BenStullsBets
ec660c4880
Root cause: the steady-state loop (#vid-loop) restarts only via a narrow
timeupdate window (last 0.08s -> seek to LOOP_TAIL_S), with native loop off.
Headless Chromium reliably lands a final timeupdate in that window, but under
real-browser decode/GPU load a timeupdate can skip past it; the clip then fires
'ended' and freezes on its last frame ('videos aren't looping'). There was no
recovery path. Add an 'ended' listener that seeks back to the tail and resumes,
guaranteeing the loop regardless of the timeupdate race. No change to normal
playback. Proven by a Playwright regression test that forces the ended state.
Co-Authored-By: Claude Opus 4.8 (1M context) <noreply@anthropic.com>
Simulator E2E (Playwright)
Browser end-to-end tests for the Human Experience Filter simulator. The §9 pipeline requires an E2E browser tier for any UI surface; this is that tier.
What it covers
tests/altitude-lock.spec.ts drives the real simulator and asserts the
altitude-morph behavior:
- Zooming to a new altitude plays the morph that matches the clip it lands on
(forward zoom-in, and
.revzoom-out). - After landing, the clip is locked — it does not change while parked (no secondary swap, no re-roll).
The played morph paths are read from window.__hefMorphs (a diagnostic array
populated by advance() in static/app.js), which is reliable even when a morph
is served from the in-memory blob cache. The locked clip is read from the
Dev-Mode #scale-name readout.
Running
Prerequisites: the project Python venv on PATH (the Playwright webServer boots
uvicorn simulator.app:app from the repo root), Node, and the Chromium browser.
# from this directory (simulator/e2e)
npm install
npx playwright install chromium
npm test
playwright.config.ts starts uvicorn on port 8099 automatically (reuseExistingServer
locally) and points the tests at it.