tanstack-db

Live queries automatically recompute SQL-like query results when underlying normalized collection data changes. You create collections tied to your API, then use live query operators to subscribe to filtered, sorted, or joined subsets for reactive UI updates.

Optimistic mutations instantly update the UI while synchronizing with the server, and automatically roll back if the server write fails. This ensures user interactions feel fast and consistent without manual state recovery logic.

A reactive client-side data layer stays in sync with server-driven data to avoid manual cache invalidation. Normalized collections with sync handlers manage inserts, updates, and deletes, keeping local state consistent automatically.

Yes, normalized collections support on-demand loading for large datasets. Sync handlers wire collections to your API, enabling local-first interactivity over server data for filtered lists and sorted subsets.

Yes, building live queries requires @tanstack/react-db and @tanstack/query-db-collection. These dependencies define collection sync modes and provide operators that map reactive queries to network requests.

A normalized client-side database prevents brittle state management by providing a reactive data layer. It handles incremental live query recomputation and cross-collection joins, eliminating the need to manually track and update dependent UI states.