Release 414 (19 Apr 2023)

General

• Add recursive member access to the JSON path language. (#16854)
• Add the sequence() table function. (#16716)
• Add support for progress estimates when fault-tolerant execution is enabled. (#13072)
• Add support for CUBE and ROLLUP with composite sets. (#16981)
• Add experimental support for tracing using OpenTelemetry. This can be enabled by setting the tracing.enabled configuration property to true and optionally configuring the OLTP/gRPC endpoint by setting the tracing.exporter.endpoint configuration property. (#16950)
• Improve performance for certain queries that produce no values. (#15555, #16515)
• Fix query failure for recursive queries involving lambda expressions. (#16989)
• Fix incorrect results when using the sequence() function with values greater than 2³¹ (about 2.1 billion). (#16742)

Security

• Disallow graceful shutdown with the default system access control. Shutdowns can be re-enabled by using the allow-all system access control, or by configuring system information rules with the file system access control. (#17105)

Delta Lake connector

• Add support for INSERT, UPDATE, and DELETE operations on tables with a name column mapping. (#12638)
• Add support for Databricks 12.2 LTS. (#16905)
• Disallow reading tables with deletion vectors. Previously, this returned incorrect results. (#16884)

Iceberg connector

• Add support for Hive external tables in the migrate table procedure. (#16704)

Kafka connector

• Fix query failure when a Kafka topic contains tombstones (messages with a NULL value). (#16962)

Kudu connector

• Fix query failure when merging two tables that were created by CREATE TABLE ... AS SELECT .... (#16848)

Pinot connector

• Fix incorrect results due to incorrect pushdown of aggregations. (#12655)

PostgreSQL connector

• Fix failure when fetching table statistics for PostgreSQL 14.0 and later. (#17061)

Redshift connector

• Add support for fault-tolerant execution. (#16860)