Superset 1.0 is a huge milestone for Apache Superset. This release holds a higher quality standard than any previous version and raises the bar for releases to come. Superset 1.0 improves on usability and delivers a new set of long awaited features focused on five major themes:
We have taken the Superset user experience to the next level with a much simpler, more intuitive UI.
Lists of Charts and Dashboards can be shown in a new format using a thumbnail grid. This makes it easier to discover and find Charts and Dashboards, especially when there are a lot of them in your Superset instance.
Visualization controls have been updated to create a more consistent and modernized interface. Refactoring includes updates to the content/labeling/ordering of these controls as well as updating their styling and simplifying their interactions and layout. This will line us up for features like drag-and-drop controls, and dynamically populated control inputs in future releases.
Users also now have the ability to create and name a new dataset or update an existing dataset when moving from SQL Lab to Explore. When going back to SQL Lab, the underlying query for the dataset will be shown, allowing users to easily make changes and update their dataset when moving between SQL Lab and Explore.
The Superset 1.0 documentation has been updated to reflect the current design and functionality.
See related PR highlights here.
Superset 1.0 makes it easier for developers to build, deploy, and maintain Superset functionality. This release is a major milestone in an ongoing effort to modernize, consolidate, and simplify the interface elements of Superset. Highly-used components have been visually updated, refactored to modern component libraries, and integrated with React Storybook for immediate visibility with consolidated test/styles. This improves the developer experience and creates a more consistent and modern aesthetic for the user.
The REST API continues to see improvement as well, with new endpoints and improvements to existing endpoints. More details here.
The Swagger interface can be used to explore and try out these changes.
See related PR highlights here.
This the most performant Superset release to date, with enhanced scalability and efficiency. In addition to many small tweaks, Charts and SQL Lab both now support asynchronous data loading. Users will feel this improvement when loading query results, especially when there are many charts in a Dashboard or when queries are long running.
See related PR highlights here.
Superset 1.0 includes a slew of new features and other enhancements. New visualization plugin architecture will make it easier to build, test, style and configure custom visualizations for Superset. Dynamic viz plugin imports will allow Superset to load data viz plugins, allowing developers to more easily use or share their custom plugins and load custom plugins on-demand from anywhere on the web. Superset has also adopted Apache ECharts as the core library for new visualizations.
Along with the Echarts integration, we are also introducing a better set of annotation features. Formula annotation, which allows users to plot any mathematical function on the chart; Interval and Event annotations, which allow users to add context to the trend in Time-Series; and Line annotation, which use a predefined chart as the source of annotation.
A redesigned home screen provides a personalized landing page that the user sees when they log into Superset, improving discoverability. It shows items relevant to the user (Charts, Dashboards, Saved Queries, etc) and serves as a hub to discover content and access recent items quickly.
Charts on dashboards have been updated to concisely show which filters are in scope, have been applied, or are in error. Context around filter changes has been increased by highlighting scoped charts when a filter is selected and increasing visibility of charts with incompatible filters applied.
Alerts and reporting have received a robust backend and UI overhaul.
See related PR highlights here.
Some of the new features in this release are disabled by default. Each has a feature flag in config.py and some require configuration of additional backend dependencies (e.g. celery, SMTP server, etc). The following table will help you get started on the list of features and their associated flags & dependencies. Feel free to contribute and expand this list.
| Feature | Feature Flag | Dependencies | Documentation |
|---|---|---|---|
| Global Async Queries | GLOBAL_ASYNC_QUERIES: True | Redis 5.0+, celery workers configured and running | Extra documentation |
| Dashboard Native Filters | DASHBOARD_NATIVE_FILTERS: True | ||
| Alerts & Reporting | ALERT_REPORTS: True | Celery workers configured & celery beat process | |
| Homescreen Thumbnails | THUMBNAILS: TRUE, THUMBNAIL_CACHE_CONFIG: CacheConfig = { "CACHE_TYPE": "null", "CACHE_NO_NULL_WARNING": True} | selenium, pillow 7, celery | |
| Dynamic Viz Plugin Import | DYNAMIC_PLUGINS: True |
This release includes hundreds of bugfixes and stability enhancements. Future major releases will have a continued emphasis on providing a stable and bug-free experience for the user.
Below is a highlight of the PRs included in this update. The full list is much longer, and can be found here.
Backwards incompatible changes and can be found here.
To see the complete changelog, see apache/superset/CHANGELOG.md