## Summary
- Convert `streamingEncoder.ts`'s safety timer from a total-render hard cap to a per-frame inactivity timeout
- Reset the timer only on `accepted === true` writes — buffered writes don't count as consumer progress
- Update the `ffmpegStreamingTimeout` config doc to reflect the new semantics
## The bug
The timer was set once at spawn and fired SIGTERM unconditionally at `ffmpegStreamingTimeout` ms — turning a "FFmpeg is hung" guard into a hard cap on total render duration. Slow-but-progressing captures (CI runner under load, large compositions, slower compositor paths after [#838](https://github.com/heygen-com/hyperframes/pull/838)'s always-clip change) regularly exceeded the 600s default and were killed mid-encode. The symptom surfaced as:
```
Streaming encode failed: FFmpeg exited with code 255
video:NNNkB audio:0kB ...
[libx264 @ ...] frame I:3 Avg QP:12.91 size: 73263
[libx264 @ ...] frame P:431 Avg QP:14.72 size: 31633
...
[libx264 @ ...] kb/s:7661.05
Exiting normally, received signal 15.
```
libx264 had encoded most frames cleanly; SIGTERM arrived during the encode, libx264 printed its end-of-encode stats, and Node observed a non-zero exit. The `audio:0kB` in stderr is incidental — `streamingEncoder` is video-only; audio is muxed later in `assembleStage`.
Downstream reproduction: `style-13-prod` fails deterministically in `heygen-com/hyperframes-internal` CI after bumping `@hyperframes/producer` from 0.6.7 → 0.6.10. Bisects to #838 widening the SDR capture path at dpr=1 — same composition shape, slower per-frame, total render now crosses 600s.
## The fix
Convert the timer to a heartbeat: each `writeFrame` that goes through to the kernel pipe (i.e. `stdin.write` returns `true`) resets it. Only true hangs (no successful frame write for the timeout window) trip SIGTERM now; "slow but progressing" renders are unbounded.
Crucially, the heartbeat does **not** reset on `accepted === false`. A `false` return means Node had to buffer the write because FFmpeg hasn't drained the pipe yet — that's not proof of consumer progress, just proof we produced. Without this distinction, a hung FFmpeg with a live Chrome would queue frames into Node's writable buffer indefinitely (no backpressure path back to the capture loop) and grow until OOM. In steady state with a slow-but-alive FFmpeg, writes alternate between `true` and `false` as the buffer drains and refills; the `true`s are enough to keep the heartbeat ticking.
Renames are intentionally avoided — `ffmpegStreamingTimeout` keeps its name and `600_000` default; only the semantics changed. The config doc spells out the new behavior so downstream consumers know what 600s now means.
## Test plan
- [x] **Slow-but-progressing capture** (`accepted=true`): 9× `writeFrame` at 900ms intervals (under the 1000ms threshold) — encoder stays alive through 8.1s. Stall past the threshold — SIGTERM fires.
- [x] **Stalled FFmpeg with live producer** (`accepted=false`): override `stdin.write` to return false; pump 9× `writeFrame` at 900ms intervals. SIGTERM still fires inside the 1000ms window — buffered writes don't keep the heartbeat alive.
- [x] Existing 33 tests in `streamingEncoder.test.ts` still pass
- [x] Lint (`oxlint`) + format (`oxfmt --check`) clean
- [ ] CI regression suite
🤖 Generated with [Claude Code](https://claude.com/claude-code)
## Summary
- `seek()` only called `seekAll()` under parent audio ownership, leaving the `<audio>` proxy playing while the timeline froze at the new seek target.
- The periodic `mirrorTime` drift-correction (`parent-media.ts`) would then yank `currentTime` back to the timeline position every ~80ms of accumulated drift, producing an audible stutter loop while the video frame stayed frozen.
- Fix: make `seek()` symmetric with `pause()` — pause the parent proxy before seeking it.
## Repro
1. Use the player in an environment where the runtime posts `media-autoplay-blocked` (mobile / autoplay-restricted contexts), promoting audio ownership to `"parent"`.
2. Start playback with audio.
3. Click anywhere on the scrubber while playing.
4. Before: audio stutters in a short loop while the video frame is frozen.
5. After: audio cleanly pauses at the new seek position.
## Test plan
- [x] Added regression test \`seek() while playing pauses parent proxy (prevents mirrorTime stutter loop)\` in \`hyperframes-player.test.ts\`.
- [x] \`pnpm --filter @hyperframes/player test\` — 110/110 pass.
- [ ] Manual repro on a device where ownership flips to \`parent\`.
🤖 Generated with [Claude Code](https://claude.com/claude-code)
## Summary
Two small fixes that together make `@hyperframes/core` + `@hyperframes/studio` consumable from non-Vite hosts (Next.js / Turbopack, Node, etc.).
### 1. `core`: ship the missing `lottieReadiness` module
The `"./runtime/lottie-readiness"` subpath export in `@hyperframes/core` claims to ship at `./dist/runtime/adapters/lottieReadiness.js`, but that file is missing from the published 0.6.6 and 0.6.7 tarballs. Consumers that import the subpath — most notably `@hyperframes/studio`'s `Player.tsx` — fail to resolve the module and break downstream builds.
**Root cause:** `packages/core/tsconfig.json` excludes `src/runtime` (those files run in a browser context and are bundled separately into the IIFE artifact). Since nothing in the included tree imports `lottieReadiness.ts`, tsc never emits a compiled output, and the file silently goes missing from the publish.
**Fix:** `lottieReadiness.ts` is a pure helper — takes `unknown`, returns `boolean`, no DOM/`window` dependencies. It doesn't belong in `src/runtime/` in the first place; the runtime-exclude rule rightly caught it. Move it to `src/lottieReadiness.ts` so the standard library build picks it up.
The subpath export **name** stays `"./runtime/lottie-readiness"` — only the exports map's underlying file path changes — so existing consumers (studio) don't need any code change.
### 2. `studio`: guard `import.meta.env` for non-Vite hosts
`packages/studio/src/components/editor/manualEditingAvailability.ts` unconditionally reads `import.meta.env`. That's a Vite-only extension; in plain ESM hosts (Next.js / Turbopack, Node, jest in some configs) `import.meta` exists but `import.meta.env` is `undefined`. Reading any property off undefined throws at module evaluation time, so the studio fails to load the moment a non-Vite host imports anything from `@hyperframes/studio`.
Guarded the read so the module is loadable everywhere; outside Vite, every flag falls back to its declared default, preserving Vite behavior.
### Changes
**core:**
- `mv src/runtime/adapters/lottieReadiness.{ts,test.ts}` → `src/`
- Update `src/runtime/adapters/lottie.ts` re-export path
- Update `package.json` + `publishConfig.exports` to point at the new dist path (`./dist/lottieReadiness.{js,d.ts}`)
**studio:**
- One-line guard in `manualEditingAvailability.ts:30` with explanatory comment
## Test plan
- [x] `pnpm typecheck` (core, studio) — clean
- [x] `bun run build` (core) — `dist/lottieReadiness.{js,d.ts}` now present
- [x] `bunx vitest run` (core) — 862/862 passing
- [x] `bun run typecheck` (studio) — clean, resolves moved file via subpath export
- [ ] Publish 0.6.8 and verify the tarball contains `dist/lottieReadiness.js`
- [ ] Verify a non-Vite ESM consumer (e.g. a Next.js / Turbopack app) imports `@hyperframes/studio` without `import.meta.env` errors
🤖 Generated with [Claude Code](https://claude.com/claude-code)
## Summary
- **Direct-timeline path** (GSAP compositions with `window.__timelines`): The player drives these via `DirectTimelineAdapter`, bypassing postMessage entirely. Rate changes sent `set-playback-rate` to the iframe but had no receiver — GSAP's `timeScale()` was never called. Fix: add optional `timeScale?` to `DirectTimelineAdapter` and call `this._directTimelineAdapter?.timeScale?.(rate)` in `attributeChangedCallback`. GSAP timelines expose `timeScale` natively, no composition changes required.
- **Audio-clock path** (compositions with audio): Three bugs caused `TransportClock` to always run at 1x when an audio element or WebAudio context drove the clock:
1. `schedulePlayback` was called without the `playbackRate` arg (defaulted to 1).
2. `onSetPlaybackRate` and `player.setPlaybackRate` didn't call `webAudio.setRate()`.
3. `TransportClock.attachAudioSource` divided by `this._rate` instead of `el.playbackRate`, cancelling the rate multiplier.
- Adds 2 regression tests to `clock.test.ts` covering the corrected audio-clock formula.
## Test plan
- [ ] Unit tests: `bun run --cwd packages/core test` — 861/861 pass
- [ ] Browser verification (Playwright headless, GSAP direct-timeline composition):
- 1x speed → ratio 0.972 ✓
- 2x speed → ratio 1.965 ✓
- 0.5x speed → ratio 0.490 ✓
🤖 Generated with [Claude Code](https://claude.com/claude-code)
Chromium throttles requestAnimationFrame in deeply nested cross-origin
iframes. In Claude desktop (Electron), the composition iframe's own rAF
loop stalls, so GSAP is never seeked and animation freezes even when
TransportClock.isPlaying() is true.
The correct fix is to drive ticks from the widget-frame rAF, which lives
one level up and is not subject to the same throttling. When play() takes
the runtime bridge path (no direct timeline adapter), the player now starts
a parent-frame rAF loop that sends "tick" postMessages to the composition
iframe on every frame. The runtime's control bridge handles "tick" by calling
seekTimelineAndAdapters(clock.now()) if the clock is playing — identical to
what transportTick does on each rAF, just driven from outside.
The composition iframe's own rAF loop is unchanged and keeps running
normally in standard browsers. Seeking GSAP twice per frame is idempotent,
so there is no regression on claude.ai or any other non-throttled environment.
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
The postMessage state path set `_currentTime` without clamping, while the
direct timeline path already used `Math.min(currentTime, _duration)`. A
final-frame state message with a frame count slightly past the end would
set `_currentTime > _duration`, causing the progress bar (position:
absolute, no overflow guard) to bleed out of the scrubber track and
visually cover the volume button, and the time display to show values
like "0:05 / 0:04".
- Clamp `_currentTime` in `_onMessage` to match the direct timeline path
- Clamp defensively in `updateTime` so the display layer never overflows
- Add `overflow: hidden` + `min-width: 0` to `.hfp-scrubber` as a CSS
safety net; remove now-redundant `border-radius` from `.hfp-progress`
(parent `overflow: hidden` handles clipping to the rounded shape)
- Apply the same `overflow: hidden` fix to `.hfp-volume-slider` for
consistency; remove redundant `border-radius` from `.hfp-volume-fill`
- Add regression test covering the postMessage over-duration case
Co-authored-by: Claude Sonnet 4.6 <noreply@anthropic.com>
terencecho