Back to Trino

Release 471 (19 Feb 2025)

docs/src/main/sphinx/release/release-471.md

4802.7 KB
Original Source

Release 471 (19 Feb 2025)

General

  • Add for textual tasks on data using OpenAI, Anthropic, or other LLMs using Ollama as backend. ({issue}24963)
  • Include split count and total split distribution time in the EXPLAIN ANALYZE output. ({issue}25028)
  • Add support for JSON logging format to console with log.console-format=JSON. ({issue}25081)
  • Support additional Python libraries for use with Python user-defined functions. ({issue}25058)
  • Improve performance for Python user-defined functions. ({issue}25058)
  • Improve performance for queries involving ORDER BY ... LIMIT. ({issue}24937)
  • Prevent failures when fault-tolerant execution is configured with an exchange manager that uses Azure storage with workload identity. ({issue}25063)

Server RPM

  • Remove RPM package. Use the tarball or container image instead, or build an RPM with the setup in the trino-packages repository. ({issue}24997)

Security

  • Ensure that custom XML configuration files specified in the access-control.properties file are used during Ranger access control plugin initialization. ({issue}24887)

Delta Lake connector

  • Add support for reading variant type. ({issue}22309)
  • Add . ({issue}25006)
  • Support reading cloned tables. ({issue}24946)
  • Add support for configuring s3.storage-class when writing objects to S3. ({issue}24698)
  • Fix failures when writing large checkpoint files. ({issue}25011)

Hive connector

  • Add . ({issue}25006)
  • Add support for configuring s3.storage-class when writing objects to S3. ({issue}24698)
  • Fix reading restored S3 glacier objects when the configuration property hive.s3.storage-class-filter is set to READ_NON_GLACIER_AND_RESTORED. ({issue}24947)

Hudi connector

  • Add . ({issue}25006)
  • Add support for configuring s3.storage-class when writing objects to S3. ({issue}24698)

Iceberg connector

  • Add . ({issue}25006)
  • Add support for S3 Tables. ({issue}24815)
  • Add support for configuring s3.storage-class when writing objects to S3. ({issue}24698)
  • Improve conflict detection to avoid failures from concurrent MERGE queries on Iceberg tables. ({issue}24470)
  • Ensure that the task.max-writer-count configuration is respected for write operations on partitioned tables. ({issue}25068)

MongoDB connector

  • Fix failures caused by tables with case-sensitive name conflicts. ({issue}24998)

SPI

  • Remove Connector.getInitialMemoryRequirement(). ({issue}25055)