Apache Superset 1.2: Release Notes
We're excited to announce the release of Apache Superset 1.2!
In this release post, I will focus on the biggest and most interesting tangible, end-user features. If you want to learn more about progress made on experimental features, I recommend reading the official community release notes for 1.2.
New Charts!
As a data exploration & visualization, new chart types are always a welcome addition.
Mixed Time-series Chart
Mixed timeseries charts let you mix multiple timeseries objects in a single visualization. Currently, this chart type requires two queries that each map to their own timeseries object.
Here's an example that mixes a line chart and a bar chart:
You can read more about how and when to use this chart recently released post on new chart types.
Radar Chart
A radar chart lets you visualize a parallel set of metrics across multiple groups or categories. Here's a quick example of a simple radar chart that visualizes parallel funnel metrics across 3 holiday seasons.
Here's a more complex example that visualizes max monthly temperatures for six hypothetical cities:
You can read more about how and when to use a radar chart in our recently released post on new chart types.
Redesigned Pivot Table
Improvements to the pivot table chart are a frequent request from the community and the committers have answered! n addition to the classic pivoting of rows and columns, you can now enable heatmap and bar chart visualizations within the pivot table itself.
Here's an example of a pivot table of popular male baby names in the United States that uses column oriented heatmap coloring:
You can read more about how and when to use a radar chart in our recently released post on new chart types.
New Databases!
Besides chart types, being able to query more data sources are new superpowers for a data exploration tool. This release also adds support for two new databases.
CrateDB
CrateDB is a distributed, time-series focused data store that has an existing Python DB API 2.0 driver with a SQLAlchemy dialect.
If you've read our previous post on adding new database connectors in Superset, you know that we're already 90% of the way there to querying CrateDB from Superset.
To showcase how easy it is to support new databases in Superset, we actually ran a community event and accomplished this live! You can watch that recording here.
Not sure what the difference between a time-series data store and a real-time data store is? We wrote about that recently too. You can read the code needed to support Crate here.
Databricks Cloud
Superset and Preset Cloud now both support Databricks Cloud, using the pyhive library. If you're curious, here's the relevant Github PR.
To learn how to connect to Databricks in Preset Cloud, check out this documentation article.
Breaking Changes
Thanks in part to the semantic versioning schema adopted by the community, Superset 1.2 doesn't contain any backwards incompatible changes! 🎉
Full Changelog
You can read the full list of PR's that made it into this release in CHANGELOG.MD.