Skip to main content

ยท 7 min read
note

โšก Rill Developer is a tool that makes it effortless to transform your datasets with SQL and create fast, exploratory dashboards. Rill Cloud can then help to enable shared collaboration at scale.

To try out Rill Developer, check out these instructions and join us over on Discord to meet the team behind the product as well as other users. In addition, once you have a project set up in Rill Developer, you can then deploy the project and collaborate with others / invite your team members by logging into Rill Cloud!

release-0 47

Sometimes you want to provide a fast and filtered access to a dashboard, perhaps to a customer or partner, and now you can! Easily create a Public Link that takes any filters that were present at the point of creation and locks them down. This ensures that however you sent the link to can only see that specific slice of your data and can't navigate to other parts of your data that could be sensitive. Optionally add a expiration date for added security!

User Groupsโ€‹

We have introduced the concept of user groups that allows you to assign users to groups. User groups are ideal for when you want to manage access to dashboards or row level data access and don't want role based access. You can now write security rules such as access: {{ has "finance" .user.groups }} to easily manage access to resources in Rill. Use the CLI to add new groups and assign users to them with a Rill Cloud UI based management coming soon!

Host static filesโ€‹

By default we now serve any files placed inside of a public folder in the root of your project and you can configure additional folders via public_paths in rill.yaml. This will allow you host for example images, geojson or shapefiles that can used in new and exciting features coming soon, stay tuned!

Security Rulesโ€‹

We have introduced the ability to utilize wildcards in inclusion and exclusion security rules to make avoid having to list all columns when defining inclusion and exclusion rules together. See more here.

Bug Fixes and Miscโ€‹

  • Added support for scheduling a report directly from a time dimension detail or pivot view.
  • Rill Cloud will now include more detailed error messages when showing an error page.
  • Added support for using subqueries for measure and threshold based filters to ensure measure filters are not limited to returning only 250 values (and thus potentially return unexpected or incorrect results).
  • Introduced styling updates for buttons.
  • Upgraded Rill to utilize GPT4o for dashboard AI generation.
  • Introduced better error handling when connection to the runtime itself has been lost.
  • Improved UX where clicking a table entry now automatically navigates to the table preview in Rill Developer.
  • Added pivot and export support for advanced measures.
  • Improved our OLAP engine connector modal so that the password visibility can be toggled.
  • Added various UX improvements and improved consistency within the connectors page in Rill Developer.
  • Fixed issue where advanced measures could produce an incorrect total.
  • Fixed issue where the pivot table would sometimes incorrectly display measure values as being null.
  • Fixed an issue when using Druid-backed dashboards where pivot tables would enter an error state when trying to fetch null or non-existant dimensional values.
  • Fixed an issue where revisiting a project page in Rill Cloud could trigger a HTTP 401 error due to a stale JWT.
  • Fixed an issue where adding mutually exclusive filters on a pivot table would cause the pivot table to crash.
  • Improved performance of the global dimension search for Druid-backed dashboards.
  • Added hardening to dimension table exports to filter out invalid advanced measures.
  • Added UX improvement so that clicking the dashboard title in the preview opens the dashboard switcher dropdown.
  • Improved the Rill installation binary so that we automatically embed the most commonly used DuckDB extensions.
  • Improved the install options UX when using the Rill install script.
  • Improved the ingestion speed of our BigQuery connector.
  • Improved the performance of our underlying parquet writer.
  • Fixed an issue where users could encounter an "empty" table when scrolling on the pivot table beyond 50 values.
  • Added UX improvement to show the reconciliation status and spinner on charts when the project is reconciling.
  • Introduced UX improvement so users can directly start a new pivot table from a dimension detail view.
  • Fixed an issue where expanding a row in the pivot table does not always reveal sub-rows when there are no columns.
  • Fixed an issue where the organization breadcrumb was missing for anonymous users in Rill Cloud.
  • Fixed an issue where duckdb.Decimal was being serialized as an object instead of a number.
  • Refactored the project status indicators in Rill Cloud, including displaying deployment errors, resource errors, and parse errors in the project deployment status page.
  • Fixed an issue where creating a bookmark with "Save filters only" enabled wasn't working properly.
  • Improved syntax highlighting in Rill Developer so that SQL is automatically highlighted in API and metrics view YAML files.
  • Improved how measure filters are applied and displayed in configured alerts.
  • Added further consistency to how Rill Cloud displays and properly reflects the underlying deployment status of a project.
  • Fixed an UX issue where there was a missing scrollbar in the Add Data modal in Rill Developer.
  • Improved Rill's support for utilizing multiple OLAP engines in the same project, including how multiple connectors are displayed in the Connector Explorer in Rill Developer.
  • Fixed an issue where Rill Developer would not work properly when using http with a non-localhost hostname.
  • Fixed an issue where loading a parquet file with infinity values would lead to 400 Bad Request errors in downstream models and dashboards.
  • Fixed an issue where Rill was not properly applying ignore_paths during rill start, which could lead to initialization issues when trying to start a large but valid Rill project.
  • Added UX improvement so that the keyboard shortcut for the Save functionality (i.e. Cmd-S) is more easily discoverable in Rill Developer via a tooltip.
  • Fixed an issue where adding or re-adding a measure to an existing pivot table with a set of dimensions already selected could lead to the wrong results being displayed.
  • Fixed an issue where users would not be able to shift-click from the time dimension detail or pivot table to copy a dimension value to their clipboard.
  • Fixed an issue where using the Athena connector with a table greater than 10GB in size would result in a glob pattern exceeds limit error.
  • Refactored how export limits are handled under the hood to make exports safer and based on the actual data size, applying a default limit of 128MB on Rill Cloud and unlimited for local exports in Rill Developer.
  • Changed the Last Triggered timestamp for alerts from executionTime to finishedOn.
  • Improved error handling in Rill Cloud to show a more informational message when a security policy results in zero measures being returned for a dashboard.
  • Improved error handling to better and more consistently surface any underlying error statuses coming from the Druid response body when using Druid-backed dashboards.
  • Fixed an issue where using comparison by time with Druid-backed dashboards would result in the dashboard to not be able to render correctly.