Skip to content

Draw instrument footprints and sampled trails in the Cesium Study pane - #72

Open
darthvader58 wants to merge 2 commits into
code-lab-org:mainfrom
darthvader58:feat/conical-scan-footprints
Open

darthvader58 wants to merge 2 commits into
code-lab-org:mainfrom
darthvader58:feat/conical-scan-footprints

Conversation

@darthvader58

@darthvader58 darthvader58 commented Nov 16, 2026 •

Copy link
Copy Markdown
Contributor

Instrument footprints in the Study pane now come from native TAT-C collect_ground_track for supplied basic, pointed and conical instruments. Conical-only studies draw the actual scan band or sector instead of showing only the satellite track. Polygon holes and multipart/antimeridian boundaries are preserved, and invalid samples are hidden according to TAT-C's footprint-center valid_obs check.

An optional Sampled footprint trail draws past valid sample polygons within the shared trail duration. All footprints use the nearest native display sample to avoid morphing vertex topology at poles and the antimeridian. Sampled footprints do not establish continuous viewing duration or a dissolved observed area. Point-event accounting is unchanged, and the existing sample budget now covers every instrument.

The legacy inline MapLibre tool is unchanged in this PR. Native compute_ground_track dissolved-area support and migration into the Study pane are tracked separately in #70. The capability inventory now corrects its previous attribution of the legacy map to that native function.

Closes #71
Closes #69

Validation:

  • Python 3.12: 560 tests passed.
  • UI/tessellation: 58 tests passed, including real-TAT-C fixtures, scan holes, layer toggles, validity, sampled trails and clock behavior.
  • Real browser/WebGL: full conical scan, hollow interior, sampled trail and shared clock checked in Cesium.
  • git diff --check passed.
  • Required behavioral eval and --compare completed. Both the prior run and this run scored 19% with two infrastructure failures. Recorded CelesTrak responses fail in the cache streaming path (Response.raw is None), so these runs do not provide a clean model-behavior baseline; keep this PR draft until that independent eval harness problem is resolved and the comparison rerun. Streaming probe passed.

Polar rendering is also corrected here: wide native polygons are partitioned for Cesium, true TAT-C boundary arcs omit artificial seams, and native precision/holes are retained. This shares the geometry helper and 19 native polar/antimeridian tessellation regressions with the small stacked follow-up darthvader58#1 (for migration PR #74). No MapLibre migration code is included in this PR.

@darthvader58
darthvader58 marked this pull request as ready for review November 17, 2026 02:42
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Use collect_ground_track for instrument footprints in the Cesium Study pane Draw conical-instrument geometry on the Cesium study pane

1 participant