжку, курсоры, рамку над стенами, клик после жеста, диалог и
пять состояний подсказки; реестр мутантов получил шесть свидетелей.
Issue: #676
User-Visible: yes
Co-Authored-By: Claude Fable 5.1 <noreply@anthropic.com>
Claude-Session: https://claude.ai/code/session_018qZfe7YS4rqEMKoVeS3GKd
Прямые и винтовые лестницы получили отдельную модель, инструменты редактора, безопасную межэтажную навигацию, вычитание из чистой площади и плоское отображение в 2.5D.
Issue: #663
User-Visible: yes
The raised room label in 2.5D View sized its own box to the 44 px touch
minimum (min-height + justify-content: center). The name therefore sat
centred in 44 px while the metrics row, placed absolutely from the label
bottom, hung (44 - name height) / 2 below it: 18.3 px at a 9 px name and
6.4 px at a 39 px one on the demo stand, against a constant 0.1 name
height on the flat plan.
The 44 x 44 px floor moves into an invisible ::before (the door-lock
pattern), z-index -1 inside the label stacking context; the label box is
sized by its text again.
demo/smoke_iso_room_label_metrics.mjs measures gap / name height at two
cameras (font x4) in Flat and 2.5D, and the 44 px floor and area-link hit
targets on labels no raised device covers. Mutant
iso-room-label-44-box-centres-name.
Screenshots: 11 frames pixel-identical, fingerprint only.
Issue: #665
User-Visible: yes
The accent fill of the active Plan/Devices/Decor tab is now a pseudo-element
stretched over the 2 px group gap and the fixed 24 px X slot of #660, so
the X sits inside the highlighted zone without any box moving: the tab,
the slot, the X and the header keep their geometry. The X takes the tab
text colour, and the keyboard focus ring moves to the same pseudo-element
so it outlines the whole zone instead of cutting it at the tab edge.
.modes becomes its own stacking context so the z-index:-1 fill paints above
the group background. On a phone the mode tabs, and the fill, stay hidden.
smoke_toolbar_stable_width: painted-pixel probes for the gap, the slot edge
and top, and the first pixel after the slot, in every editor at every width
above 480 px; the X colour equals the tab colour. The existing geometry
and hit checks are unchanged and green. Mutant
toolbar-active-highlight-stops-at-tab.
Screenshots: 11 frames pixel-identical, fingerprint only.
Issue: #666
User-Visible: yes
The stage is pointer-events:none, but that is not an inherited ban: the
shared marker styles opt the 44 px floor (.dev::before) and the painted
capsule (.device-shell-frame) back in for the interactive plan (#564).
The static card imports them whole, so since v1.76.0-beta.2 it showed a
hand cursor over every marker and swallowed clicks that do nothing.
The card now closes every descendant opt-in (.hp-static-stage *, ::before,
::after). smoke_space_card asserts the browser hit test itself
(elementFromPoint over the marker and a 9x9 grid across the stage) and
the cursor, not the stage computed style that stayed green through the
regression. Mutant static-card-descendants-hit-testable.
Screenshots: 11 frames pixel-identical, fingerprint only
(npm run docs:accept -- --identical).
Issue: #664
User-Visible: yes
Переход снова интерполирует общий бюджет сцены и панели инструментов вместо измерения скрытого исходного кадра. Документационные кадры попиксельно не изменились.
Issue: #660
User-Visible: yes
- settings.volumetric_view (General settings › Display, third switch) replaces
the alpha entry, the header projection toggle and the phone-menu item; one
rule for card, sidebar page and kiosk; editors stay Flat; backend accepts
only a boolean, support package copies it.
- Raised tiles in 2.5D View: no ring, rounded body min(0.275 D, 0.3 h), edge
0.1 D with the lab filters, ×1.12 size also in layout/collision, lift
0.075 D, one floor-shadow layer under all markers (theme × floor table),
frames hugging tile and edge (Alert > Focus > Selected > Hover), forced
colours without edge and shadow (src/iso-tiles.ts, styles/iso-tiles).
- Soft sun wash instead of projected Flat wedges (src/iso-sun.ts): same gates
and windows as sun_rays, length from elevation, parallelogram along the sun,
tone and streaks by floor lightness, sill line.
- Walls take the user's wall colour (top opaque, sides × .77/.68/.60); theme
rules for walls/openings/labels removed; furniture uses the Flat stroke rule.
- Smokes smoke_iso_tiles/iso_sun/iso_theme_walls/volumetric_setting, 16 new
mutants, acceptance scenes STAGE6_ACCEPTANCE_SCENARIOS (golden with their
Linux CI baselines), ACCEPTANCE.md side by side.
Issue: #649
User-Visible: yes
На окне не шире 480 px шапка карточки и страницы бокового меню занимает одну
строку ≤ 56 px: вкладки пространств (без шестерёнок и «+», прокрутка вбок,
текущая вкладка докручивается в видимую область), масштаб, одна шестерёнка
«Действия и настройки» и у админа слот крестика #647. Заголовок скрыт.
Меню шестерёнки (src/header-menu.ts): редакторы, настройки и добавление
пространства, общие настройки, PDF, помощь, сводная панель и объёмный вид
(alpha) — в фиксированном порядке и с теми же условиями и действиями, что у
прежних кнопок. Пункт закрывает меню; Escape возвращает фокус на шестерёнку;
касание вне меню поглощается прозрачной подложкой. На > 480 px шапка прежняя,
в киоске меню нет.
- три кнопки шапки вынесены в renderHeaderActions — шапка карточки не растёт;
- сводная панель отдаёт свои пункты методом menuItems();
- i18n: title.header_menu (en/ru/de/fr);
- потолок initial View 291 000 → 292 600 (замер 291 635), база bundleBytes;
- смок smoke_mobile_view_header, юнит header-menu, 10 мутантов; смоки
support_feedback, gear_tabs, modes, toolbar_stable_width переведены на
меню на ≤ 480 px; подпись вкладки в span.tabtitle без pointer-events;
- USER-GUIDE.ru, UX-MODES, TOUCH-SUPPORT, STYLING-HOOKS, data-hp-contract,
оба changelog.
Issue: #616
User-Visible: yes
Счётчик «N устр.» и ключ count.devices удалены вместе с правилом
`.head .count` (скрывало счётчик на ≤1100 px). Крестик закрытия редактора
вынесен из активной кнопки режима в отдельный слот `.editor-close-slot`
сразу после группы кнопок режимов: 24 × 24 px (--hp-editor-close-size),
одинаковый в редакторе и вне его; вне редактора слот пуст, aria-hidden,
без фокуса и pointer-events. Кликабельная зона × — весь слот (≥24×24, #195),
глиф 13 px; × виден на любой ширине, где видны кнопки режимов.
Смок smoke_toolbar_stable_width (1400/1000/768/390 px, en/ru): ширина
шапки и отступы кнопок стабильны во всех переходах, × виден, попадание
в край зоны закрывает редактор, нет нового overflow. smoke_editor_tabs,
smoke_edge_cases, smoke_hidden_flag переведены на слот/список устройств.
6 мутантов. docs: UX-MODES.md, STYLING-HOOKS.md, data-hp-contract.json.
Issue: #647
User-Visible: yes
Транзитная геометрия Resize теперь только вычисляет центр для текущего кадра и не удаляет сессионную позицию. Окончательная геометрия по-прежнему очищает точку, если она действительно вышла за комнату.
Issue: #645
User-Visible: yes
Кнопка сохраняет положение в пределах комнаты только на время сессии редактора,
корректно отменяется при pinch/cancel и учитывается подписями Resize.
Issue: #645
User-Visible: yes
General settings colour tiles (room fills and glow) paint their colour at
its opacity over the checkerboard again, so "No light sources" at 0 % no
longer looks like solid grey. The label ink is chosen from the visible
colour (hex mixed with the checkerboard by alpha). Colour plates
(flat-swatch without cover-swatch) stay solid — now witnessed by the
- hp-color-opacity: one `_solidSwatch` condition drives background and
opacity; cover-swatch keeps the checkerboard
- editors/color-tile-ink.ts: pure isLightHex/colorTileInk, unit-tested
- smoke_dialog_polish_605: tileAlpha, defaultsDiffer, noneInkDark,
liveAlpha, platesFlat, plateSolid; oneSurface drops the opaque check
- mutants M-615-tile, M-615-plate, M-615-label
- USER-GUIDE (ru/en): "Colour tile" row; both changelogs
Issue: #615
User-Visible: yes
На вкладках «На плане» и «Скрытые» — флажки строк, «Выбрать все (N)» по
всему отфильтрованному набору (включая строки за «Показать ещё»), панель
«Выбрано: K · Скрыть/Показать выбранные (K) · Снять выбор» в отдельном
контейнере .device-inbox-batch вне .device-inbox-filters.
- src/device-inbox.ts: чистые inboxVisibilityAllowed (один источник для
canHide/canShow и выбора), selectableInboxRows, effectiveInboxSelection,
applyInboxVisibility — свёртка одиночных действий; одиночное = пакет из
одной строки.
- src/device-inbox-batch.ts (новый, lazy editor): выбор, панель, строка-
флажок и writeInboxVisibility — одна запись houseplan/config/set с
expected_rev на пакет, busy='__batch__' (inert), откат маркеров при отказе,
выбор сохраняется и сужается. Одиночный _setInboxHidden рантайма — тонкая
обёртка над той же записью; рантайм стал короче базы на 14 строк.
- сброс выбора при смене вкладки (клик и стрелка), поиска и «Только новые».
- i18n en/ru/de/fr: 9 ключей по ТЗ §8.
- смок demo/smoke_device_inbox_batch.mjs (AC1–AC8, AC10), юниты
test/device-inbox.test.mjs (AC2, AC4), 4 мутанта в реестре.
- docs: USER-GUIDE ru/en, FILTERING.md, оба changelog.
- monolith-baseline: hostRefs 4948→4959, bundleBytes 2500387→2505569 —
новая фича (состояние выбора и запись идут через порт хоста).
- bundle-budget: LAZY_EDITOR_GZIP_CEILING 245 400 → 246 600 (замер 245 610,
+1 190 Б gzip от базы 244 420); initial View 289 449 → 289 765 (+316 Б,
строки en-словаря), в пределах потолка.
Golden device-inbox-* изменятся ожидаемо — пересъёмка предрелизным гейтом
на Linux CI (§11.4).
Issue: #618
User-Visible: yes
План уходил base64 в WebSocket-кадре: файл больше ~3 МиБ давал кадр больше
4 МиБ, HA закрывал сокет до обработчика, и обещанные 8 МБ были недостижимы.
- бэкенд: HouseplanPlanUploadView (POST /api/houseplan/plans/upload), потоковый
предел MAX_PLAN_BYTES (read_bounded), общий writer store_plan_upload для view
и ws_plan_set (контракт WS без изменений);
- карточка: stagePlanFile/uploadPlanFile/renderPlanBackdropGuard в
backdrop-pick.ts для обоих рантаймов; PlanFilePayload хранит Blob вместо b64;
SVG больше предела — тост при выборе, растр — диалог #39 только с уменьшенной
копией, копия больше предела не попадает в staging, 413 называет предел;
- i18n toast.plan_too_large, backdrop.over_limit_body (en/ru/de/fr);
USER-GUIDE ru/en, CHANGELOG ru/en, docs/testing-notes (#617);
- тесты: test/plan-upload-limit.test.mjs, tests_backend/test_plan_upload.py,
test_ha_upload.py (#617), smoke_plan_upload_limit.mjs; три смока переведены
с b64/plan/set на blob/fetchWithAuth; мутанты plan-upload-*;
- база монолита: hostRefs +3 (общий хелпер плана вместо двух копий в рантаймах,
новый тост уменьшенной копии).
Issue: #617
User-Visible: yes
