All notable changes to this project will be documented in this file.
The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.
- Performance
- data compression using pako
- implement progressive loading with LTTB (Largest-Triangle-Three-Buckets) sampling
- intersection observer for lazy loading visualizations
- optimized chunk sizes and sampling
- loading progress indicator for large datasets
- zlib compression on server side for data transfer optimization
- lazy loading using intersection observer for on-demand graph renders
- data sampling to reduce point density for large datasets
- chunk loading to prevent browser freezing
- memoization and React.memo to prevent unnecessary re-renders
- debounced resize handling
- feature toggles for fine-tuning optimization parameters
- Deployment
preswald deploy
now works for local and cloud deployment via cloud-runpreswald stop
works for stopping local deployments
- Misc
- Moved setuptools from dev to core dependency in response to our first github issue
- Set log levels via preswald.toml + CLI
- MVP of State Management
- Atom and Workflow classes provide notebook-like DAGs
- Support for selective recomputation, caching
- Basic dependency visualization and analysis
- Initial release of Preswald:
- Core functionality for building simple, interactive data apps.
- Support for Markdown, data connections, and rendering tables.
- Basic theming and layout configurations.
- Full CLI support for project management.