The probe exposes settled() and the first-frame outcome; scripts/qa/ready.mjs
waits for drawn frames with the camera at rest and for a pick at the click
position, with a deadline and its reason in the error. The strict scenarios
use these instead of fixed pauses and loading-screen polling, and the
hidden-switch scenario checks how the loading screen was lifted instead of
how long it took.
Each switch scenario now records a draw ledger per engine and runs a
contract monitor; both write to the scenario's evidence file only, with an
empty allowlist until browser runs show what differs on purpose.
Navara now matches Cesium where the review found it diverging: a flight
holds the camera inputs, camera.changed counts turns as Cesium does,
render requests made during a frame survive it, follow pinches zoom,
ground widths follow the pixel ratio and Cyber sonar reaches glyph caches
and models. On Cesium, a staged imagery swap keeps loading tiles, setView
replaces a running flight, and a cleared weather detail goes away.
The contract gains MoverSet.failed, maps.terrainRevision,
render.maxTextureSize and surface.sampleTerrainMostDetailed on every
engine, so layers retry failed models, key road heights by terrain,
size weather textures to the GPU and place Bhote Koshi cards on detailed
terrain.
Engine switches keep the live view, follows, focus, layer options and
cards: a pick hands over the exact pose, applies no address options and
pans no carried-over card; a history visit still applies its entry.
Layers put GhostMaps pins, cable landings, OSM Places and directions on
the measured surface without per-frame sampling, draw the Cockpit contact
dot, hide disabled CCTV on attach, and keep ALPR running outside secure
contexts. KML imports draw raised polygon fills, rotated and coloured
ground overlays and north-referenced icon headings, and report files
with nothing to draw.
QA scripts read the contract instead of Cesium internals (test:track is
109/109 again), wait for the camera to rest before clicking, and give
cold switches and parked engines realistic time. Docs follow the code.
- src/geo drops the ported debug argument checks and the tests that
bitExact already covers; package boundaries accept a directory entry,
so src/geo/ is listed once per group.
- Cesium no longer carries movers and lines nobody draws with: they are
optional engine capabilities. The fake engine and the Cesium camera use
the shared cameraMath; unused contract tables, clock.ticked, the
'finest' surface detail and Navara click statistics are gone.
- ALPR keeps one overlay core and one presentation model for both
engines, bikeshare one station bookkeeping, the Navara ground effects
one definition helper; image atlases, view centres, mover banks and
colours are shared.
- The legacy switch route, the clock offset and rate controls and other
unused exports are removed.
- The acceptance runs the Navara draw, geo import and scope checks;
scenarios use the harness switch and page helpers; the main chunk
check keeps the size budget only and runs once in CI.
The long soak's heap slope counted the per-aircraft dead-reckoning history (POSITION_HISTORY_LIMIT fixes for 10 000 aircraft, about 14 MiB over the first 2.5 min of polls) as switch growth: a run with the same workload and no switches grows as much and then levels off. With --assert the soak now waits until the flights feed has been read more times than the history keeps before it switches; over 500 switches the heap then grows 1.07 KiB per switch.
qa-strict-context-loss records, only when no non-blank frame came, how many frames the running and the watched engine drew in 3 s and how much of the last one differs from its first pixel.
Every scenario runs through scripts/qa-strict-harness.mjs: each page, frame and worker gets an ordered block list before it runs (paid hosts and routes first, then only the preview, declared fixtures, keyless probes and cached free tile hosts), switches with real input on the picker, and records its assertions per layer and engine. The runner reads the catalog from the live page, fails any layer without an assertion on either engine, can build and serve a keyless checkout itself, and adds rapid, hidden-tab, context-loss, failure-injection, cold-chunk and soak scenarios. The main chunk gets a 1% size budget.
German Khokhlov