- Introduced IPC channels for getting and setting the active search space, enhancing user experience across the application.
- Updated the preload script to expose new API methods for active search space management.
- Modified the main window and quick ask functionalities to sync the active search space based on user navigation.
- Enhanced the desktop and web applications to allow users to select and manage their default search space seamlessly.
- Implemented automatic synchronization of the active search space during login and navigation events.
- Replaced icons for "General Assist," "Quick Assist," and "Extreme Assist" shortcuts to better represent their functionalities.
- Updated descriptions for each shortcut to enhance user understanding of their actions.
- Refactored the layout of the shortcut recorder for a more streamlined user experience.
- Introduced a new "General Assist" shortcut, allowing users to open SurfSense from anywhere.
- Updated shortcut management to include the new general assist functionality in both the desktop and web applications.
- Enhanced the UI to reflect changes in shortcut labels and descriptions for better clarity.
- Improved the Electron API to support the new shortcut configuration.
- Updated the development script to include a build step before launching the app.
- Refactored the registration of quick ask and autocomplete functionalities to be asynchronous, ensuring proper initialization.
- Introduced IPC channels for getting and setting keyboard shortcuts, allowing users to customize their experience.
- Enhanced the platform module to support better interaction with the Electron API for clipboard operations.
- Improved the user interface for managing keyboard shortcuts in the settings dialog, providing a more intuitive experience.
- Added IPC channels for getting and setting auth tokens in Electron.
- Implemented functions to sync tokens from localStorage to Electron and vice versa.
- Updated components to ensure tokens are retrieved from Electron when not available locally.
- Enhanced user authentication flow by integrating token management across windows.
Wrap scroll handlers in thread.tsx, InboxSidebar.tsx, and
DocumentsTableShell.tsx with requestAnimationFrame batching so scroll
position state updates fire at most once per animation frame instead of
on every scroll event (up to 60/sec at 60fps). Add cleanup useEffect to
cancel pending frames on unmount.
Fixes#1103
- Updated maximum file size limit to 500 MB per file.
- Removed restrictions on the number of files per upload and total upload size.
- Enhanced handling of user-mentioning documents in the knowledge base search middleware.
- Improved document reading and processing logic to accommodate new features and optimizations.
Components were calling useSearchParams() at the top level but only
reading the value inside useEffect or callbacks, never in JSX. This
subscribed the entire component tree to every URL query change.
Fix: read from window.location.search directly inside the effect so
no React subscription is created.
Changes:
- new-chat/page.tsx: read commentId inside effect + popstate listener
for SPA back/forward support; removes subscription from 1500+ line tree
- dashboard/page.tsx: read window.location.search at redirect time;
removes searchParams from dep array
- public-chat-footer.tsx: one-shot mount read for action=clone param
- TokenHandler.tsx: one-shot mount read for token + refresh_token params
Implements Vercel React Best Practices Rule: rerender-defer-reads (5.2)
Apply content-visibility: auto to off-screen list and sidebar items
to skip browser layout and painting until they scroll into view.
Changes:
- globals.css: add .list-item-lazy (48px) and .sidebar-item-lazy (40px) utilities
- DocumentsTableShell: apply list-item-lazy to desktop <tr> rows and mobile card divs
- AllPrivateChatsSidebar: apply sidebar-item-lazy to thread items
- AllSharedChatsSidebar: apply sidebar-item-lazy to thread items
- InboxSidebar: apply content-visibility inline (80px intrinsic height) to inbox items
Implements Vercel React Best Practices Rule: rendering-content-visibility (6.2)
- Pass is_public from request body in create_prompt route
- Only bump version on content field changes (name, prompt, mode),
not on is_public toggle
- Add prompts query and mutation atoms (atomWithQuery/atomWithMutation)
with TanStack Query caching, replacing manual useEffect fetches
- Update PromptPicker, PromptsContent, and CommunityPromptsContent
to consume shared atoms instead of local state
Remove hardcoded DEFAULT_ACTIONS and icon map from prompt-picker,
fetch all prompts from the backend. Simplify Zod types to match the
single-table schema (drop source/system_prompt_slug/is_modified,
add version). Update PromptsContent empty state copy.
