- move resolveR2PlaybackUrl into lib/video-upload-validation.ts so the compare
view and the main video page cannot drift apart
- validate the resolved URL with isPlayableVideoUrl before it reaches <video src>
- add a per-player cooldown so a follower that cannot keep up is not seeked
every second, which would stutter rather than correct
The compare page predates the r2 upload provider: r2 versions fell
through to a URL-safety check that throws on app-relative upload URLs,
so their panels rendered nothing and registered no player — the shared
controls drove an empty list and nothing played.
- add an R2Panel mapping a plain video element over the app upload
route to the shared adapter, using the same playback-url resolution
as the main video page
- make play/pause state detection work without the YouTube API loaded
(numeric fallback), so bunny/r2-only comparisons can pause
- re-sync panels that drift more than 350ms from the source player
once per second so playback stays aligned, not just starts aligned
Verified against the running app: both versions play in lockstep
(0.000s measured drift), pause together, and seek together.
Co-Authored-By: Claude Fable 5 <[email protected]>
Comment content was stored with newlines intact but rendered inside <p>
elements with default white-space, collapsing line breaks into single
spaces. Add whitespace-pre-wrap (and break-words) to the comment/reply
render wrappers in the comments pane and the compare-versions view.
- Added support for self-hosted S3 video uploads with new environment variables: OPENFRAME_ENABLE_S3_VIDEO_UPLOADS and OPENFRAME_MAX_VIDEO_UPLOAD_BYTES.
- Updated .env.example and .env.docker.example to reflect new configuration options.
- Enhanced Content Security Policy to include origins for S3-compatible storage.
- Updated dependencies for AWS SDK to support new features.
- Refactored upload logic to accommodate both Bunny and S3 upload providers.
- Updated documentation to clarify the usage of direct uploads and S3 configurations.
- Closes#11
- Moved registration logic to a new client component for better separation of concerns.
- Integrated invite code requirement based on feature flags in the registration process.
- Enhanced dashboard functionality to conditionally enable Bunny uploads based on feature flags.
- Updated various components and API routes to check for Bunny uploads and Stripe billing feature flags.
- Added new feature flag utilities for managing feature toggles in the application.
Replace per-frame React state updates with ref-based tracking and direct DOM
mutation for the progress bar, playhead, and timecode. React state is now
throttled to ~4 commits/sec (~250ms), eliminating unnecessary re-renders on
every animation frame during playback.
Co-Authored-By: Claude Sonnet 4.6 <[email protected]>
- Implement optimistic updates for comment creation, deletion, resolution, tag, and content changes
- Enhance user experience by providing immediate feedback for comment actions
- Integrate `sonner` toasts for success and error notifications
- Improve video data fetching error handling and logging
- Refine active video version selection logic for robustness
- Simplify video detail not-found page, removing dynamic project link
- Introduce dedicated error pages for dashboard and video routes
- Add specific not-found pages for dashboard, projects, videos, and settings
- Implement global `not-found.tsx` for general unhandled routes
- Integrate root and dashboard layouts with ErrorBoundary and Suspense
- Add new UI components: Accordion, Hover Card, Menubar, Navigation Menu, Select, Tabs
- Update Navbar to utilize the new Navigation Menu component
- Enhance `button` component with a `link` variant for better styling
- Refine existing UI components (dialog, dropdown, input, etc.)
- Update Tailwind config with new colors and animation extensions
- Implement loading states for dashboard, project, and video views
- Introduce skeleton UIs for settings pages
- Add skeletons for Navbar and charts
- Enhance `Skeleton` component with a text variant
- Improve perceived performance during data fetching
- Update `.gitignore` with `Optimization.md`
- Removed Telegram setup instructions from settings page.
- Enhanced video and version comment APIs to include tag information.
- Added new CommentTag model in Prisma schema for managing tags.
- Created API routes for managing tags (GET, POST, PATCH, DELETE).
- Updated WatchPage to support tag selection and display.
- Introduced keyboard shortcuts modal for improved user experience.
- Added tag selection dropdown in comment input area.
- Add R2 client setup and audio upload functionality in lib/r2.ts.
- Create audio file cleanup functions in lib/r2-cleanup.ts to delete voice files associated with videos, projects, and workspaces.
- Enhance rate limiting in lib/rate-limit.ts with new action-specific limits and improved IP validation.
- Introduce a unified rate limit check function that returns a 429 response when limits are exceeded.
- Update package.json to include the AWS SDK for S3.
- Implemented edit dialog for updating video title and description.
- Added functionality to create new video versions with URL validation.
- Included delete confirmation dialog for video removal.
- Enhanced UI with loading indicators and error handling.
- Updated video card layout for better user interaction.
feat(youtube): extend YT namespace with playback rate methods
- Added methods to set and get playback rate.
- Included method to retrieve available playback rates.
Initialize OpenFrame project with core scaffold and UI foundation.
- Project scaffold: Next.js 16.1 (App Router) + Bun runtime
- UI: shadcn/ui + TailwindCSS; landing, dashboard, project, and auth UIs
- Video support: provider abstraction (YouTube first), video player page with timestamped comments and custom timeline
- Auth & DB: NextAuth skeleton and Prisma schema (Postgres) included
- UX: dark-mode toggle, full-width layouts, comments sidebar, video route moved to /watch/[videoId]
- Dev: added YouTube iframe integration, custom controls, and TypeScript types
- Next steps: DB migrations, API routes (CRUD), real data wiring, voice-recording & sharing