v1.8.2
This release includes bug fixes for filter handling in CubeScan, join-hint enrichment loops, and masked member visibility in pre-aggregations. It adds support for anchoring rolling windows on multi-stage include date ranges and improves performance in UNION ALL planning, native serde deserialization, and plan SQL template memoization.
1.8.2 (2026-10-08)
Bug Fixes
- cubesql: Keep filters above limited CubeScan (#12071) (4b3b59d)
- schema-compiler: Stop join-hint enrichment looping on a cyclic join tree (#12149) (ba1736f), closes #9696
- tesseract: Mask masked members read from pre-aggregations (#12152) (bbbaea4)
Features
Performance Improvements
Source: original entry ↗
More from Cube
Follow Cube to get its new changes in your feed and email digest.
Stop rollupJoin and rollupLambda from leaking memory through memoized references
Fixed a memory leak in the schema-compiler where rollupJoin and rollupLambda functions were holding memoized references. This resolves performance degradation in long-running processes.
v1.8.5
Release 1.8.5 includes bug fixes for cubesql subqueries and databricks-jdbc-driver time handling, a new environment variable for controlling scheduled refresh timers, and performance improvements for pre-aggregation matching.
v1.8.4
This release includes bug fixes for API gateway SQL injection vulnerabilities, client state management issues, and decimal calculation accuracy in CubeStore, plus new CLI support for managing custom roles and improved leaf merge optimization in Tesseract.