⚡ Rill Developer is a tool that makes it effortless to transform your datasets with SQL and create fast, exploratory dashboards.
To try out Rill Developer, check out these instructions and let us know over on Discord if you encounter any problems or have ideas about how to improve Rill Developer!
This release has lots of refinements to the Open Beta experience that we launched at the end of May as well as an upgrade to duckDB
Upgrade to duckDB 0.8 - Rill is now running on duckDB
0.8 with improvements in data ingest performance as well as
FLOAT division by default with no need to
CAST get decimal-level precision. This is a breaking change and will require that you delete your database
rm -rf stage.db* and run
rill start to recreate the database with the latest version. Unfortunately we do not support their new
PIVOT feature for dynamic lists of values during modeling, only static values, because it is not designed to work with the ephemeral data views we use for Rill dashboards.
- Will work:
PIVOT iris on Species in ('Iris-setosa', 'Iris-virginica') USING first(SepalLengthCm);
- Will not work:
PIVOT iris on Species USING first(SepalLengthCm);
Refine default time grain choices for each time range - All time ranges have a natural time grain that makes the most sense to show in the dashboard. We have some logic in the frontend today that gives us this functionality, but it’s far from perfect. These updates should set better defaults that will feel more "natural" given your data.
Support domain whitelisting for role-based access - On launch, Rill asked admins to grant view access to a hosted dashboard at the individual email level. In other words, you had to be explicit and add users one at a time. This can be quite arduous and is more fine-grained than many organizations need. These changes allow a domain to be whitelisted more broadly, but only for the same domain as the admin's email.
Many small improvements to our cloud platform - We continue to refine our cloud experience through many small changes. Of note - now schemas are relaxed in glob sources to prevent error cases when inferred types change, we better optimize requests and query timeouts to prevent runaway queries hogging resources, and deployment flows support "publishing" a project in an error state before asking for credentials to help you remember your project state.