Cube — Data, Analytics product updates and releases, tracked on megachangelog.
Optimized CubeSQL to merge view joins that share a cube member into a single CubeScan operation, improving query performance and efficiency.
This release includes bug fixes for access policies with row filters, JavaScript number deserialization, and Presto driver header handling, plus an EXPLAIN ANALYZE DETAILED feature for per-query trace in Cube Store.
This release includes bug fixes for the API Gateway WebSocket cache parameter handling, improved subscription ID handling and connection management, and corrected deserialization of large JavaScript numbers that exceed u64/i64 range.
This release introduces HttpQueryResultCompleted for zero-column results and unique key filter pushdown below last-row dedup in cubestore, along with a performance improvement that combines session initialization into a single ALTER SESSION for the Snowflake driver.
This release includes bug fixes for SQL push down filter parameters, nested CTE generation, pre-aggregation table naming issues, and SQL planner tests. It also adds support for accepting a list of API secrets for verification in the API gateway.
This release includes bug fixes for the QuestDB driver's HAVING clause support, improved error parsing UX in CubeSQL, new links feature in the data model, and native Rust time-series generation capabilities in Tesseract.
This release includes bug fixes for Athena custom-granularity pre-aggregation builds and Tesseract JOIN-model assembly, along with new features for universal error prefixing in CubeSQL, distributed query cancellation for Athena, and multi-stage filter and grain directives in Tesseract.
This release includes new features for cubesql including pre-aggregation indicators and filter rule flattening, plus performance improvements to QueryResult with columnar storage migration.
Bug fixes for schema-compiler duplicate cube and view names in JS models, plus enablement of native orchestrator for all cases.
This release includes bug fixes for cubestore-cli reconnections and tesseract pre-aggregation usage, along with new features for grouped joins with SQL push down, timezone format parsing, Arrow format support, and default value filters for views.