Commit Graph
1073 Commits
Author SHA1 Message Date
Sameh Khamis bd99000433 Merge remote-tracking branch 'origin/main' into pr/wildlife-sightings 2026-11-05 16:00:37 -08:00
Denis Groshev 427463cc45 fix(news): list Local newest first
Local ranked its reports by severity first. A country's own news now
carries no severity, so in a conflict area GDELT's reports (severity 1
and up) could crowd every one of its stories out of the 40 listed.
Local is now newest first; severity still shows by colour.

Also: readTitle reads Conflicts headlines only, the economy category's
comment and the attribution popover name World News API's category,
and KNOWN-ISSUES names Core's ids for World News API's categories.
2026-11-06 00:45:33 +01:00
Denis Groshev 1e78ac931c Blame a GDELT export only while GDELT answers, and fix the review's notes
An export is given up on (never published, unreadable, or failing three
times) only while GDELT itself answers: its list of exports read, or
another export of the same round read. An outage that answers 404, or junk,
for everything now gives up on nothing, and a failing file earns one strike
a round rather than three within the first seconds' rounds.

Docs: News's share token is 9; territories without a country code are named
as themselves too; the changelog fragment credits one pull request. Unused
exports dropped from the rules and stale scaffolding from the ledger tests.
2026-11-06 00:45:33 +01:00
Denis Groshev 34c75ff5de Serve the News layer without a news server: GDELT, Liveuamap and World News API
With GEV_NEWS_ORIGIN unset, Core answers the News layer's routes itself.

Conflicts need no key: the last 24 h of GDELT 2.0 event exports, physical
violence only (CAMEO 18-20), one report per story (copies of one agency
story merged), placed where GDELT put it, a town the article's own words
name winning over a bare country, and titled by the article's page title
from the GKG export of the same quarter hour (the newest 3 h), else by its
address words. GDELT's actor coding is never shown. Exports are read up to
the newest GDELT lists; one that cannot be read is given up on without
holding the rest back, and an outage of two hours answers 502 rather than
an empty feed. With LIVEUAMAP_API_KEY, Liveuamap's mapped items join them.

Local reads World News API by the source country at the view centre
(WORLDNEWSAPI_KEY, at most WORLDNEWSAPI_DAILY_READS a day), with the
Conflicts reports placed inside the area. Disputed areas (11 from Natural
Earth, bundled) and territories without a country code are named as
themselves and read from no country's outlets; a country's outline leaves
them out.

Categories, relevance and severity come from shared rules; keyed answers
are cached and served while they refresh, and no key or address is logged.
2026-11-06 00:45:31 +01:00
Denis Groshev edb5837ae5 Name the Local area, say what each News section needs, and credit its sources
Local reads the area at the view centre (`/outline?at=`) and asks the feed
for its reports (`at`): the area is outlined faintly on the globe and named
at the top of the section ("News from Kazakhstan"; within 500 km at sea).
A disputed area is named as itself. A read that a newer look replaced, or
one left behind by a switch to Conflicts, is dropped, and a server without
the area route is asked once.

A section read from a key the server lacks says which key it needs
(503 `no_key`), that the key was refused (`key_rejected`) or that its
daily allowance is spent (`key_quota`); the other section keeps reading.
Liveuamap and World News API join the key registry.

Cards name the service a report came through ("via GDELT Project"), and
Liveuamap and Natural Earth are credited once their data shows. Economy is
a category of its own. Disputed-area labels are smaller and show the name
only; outlines keep up to 300 parts.
2026-11-06 00:45:31 +01:00
Denis Groshev 5bb489d1fb Put news on the globe, with a feed panel in the right rail
Events ▸ News draws the last 24 hours of reports from curated hotspots
(Ukraine–Russia, Israel–Gaza–Lebanon, Iran–Hormuz, Yemen–Red Sea, Sudan,
Sahel) and developing situations elsewhere, from any server that answers the
`/api/news` read routes.

- Globe: one marker per merged event, clamped to the ground and culled at the
  horizon; one beacon per hotspot when zoomed out and a pulsing beacon for a
  developing situation; hotspot outlines with disputed areas dotted and
  labelled by who administers and who claims them; picture cards for at most
  ten major or fast-growing events with two or more sources, never
  overlapping, switchable off.
- Panel: hotspot chips, a severity filter (significant and above by default),
  categories, minor news off by default, and a scrolling list of event cards
  with picture, headline, place, sources, age and the sides that reported it.
  "In view" lists what is near the map below 3,000 km once the camera
  settles. A card flies the globe to its event and opens it with a link to
  the publisher. Reduced motion stops the pulse and the flights.
- Core serves no news itself: the local server proxies GEV_NEWS_ORIGIN when it
  is set, and answers 503 unavailable otherwise, so the layer reads "not
  configured" once instead of retrying.
- Share token 9 (enabled-only); voice stays off for this layer.
2026-11-06 00:45:24 +01:00
Denis Groshev dd83812a21 Hold Cockpit regional news to the last 24 hours, and tell its states apart
Regional News showed "REGIONAL NEWS UNAVAILABLE" above populated rows and a
static "RECENT" footer over months-old articles. The status banner kept
`display: grid` when hidden, every non-empty answer carried the unavailable
text, and nothing limited or sorted articles by age.

- One window (regionalNewsWindow.js, 24 h): the local proxy asks Google News
  with `when:1d` and GDELT for 24 h, drops undated, older or future-dated
  articles, sorts newest first, and reports newsWindowHours/newsFetchedAt.
- The Cockpit labels four states: fresh (banner hidden, "LAST 24 H"),
  kept after a failed refresh ("CACHED · REFRESH FAILED"), no news in the
  window, and unavailable. The footer lives in the news state, so it
  survives page and tab changes, and the client drops out-of-window rows.
- `.cockpit-brief-status[hidden]` now hides the banner.
2026-11-06 00:45:19 +01:00
Sameh Khamis e3f668831c Merge branch 'main' into denis/gev-cockpit-entry 2026-11-05 14:01:31 -08:00
Sameh Khamis 63fd559496 Keep the ridden aircraft's layer when Contacts ends, and clear its rows from Live Signals
Riding an aircraft now adopts its layer the way a direct selection does, and
Contacts gives up a dependency the user adopts, so ending Contacts mid-session
(one of its layers off, or Contacts itself) no longer turns off the ridden
layer and ends the Cockpit. When Contacts goes off, Live Signals drops the
rows built from its snapshot instead of leaving stale distances marked
CURRENT, and the feed-off warning is no longer duplicated.
2026-11-05 14:01:17 -08:00
Said Turksever bd8ef09042 Merge branch 'main' into fix/keyless-unicode-search 2026-11-05 21:44:18 +00:00
Yehe LiuandClaude Opus 5.5 14f88b3e08 feat: Wildlife Sightings layer from iNaturalist and Acartia
Recent community wildlife observations, as a new Wildlife group in the
layer panel.

- Server: keyless iNaturalist proxy (v2 observations with only the
  projected fields; undocumented v1 UTFGrid density tiles, decoded on the
  server) sharing one paced queue (burst 4, ~1/s, 60/min, 8,000/day budget,
  up to four in flight), in-memory caches and upstream-only cooldowns;
  Acartia Salish Sea marine-mammal feed rounded to 0.01 degrees with
  comments dropped.
- Layer: grid counts drawn as a soft view-ranked density glow; real sampled
  sightings as taxon badges at their true positions with ambient photo
  thumbnails and labels; obscured records stay privacy cells; a clicked
  sighting stays pinned through camera moves and refreshes.
- Anchored photo card per selection with the single iNaturalist link;
  attribution in the layer row and the Data attribution popover.
- Filters (CONFIRMED, taxa presets) round-trip in share links with token 9.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-05 16:17:12 -05:00
Trygve Hemminghytt 47ea4aff1d Merge remote-tracking branch 'origin/main' into feat/datacenters-norway-refresh 2026-11-05 19:07:11 +01:00
German Khokhlov 067383fda8 fix(datacenters): drop the redundant seen set and test skipped against exports
The merge test now passes an exported newly tagged feature, so removing
'- set(fresh)' from the skipped report fails it.
2026-11-05 21:42:22 +04:00
German Khokhlov b5cef8daf4 Tighten the Rio comments, notes and credits
Name each Rio feed in its credit instead of repeating the licence, drop the
deprecation history from the bus note, list Rio among the feeds with a
route-name table, and shorten the route-table header and test comments.
2026-11-05 21:29:21 +04:00
German Khokhlov da00b096d2 Fade a vessel's callout by its card's painted opacity
The callout now asks for the opacity its own card paints with this frame,
keyhole fade included, and fades out as the card fades in. A card faded out
at the keyhole rim leaves the callout readable, and a card fading out after
leaving its source still hides it. The vessel card id is cached on the
record, and the allocation gate covers the lookup.
2026-11-05 21:21:02 +04:00
Said Turksever bc65c274da fix(geocode): keep a leading "the" in keyless name matching
foldPlaceName drops a leading "the", so "The Hague" over upstate New
York accepted Hague, NY as an exact nearby match. Split the script-aware
fold into foldPlaceText and use it for keyless toponyms; foldPlaceName
keeps its behaviour for the packs and street matchers.

Also: changelog fragment credits #1585, folding test pins the Latin
letters NFD does not decompose (ß, ø, ł), and CURRENT-STATE notes that
non-Latin names match only when spelled with their marks.
2026-11-05 17:13:00 +00:00
Mert Kaya e4fda9d4e6 fix(geocode): preserve non-Latin names in keyless search 2026-11-05 19:09:00 +03:00
German Khokhlov a67c7969fd fix(transit): read Rio speeds as the m/s SMTR publishes
Both SMTR feeds now publish Position.speed in m/s. Compared with how far
vehicles moved between reports on 2026-11-05, the published values ran
3.1x (buses) and 3.6x (BRT) below the observed speed, so the km/h
conversion divided real speeds by 3.6. Drop speedUnit and the conversion.
2026-11-05 13:48:11 +04:00
German Khokhlov e5dd7c71ac Merge main into feat/rio-transit-gtfs-rt 2026-11-05 13:35:27 +04:00
German Khokhlov 652f3c8ea1 Hide a vessel's detection callout while its own card paints
With Vessels and Detection on, a vessel with an ambient card on the map
also got its detection callout next to the bracket. The callout lane
paints beneath the cards, so its text showed through the card's plate
and covered the speed and course.

The callout now gives way only to a card the shared overlay actually
paints in the same frame. The overlay paints part of the published
cards (shared budget, collisions, UI chrome), so a vessel whose card
was dropped keeps its callout.

The detection lane paints before the cards publish their paint rects,
so getOverlayPaintRect has nothing for the current frame there. The new
isOverlayEntryQueued answers from the paint queue that the overlay
solves before painting. Vessel detection objects name their card in
cardSourceId/cardEntryId.
2026-11-05 12:53:17 +04:00
Itallo BrandãoandClaude Opus 5.5 dcd0f2cccd feat(layers): add shipwrecks layer (EMODnet + Wikidata)
A new Events layer maps shipwrecks with a card per wreck:
- EMODnet Human Activities heritage shipwrecks (about 6,500 in European
  waters, CC BY 4.0), skipping rows taken from wrecksite.eu.
- Wikidata shipwrecks with a Wikipedia article plus sinking events
  (CC0); an event folds into its ship for the date and death toll.
- A same-origin /api/shipwrecks proxy refreshes once a day and serves
  the last snapshot as stale for up to a week.

Markers are a boat icon coloured by era and sized by deaths; four era
chips filter the layer. The card follows the air-accidents layout
(deaths first). Text that names people is dropped; victims appear as
a count only.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-05 04:42:59 -03:00
Itallo BrandãoandClaude Opus 5.5 6172be3995 feat(transit): give vehicle icons a readable front, and none without a course
At the ~20 px the fleet renders at, the bus, tram, metro and unknown
glyphs were near-symmetric boxes: rotated along the course, nothing showed
which end was the front. Each now has a front built into its outline (an
arrow-like nose for the bus, a pointed cab for the tram, a chamfered cab
for the metro), so it survives the tint, the halo and the sensor profiles,
which keep only the outline.

A vehicle with no known course sits at rotation 0, where a pointed glyph
would claim it is heading north. Those vehicles, fleet and selected alike,
now get a symmetric outline (rail and ferry included) and switch to the
pointed one once a course arrives. The raster cache doubles to a fixed 72
variants.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-05 03:39:15 -03:00
Itallo BrandãoandClaude Opus 5.5 b37025414d test(firms): pin the refresh call count and read sources from production
Review follow-up for #1552:
- environmentProviders: assert exactly 3 upstream calls per refresh again,
  so a returning fourth source would fail the test.
- firmsProxy: build the test's SOURCES from server/providers/firms.js, so
  there is one list instead of a mirrored copy.
- firmsProxy: drop the unused fourth count entry.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-05 02:57:27 -03:00
Itallo BrandãoandClaude Opus 5.5 ad48305dac feat(transit): name Rio routes from SMTR's published GTFS
SMTR's realtime feeds carry internal route keys, so a Rio card read
"Route O0731AAN0A" where riders know line SN731. Join both feeds to the
route_short_name in SMTR's static GTFS routes.txt (CC BY 4.0, same
publisher and terms). The keys cannot be parsed: O0014AAA0A is LECD135.
A 2026-11-04 snapshot found all 282 bus and 21 BRT routes in service in
the table; an unknown key still falls through raw.

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-04 22:49:54 -03:00
Sameh Khamis 464c29a9ed Draw the last native world labels through the label overlay
The Directions A and B letters and a data pack's media label were the two
Cesium text labels left, kept by an allow-list in noCesiumLabels.test.mjs.
Both now publish entries to the shared world-overlay host, and the
allow-list is gone, so any Cesium label in src is a test failure.

Directions publishes the letters to the directions-markers source; the
markers keep only their points. A clicked endpoint's letter hangs over the
surface that was clicked. One placed without a click stays hidden until its
ground-floor cell resolves, then requests a render, so a still camera shows
it; a cell that settles with no height falls back to the ellipsoid. Each
position is computed once.

A pack resource handle gains label(entry). The application passes its
shared overlay host to SceneDirector through dataPacks.overlayHost; without
one a media pack fails to load rather than drawing a native label. Each
labelled resource takes a source from a pool and returns it on dispose, as
the host never removes a source, so two presentations of one pack never
clear each other's labels and repeat loads add no sources. A label written
after dispose is ignored. The media label is protected, so it always draws
as before.

Live (Denver, keyless globe): a clicked A and a programmatic B both anchor
on their dots over 1.58 km ground. Documented in docs/CURRENT-STATE.md and
docs/DIRECTOR-DATA-PACKS.md.
2026-11-04 12:30:31 -08:00
German Khokhlov 908cce915a Merge remote-tracking branch 'origin/main' into claude/hal-676-gev-fire-intensity-742bd1
# Conflicts:
#	src/i18n/locales/es.json
#	src/i18n/locales/ru.json
#	src/i18n/locales/uk.json
#	src/i18n/locales/zh-CN.json
2026-11-04 22:45:08 +04:00
Itallo BrandãoandClaude Opus 5.5 84c1fef8d3 fix(firms): stop requesting the retired Suomi NPP source
NASA stopped delivering Suomi NPP data on 2026-11-02 (FIRMS news page).
With a live MAP_KEY on 2026-11-04, VIIRS_SNPP_NRT came back ok with zero
rows while NOAA-20, NOAA-21 and MODIS returned 57k, 49k and 11k, so the
source only spent one transaction per refresh. The proxy now requests
NOAA-20, NOAA-21 and MODIS NRT. The "N" satellite label stays, so cached
or archived Suomi NPP rows still read SNPP.

Tests count sources from the list instead of assuming four, and a
regression test keeps the retired source out of the production list.

Refs #849

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-04 15:15:34 -03:00
Sameh Khamis 9bdeffeca3 Ask for heights through one service per scene
About 20 sites called scene.sampleHeight, globe.getHeight,
scene.clampToHeight or the MostDetailed variants directly, each with its own
idea of a miss (undefined, NaN, a throw). They now ask heightsFor(scene) in
src/queries/heights.js:

- now(point, { against, exclude }) answers from what is loaded, or null;
  against is 'surface' (what is drawn) or 'terrain' (loaded terrain only).
- clampNow(position) moves a position onto the drawn surface, or null.
- batch() and clampBatch() wait for the most detailed tiles: null per
  missed point, a rejection when the whole query fails.
- supported() reports { surface, clamp } without allocating.

Cesium still answers now() synchronously, so nothing changes on screen and
nothing is slower: in the live app 1,000 calls cost the same through the
service as direct (terrain 0.5-0.7 ms, surface 2-6 ms a call, run-to-run
noise larger than the difference). Every site already treated a miss as
"hold the last value or fall back"; a renderer that can only answer from a
cache or asynchronously can now plug in at one place.

- A Bhote Koshi evidence point whose terrain sample misses stays at its
  fallback elevation instead of reading the miss as sea level.
- The mesh floor sampler skips a scene without depth-texture support
  instead of spending its budget on probes that throw.
- The world annotation probe's trace reads hit/missed.
- A source scan forbids direct height calls outside the service.

The service is exported as ./queries/heights and documented in
docs/APPLICATION.md.
2026-11-04 09:24:30 -08:00
Sameh Khamis 02d492a079 Take pointer input through one dispatcher per scene
Every layer and tool bound its own ScreenSpaceEventHandler and picked the
same click again: with the usual layers on, one click ran about 26
scene.pick and 2 drillPick passes (120-215 ms in NYC with 18 layers). Pointer
input now comes in through src/input/dispatcher.js, one handler per scene.
A click is picked lazily and at most once per kind of query, and every
subscriber reads the same ClickEvent (pick, hit, id, ownerId, gesture,
drill, ground). The same click now costs 3 picks and 2 drills, 17-70 ms.

- Ambient onClick subscribers run only while no tool holds the pointer,
  checked once per event; tools subscribe with their lease. Moves and
  releases always reach ambient subscribers, so hover and drags end cleanly.
- Camera input and the viewer's stock select/track clicks are ref-counted
  holds (suspendCamera, suppressStockClicks). The dispatcher owns the
  controller's enableInputs, so a Cesium flight that ends no longer turns
  input back on under a drag or the cockpit.
- The draw tool, imagery box, CCTV gizmo, cockpit, directions and every
  selecting layer use it; bindTrackingClickGesture is gone (the gesture is
  on the event). A source scan forbids new handlers, enableInputs writes and
  stock-click removal outside the dispatcher.
- The sensor-style pick correction transit had now applies to every layer.
- Radio's 8-pick near-miss ring is one 17x17 windowed drill.
- A missed pick passed to pickContact is no longer picked again, and a
  touch tap's contact search shares the click's drill.
- A gizmo drag whose release Cesium never reported ends on the next press
  instead of leaving the camera held.

The dispatcher is exported as ./input with the pointer-lease functions, and
documented in docs/APPLICATION.md.
2026-11-04 08:48:51 -08:00
Itallo BrandãoandClaude Opus 5.5 1b0b4a10dc Merge main into feat/rio-transit-gtfs-rt
Resolves the DATA_SOURCES.md conflict by keeping main's rewritten Transit
and Radio bullets and restating the count for this branch: eighteen feeds
covering seventeen regions (the two Rio feeds share one region).

Co-Authored-By: Claude Opus 5.5 <[email protected]>
2026-11-04 11:53:46 -03:00
Sameh Khamis 83ec9473b6 Delete the billboard and model mirrors; models become layer-held records
Aircraft draw from GPU icon sets, so the billboard mirror, the billboard
treatment and the aircraft visual-state module had no callers left. 3D models
are now records the layer holds under the contact's id (modelRecord.js):
show and placement are written through showModel/placeModel and read back
from the record; renderer-measured values stay on record.model.

The tracked-model click check only matches when a tracked model exists, so
an id-only pick is never mistaken for a click on the tracked plane.
2026-11-03 22:36:27 -08:00
Sameh Khamis c9f1fcd425 Draw FIRMS detections from a GPU icon set 2026-11-03 22:09:57 -08:00
Sameh Khamis 8aaf5564d7 Draw CCTV cameras from a GPU icon set 2026-11-03 21:47:16 -08:00
Sameh Khamis e0eb771217 Draw live AIS vessels from a GPU icon set 2026-11-03 21:16:19 -08:00
Sameh Khamis 816002e3b9 Share the aircraft 3D-model regime between flights and military 2026-11-03 20:55:09 -08:00
Sameh Khamis b027bde6ad Draw military flights from the shared GPU icon set
The flights icon planning and presentation move into
`createAircraftIcons` (src/data/aircraftIconLayer.js), and the motion
regime both layers dead-reckon from into src/data/aircraftMotionRegime.js.
Flights keeps only its grounded-floor pass on top.

Military now draws every contact as a row of that icon set, like
flights: the billboards, the cull positions and the rotation-pass
signature go. Tracking hides the row and untracking re-plans it from
the tracked course. Queries read the drawn row position and judge the
horizon as the shader does. Snapshots that arrive before init or after
destroy are ignored, and per-contact lookups find no contact then
instead of throwing.

With about 120 live military contacts, preRender p95 drops from
0.7-0.8 ms to 0.2 ms on the same views. Render time and frame gaps are
unchanged or slightly lower.
2026-11-03 20:30:43 -08:00
quanticsoul4772 0b91938f7d Merge branch 'main' into upstream/datacenter-ground-sampling 2026-11-03 20:30:27 -08:00
Sameh Khamis fa3a3a3bca Draw civil flights from an icon set moved on the GPU
Every flight is now a row in the layer's icon set. Its motion segment
follows the same regime `_deadReckon` picks (interpolate between fixes,
warm up backward from the oldest, coast forward from the newest), and
one shared `_motionRegime` makes that choice for both. The shader moves
every untracked contact between polls, so the 12 Hz fleet tick no
longer touches each billboard. It now only re-plans expired rows,
floors grounded contacts, advances focus while needed, and runs the 3D
model regime.

Grounded contacts the floor lifts are pinned to a hold at the floored
position. Their nose glides on the GPU. Contacts within 50 km re-floor
every tick, and farther ones take turns about once a second. Low
contacts are judged for the horizon at no lower than 12 m above the
ellipsoid, as before. Below hover speed a contact's nose holds. A row
whose fleet model is drawn keeps its focus updated, because the model
reads it. Untracking re-plans and re-floors the icon from the tracked
course.

On the same views with about 6,700 OpenSky flights, frame gaps at p95
drop from 28-40 ms to 17 ms, and the fleet tick from 16-37 ms to about
2 ms at most.
2026-11-03 19:40:39 -08:00
Sameh Khamis 37dd4058c1 Add columnar icon sets drawn by an instanced Cesium primitive
Icon rows live in typed-array columns: one packed motion segment (hold,
lerp or arc, with a course glide), style, colour and pick ids per row.
The primitive draws every row in one instanced call, dead-reckons in the
vertex shader, matches the billboard treatment the fleet layers use, and
picks and drill-picks like a billboard. Cyber sonar covers icon sets.
2026-11-03 18:35:12 -08:00
Sameh Khamis c66f8f039b test(cctv): make the oversized Houston catalog fixture valid JSON
The streamed oversized body was invalid JSON, so the loader returned
nothing whether or not the byte cap applied. Pad a valid catalog with a
usable camera instead, so only the cap can refuse it.
2026-11-03 15:04:05 -08:00
Sameh Khamis f120989a2b docs(cctv): say the Houston pack needs TxDOT's written permission to enable
TranStar's Content Usage terms prohibit copying, distribution and
external framing of its images without TxDOT's prior written
permission. The opt-in text only said the pack was off unless
CCTV_HOUSTON_ENABLED=1, without saying what an operator needs before
setting it. Say "enable only if you hold TxDOT's written permission" in
.env.example, the README, DATA_SOURCES.md, CURRENT-STATE and the
changelog, matching the NYC and Florida packs, and fix the column of the
.env.example comment.

The houston-cctv credit now reads "(Houston, when enabled)" and links
the Content Usage terms, like the Florida credit. Where the docs said
frames are "never stored or redistributed", they now say what happens:
frames are relayed on demand, buffered only per request, sent with
no-store and not archived.
2026-11-03 15:03:37 -08:00
Sameh Khamis 8cd0cc82f5 fix(cctv): read TranStar dir as the camera's side of the road, at low confidence
TranStar's `dir` field is the side of the freeway a camera sits on, not
the way it faces. TranStar's own map script (translateCameraDirection)
shows it as "Traffic closest to the camera is traveling": South is
eastbound, North westbound, West southbound and anything else
northbound. The live catalog agrees: I-10 cameras are North/South and
I-45 cameras East/West.

Treating `dir` as a high-confidence facing pointed most Houston cameras
across or away from the road they watch, and dropped the "estimated"
marking. Map it to the travel direction instead (S 90, N 270, W 180,
E 0) as a low-confidence prior with the low-confidence pose, and keep
the id-hash fallback for None or unknown values.
2026-11-03 15:02:30 -08:00
Sameh Khamis 09bdc314d0 Merge branch 'main' into fix/houston-cctv-pack 2026-11-03 15:01:13 -08:00
Sameh Khamis 08985d5083 Keep installation and local GeoJSON pick state off Cesium entities
Installation selection fills and outlines record the installation they
belong to in the layer's installationMarkers store instead of an
installationId expando, and the click handler resolves them from there.
Named-marker points keep their { installationId } pick payload, which is
the app's own object, so sibling layers treat a point click as before; the
marker remembers its record id for visit() instead of reading the point.

The local GeoJSON package tracks the entities each layer loaded in a
per-instance set instead of a __localLayerId tag, and passes the stem
base to registerEntityContext as targetPosition instead of hanging it on
the entity; FIRMS does the same, and voice targeting reads the context
record. QA scripts find a layer's features by their context id.
2026-11-03 14:57:31 -08:00
Sameh Khamis 585392cbcf fix(drone-airspace): refuse cross-site requests to the FAA proxy
A foreign Origin or Sec-Fetch-Site: cross-site request was answered and
counted against the client's 60-a-minute quota, and could start a refresh.
Run the keyless same-site checks the other map routes use before rate
limiting and cache acquisition; reverse-proxy forwarding headers stay
accepted. List the route with the keyless map routes in SECURITY.md.
2026-11-03 14:35:14 -08:00
Sameh Khamis 8e63420f06 fix(drone-airspace): show stale FAA snapshots as stale and cap them at a day
The source dropped the proxy's stale flag and fetch time, so the layer row
read nominal and freshly updated through an outage, and the proxy served its
last snapshot indefinitely. Carry stale and fetchedAt through the source as
Fire Perimeters does, date a stale snapshot by its fetch, and stop serving a
snapshot more than 24 hours old.
2026-11-03 14:34:01 -08:00
Sameh Khamis 6bdedd8523 fix(drone-airspace): refuse empty or incompletely paged FAA snapshots
An upstream answer whose zones all fail validation normalized to an empty
list, and paging that stopped at the page limit while the service still
flagged more records was accepted as complete. Both were cached as a fresh
snapshot for an hour and replaced the last good one. Treat them as failed
refreshes so stale-on-error keeps serving the previous snapshot.
2026-11-03 14:33:10 -08:00
Trygve Hemminghytt 1f0a8b5f16 fix(datacenters): report newly tagged features osmium cannot export
A still-tagged feature without exportable geometry already keeps its old
line. A newly tagged one had nowhere to go and was dropped silently; it is
now listed as skipped with a warning, so a refresh never loses a site or
hides a broken one.
2026-11-03 22:58:09 +01:00
Sameh Khamis 0889cd0cb7 Answer earthquake, CCTV pick and tracked sprite reads from the app
Earthquake analyst records come from the events the layer drew, gated on
its own enabled flag, instead of reading each entity's position and
property bag back; the discs no longer carry the bag. CCTV coverage rays
and projection planes record their camera id in a layer pick store
instead of a cctvCameraId entity property, with the same ownership checks.

The tracked flight, military and local ADS-B billboards record their size
and distance scaling as the entity record's sprite, which the close-range
camera guard and the screen annotations' tracked footprint now read in
place of the billboard's property getters.
2026-11-03 13:32:53 -08:00
Sameh Khamis 05904edfc9 Read aircraft model placement from the app's mirror
Flights, military and local ADS-B now write their models' show and world
matrix through setModel in src/data/modelMirror.js, and read them back
through modelState: the fleet and tracked handoffs, the tracked visual and
trail anchors, detection anchors and modelAnchorWorld. A matrix is copied
into the app's record and into the model's own matrix, so the shared scratch
can be passed directly and models still never share one matrix object.

What Cesium measures (ready, computedScale) is still read from the model,
and colour stays live as on billboards, because cyber sonar writes it too.
Test seams mirror the models they install, as they do billboards.
2026-11-03 12:58:50 -08:00