Release 464 (30 Oct 2024)

General

• ⚠️ Breaking change: Require JDK 23 to run Trino. (#21316)

• Add the Faker connector for easy generation of data. (#23691)

• Add the Vertica connector. (#23948)

• Rename the fault-tolerant-execution-eager-speculative-tasks-node_memory-overcommit configuration property to fault-tolerant-execution-eager-speculative-tasks-node-memory-overcommit. (#23876)

Accumulo connector

• ⚠️ Breaking change: Remove the Accumulo connector. (#23792)

BigQuery connector

• Fix incorrect results when reading array columns and bigquery.arrow-serialization.enabled is set to true. (#23982)

Delta Lake connector

• Fix failure of S3 file listing of buckets that enforce requester pays. (#23906)

Hive connector

• Use the hive.metastore.partition-batch-size.max catalog configuration property value in the sync_partition_metadata procedure. Change the default batch size from 1000 to 100. (#23895)

• Fix failure of S3 file listing of buckets that enforce requester pays. (#23906)

Hudi connector

• Fix failure of S3 file listing of buckets that enforce requester pays. (#23906)

Iceberg connector

• Improve performance of OPTIMIZE on large partitioned tables. (#10785)

• Rename the iceberg.expire_snapshots.min-retention configuration property to iceberg.expire-snapshots.min-retention. (#23876)

• Rename the iceberg.remove_orphan_files.min-retention configuration property to iceberg.remove-orphan-files.min-retention. (#23876)

• Fix failure of S3 file listing of buckets that enforce requester pays. (#23906)

• Fix incorrect column constraints when using the migrate procedure on tables that contain NULL values. (#23928)