From review:
- An answer the view picked ("Springfield" over Illinois) is reused or
shared only within the same ~2° view cell, so asking again elsewhere no
longer gets the first state's area.
- Admin names compare with accents folded on Latin letters only: "Quebec"
matches "Québec" and "Cordoba" asks between Argentina and Spain, while a
Thai or Devanagari mark still tells names apart.
- A Natural Earth name used by several features is one place with every
part when they touch (the Great Rift Valley, the Lau Group across 180°)
and a question when they are far apart (Flores); natural places can now
be candidates.
- The AREAS rule and the area tool descriptions are tighter, leaving a
~270-character margin under the 55,500 prompt ceiling, and resolve_area
no longer names osm_query when that tool is not offered.
- A zoom during a pivoting tilt now scales the range the tilt holds,
as it already did for orbit, instead of being undone on the next
frame while reporting success. The zoom limits no longer move a view
the wrong way: a zoom in leaves a view closer than 80 m in place, and
a zoom out leaves an orbit beyond 5,000 km where it is.
- A grab, touch or wheel on the map while a voice fly-to is in flight
drops the motion queued behind it. Cesium lands the flight anyway,
so the orbit or tilt would otherwise take the camera back. Chained
fly-tos in one reply keep the earliest snapshot.
- A pivot tilt toward the horizon stops lowering the camera at 150 m
above the pivot and tilts the rest in place, so a close view no
longer ends inside tiles or a hillside.
- pitchDeg is an absolute angle. A direction that contradicts it is
refused, judged from the pitch the swing starts at, and the tool
description says so.
A radius the user gives always draws a buffer, but the enclosing outline
was still looked up first (up to 6 s and a guarded Nominatim request) and
then thrown away. It is no longer requested. The slow-outline test now
checks the default 250 m fallback it was meant to cover.
Three resolver fixes from checking the review focus:
- A natural region named like a country or state no longer hides it.
"India" was the peninsula, "Japan" an island group, "Florida" the
peninsula and "Punjab" a desert, so bare "Punjab" never asked which one.
The region still answers when asked for by level, and when no unit truly
has the name ("Great Britain" finds the United Kingdom only by alias).
- Same-named units get their own candidate: "Washington" offered WA and DC
with one id, so either choice was DC and a DC lookup overwrote a stored
Washington State. Ids now carry the region code, and choices show it.
- A bundled unit that only shares the name ("Azad Kashmir" for "Kashmir")
no longer ends the search; the outline ladder still looks, and the
mismatch is kept if it finds nothing. Retired zones stay refused.
osm_query handed its places to the analyst with no presentation, so "which of
those have emergency?" after "hospitals in Kathmandu" was answered as "anywhere
in the loaded data". The handoff now carries the scope the places were found
in ("inside Kathmandu" with its areaId, or "in view"), as an analyst count over
that area would.
Every name ending in "Zone" was kept without a level and refused as a retired
administrative identity. That is right for Nepal, which replaced its zones in
2015, but zones are current units elsewhere: Ethiopia's "Gamo Zone" was
refused. The selector now keeps the verified level unless display_name places
the zone in Nepal or names no country, and the client no longer reads "Zone"
as a retired level when the source verified one.
The cancelled and failed branches ran before the say-line branch, so a
cancelled panel change no longer said "Cockpit exited; the voice panel was
left as you last set it", and AREA_TIMEOUT / OSM_TIMEOUT got "tell the user it
failed" instead of the tool description's "retry once". A cancelled or failed
result with a say line now has it spoken, and those two timeouts are retried
once. Both providers share this instruction.
"Fly to Paris and orbit it" reaches the app as two sibling tool calls,
which OpenAI runs at once. With a place that needs a lookup, the orbit
started over the old view and the late flight then replaced it; with a
known place, pan and rotate cancelled the flight already in the air.
Orbit, pan and rotate now wait for a pending voice flight the same way
tilt does: up to 30 s, and not run (saying so) if the flight is
cancelled, finds nothing or times out. Stop and rotate_globe do not
wait. Orbit keeps its own arming for flights the voice runner does not
track.
"Fly to London and tilt the camera" reaches the app as two sibling tool
calls, and OpenAI runs them concurrently. The tilt's navigation step
cancelled the flight already in the air, so the camera tilted where it
was; when the place still needed a lookup, the late flight replaced the
tilt and London came out top-down.
The voice runner now tracks its latest flight. A tilt sent while that
flight is still looking up or flying waits for it to land (up to 30 s),
then tilts at the destination. If the flight is cancelled, finds nothing
or times out, the tilt is not run and says so.
A voice fly-to frames a city top-down, at pitch -90. The tilt clamp
rejected any step that left the pitch below -89, so the first frame of
"tilt up" stopped the motion and "tilt down" was refused: from the view
voice lands on, no tilt ever moved the camera. A bare "tilt the camera"
also failed because tilt required a direction.
- Clamp tilt only in the direction of travel, so -90 can tilt up.
- Swing the camera about the ground under the crosshair, so the place
stays centred and the heading is kept. A tracked view still tilts in
place, and a pivot tilt yields when tracking takes the camera.
- Add an optional move_camera pitchDeg for "tilt 45 degrees"; a tilt
with neither angle nor direction settles at -45 (oblique view).
- Re-derive the voice tool schema pins for the new argument and wording,
and keep the Realtime prompt under its 52,000-character budget.
Picks up the Radio Next/Previous fix (#1474) and the Codespaces docs
(#1515). The one conflict is the Radio cycle assertion in
gevActions.test.mjs: main's new fallback flag is kept in this branch's
layout.
- strip control and format characters from OSM tag values, keeping the
zero-width (non-)joiners Persian and Indic names spell with, and cut
them at 120 code points so a name cannot steer or flood the model
- adopt the nine new geometry test files in format-scope.json; five of
them change by formatting only
Both stack parents (#837, #842) are on main, so the voice geometry branch
now builds on main directly. Conflict resolutions keep main's behaviour and
add the geometry work on top: osm_query gating on every provider, restored
admin identity for Nominatim outlines, merged analyst follow-up memory, and
the cancelled Globe View reset instruction ahead of the generic cancellation.
Resolve the CURRENT-STATE.md layer table by keeping main's realigned
table and adding this layer's row. Main added an Arabic locale, so
translate the 12 layer strings and list ar.json in the drone-airspace
package.
Draw and Import used to be reachable by voice through DISPLAY. set_panel_open
now takes 'utilities-panel', and "utilities", "draw", "drawing" and "import"
reach it. This adds one value to the Realtime tool schema, so the pinned
payload digest, enum list and serialized length are re-derived with a note.
The show_data_layers_menu test now finds a row only inside the list that
really holds it, so dropping #utilities-toggles from the lookup fails it.
main moved the panels to Vue views and a one-open-panel rail, and removed
the two templates this branch edited. This merge carries the branch onto
that architecture; it adds no behaviour of its own beyond the port.
- The Utilities panel is a RailPanel in LayerPanelsView with its own rail
item (handyman). LayerPanel renders the Directions rows into its list with
LayerRows, and falls back to Data Layers without it.
- Draw leaves DisplayPanel for DrawControls, DrawMarkList and DrawSwatches,
which render what the draw tool publishes to the display store and call
its actions, as GeoImport does. drawMarks.js words the rows; the DOM-built
drawMarkList.js is gone. A row's fly-to is an immediate navigation like
Import's ('drawing' gets its Cockpit refusal), the name field keeps a
draft through v-field-value, and the colours are one radio group with a
single tab stop.
- UiSegmented draws an option's icon, named by its label, for the shape
buttons.
- The catalogs take main's languages with this branch's strings, Arabic
included, and the tests follow the views to their new homes.
Each overlay is now its own Infrastructure row instead of a chip on one
shared row: sea-marks (token 6) and ski-pistes (token 7), both
enabled-only, built by createOsmOverlayLayers() from one descriptor
list. Drops the option group and its share-codec defaults, gives each
layer its own name, voice entry and CURRENT-STATE row, and shifts the
token-ledger fixtures to the next free digit 8.
Main moved on since this PR opened; the overlay layer is ported to the
current contracts:
- Layer-state token 6 (j belongs to transit; new single-character
tokens are digits) with a reservation-ledger row, and the option owner
renamed to the layer id so the compact share codec can encode it.
- Constructed per catalog by createOsmOverlaysLayer() in
constructCatalog.js; src/standalone/data.js no longer registers layers.
- Drapes through maps/imageryHost.js, so the overlays show on Google 3D
(tileset imagery) as well as the globe stacks, and follow map-stack
switches. Imagery layers are created on enable and destroyed on
disable/destroy.
- Translated chip labels and layer name in every locale, LAYER_NAME_KEYS,
Infrastructure panel group, voice manifest entry.
- Attribution: OpenSeaMap tiles CC BY-SA 2.0, OpenSnowMap credit as
required, OSM ODbL credit for both; bulk-download note in DATA_SOURCES.
- CURRENT-STATE row, changelog fragment, share-codec absent defaults,
updated token-ledger fixtures and voice schema pins.
Chat sits next to the mic in the command dock. Replies are silent and never
request the microphone; the first message connects with the selected provider
and runs the same map actions as voice through each provider's text API
(OpenAI OPENAI_CHAT_MODEL, Gemini GEMINI_CHAT_MODEL via generateContent, which
never starts Live). Only one mode runs at a time.
- Chat panel rendered by the voice control presenter: conversation, recent and
quick-command suggestions, Up/Down history, attachments (map-layer
references or small UTF-8 text files, sent only with the message)
- Chat and voice panels sit above the dock or dock into the right rail's
column; minimized bars sit centred above the dock, and height animates on
swaps
- A centred panel steps down to its bar for the Location and Visual Presets
trays without moving focus, rests above an open tray, and docks right while
expanded over one; it returns to the centre once the trays close
- Minimized voice previews the last sentence from each side; the empty Voice
Preview suggests things to say
- Settings opened from chat keep chat's frame and show the shared AI provider
and quick commands; voice Settings show Back to voice
- Closing chat ends a typed session; the voice panel's close turns voice off
- The dock widens its voice control for the chat button and restyles its
wings, which show icons only at 900px and below; assistant panels share the
left panels' glass, and Cyber keeps its own surfaces
- Server text-chat handlers with body, time and per-IP limits
(GEV_RATELIMIT_*_CHAT_PER_MIN), documented in SECURITY.md; tests and all
eleven locales
Co-authored-by: Darren Dignam <[email protected]>
Han and kana have no word edges, so a CJK forbidden word now counts anywhere
in the reply. The proxy no longer counts words: the model reads Rue de Rivoli
as one, so only the 80-character word-boundary cap remains.
Also restore src/voice/gevActions.test.mjs, reformatted by mistake in 818d2c82.
- The proxy keeps combining marks, in-word apostrophes and name dots
(ar vowel signs, fr d'Orléans, ja ・), normalizes to NFC, caps every
summary in characters and cuts at a word boundary.
- feedState is nullable instead of a `none` token, and the summary has at
most five words: "exactly five" made the model pad short views with
"no data" or "unavailable".
- The HUD caches only the model's words and adds the current warning on
every paint; an RTL locale gets a leading right-to-left mark.
- The echo check also rejects an all-Latin reply in a non-Latin language,
translated layer ids and, with every feed healthy, the UNAVAILABLE word;
places are matched by word and one-word layer names are skipped in
Latin-script locales.
- The warning tag names the worst layers first; a partial analyst card
shows one partial chip.
main moved the Radio markup into Vue views and rebuilt the UI tests on
stores. Resolved:
- src/ui/templates/context.html: deleted in main; the new privacy note now
lives in src/ui/views/panels/RadioPanel.vue.
- src/ui/radioControls.test.mjs: main's store fixture; the Next/Previous
autoplay test drives store.act('cycleRadio').
- src/voice/gevRealtime.test.mjs: main's createTestVoiceUi; the stop and
select handoff cases stay one table-driven test.
- src/data/radioNextState.test.mjs reads the published gev-radio view.
- src/i18n/locales/ar.json: the new note translated for the Arabic locale.
Layers whose data has a time of its own follow the page's world clock from
the start; live feeds such as flights, military, vessels and transit stay
live. Launches register the span of the launches loaded, so turning them on
brings up the timeline. Each timed layer registers its coverage once and
reports none while off, telling the clock after each step has settled. The
launch feed keeps its own window, and an orbit ring is drawn again once the
world time has moved far from when it was drawn.
Choosing a language in the picker now applies it at once. The choice is
stored and ?lang is kept in step through replaceState, and the picker keeps
focus.
Code that writes translated text itself re-presents it on a switch:
onLocaleChange() lets an owner listen, removed at teardown, and status
writers hand over a function of the locale (voice cards, scene director,
SDR, draw hints, imagery split, map sources, Bhote Koshi media) so the
words are rebuilt rather than re-sent. The HUD, cockpit, radio, CCTV,
launch, awareness and credit texts follow the same hook, and
applyDocumentTranslations can be applied again without undoing text code
wrote since.
The unsaved-work registry and persistLocaleAndReload existed only to guard
the reload, so they go, with their catalog entries.
A panel's collapsed, dock-pinned, dock-pinned-top, layout-auto-collapsed
and accordion-collapsed classes are drawn by its root's v-panel-state from
the gev-panel-chrome store. PanelChrome, its restore, the layout
controller, both rails and the side accordion present each panel's state
there instead of writing its classes; the rails call onCollapse(panel,
{collapsed, auto}). The directive applies a change in a synchronous
watcher, since the rails measure a panel in the task that opens or closes
it.
Destroying the chrome inside Cockpit now puts back the panels Cockpit's
entry closed while it can still present them. Without a shell, the voice
tools' set_panel_open reports that there is no panel to open. QA scripts
present the states they set instead of writing the classes.
PointerReticle keeps the turn's timers and tool tracking and presents
where the user pointed to the gev-pointer-reticle store, one entry per
reticle; the command dock teleports each to the body with the same
class, aria-hidden, hidden state, target and position. The voice QA
reads the card and reticle after Vue's flush, and the application guide
says leaving out the command dock leaves out the voice control and its
reticle.
createVoiceControl now returns a presenter instead of building the
control with innerHTML: the session, cost, input, card and panel owners
write its state by part and listen for interactions, and VoiceControl.vue
renders it in the command dock, between the location bar and the visual
presets as before. Focus moves wait for the render; the waveform's bars
publish apart from the control and only when they change; the minutes
field keeps a draft until its owner writes it. Owner tests use a real
presenter on their own store, and view tests render the control.
initKeySetup keeps the setup requests, the ChatGPT OAuth flow, the pasted
field values and the dialog's keyboard, and publishes the POWER UP chip,
the dialog, its key rows and the status line to the gev-provider-settings
store. ProviderSettingsView renders them and sends clicks and typing to
the connected owner; in Cyber the chip is teleported into the top
toolbar. A successful save renders the rows afresh so no field keeps a
key, and the status line announces a repeated message again. The voice
action opens and closes the dialog through the same controls.
Every scrolling surface carries the kit's .ui-scroll, which
draws the thin themed scrollbar from --scroll-thumb and --scroll-track; the
Context panel and others drew the platform's light scrollbar because the
page declares no dark color-scheme. The per-panel scrollbar rules go; Cyber
sets the two tokens and Cockpit's HUD windows keep their accent tint.
Selects draw .gev-field's own chevron instead of the platform arrow, the
voice selects join .gev-field, number fields keep dark spinners, and the
Street Level range uses the shared slider in its green.
reconcile voice outlines with main f13c60c, retaining German/Japanese,
Operator HUD retirement and Context/Radio cleanup, named-building
containment, courtyard-hole, non-Latin identity, attribution, cancellation
and antimeridian repairs
keep outline callouts above the open voice preview when an above-panel
slot fits without colliding with another callout,
and clip the compact dock location label on narrow screens