Add Trino 415 release notes

[colebow]

Description

Assemble the release notes for the upcoming Trino 415 release. Fixes #17135

Release notes

(x) This is not user-visible or docs only and no release notes are required.

Verification for each pull request

Format: maintainer, PR/issue number, ✅ / ❌ rn ✅ / ❌ docs
✅ rn - release note added and verified, or assessed to be not necessary, set to ❌ rn before completion
✅ docs - need for docs assessed and merged, or assessed to be not necessary, set to ❌ docs before completion

19 Apr 2023

• Fix testFilterWithUdf #17106 ✅ rn ✅ docs

20 Apr 2023

• Upgrade Phoenix to 5.1.3 #16778 ✅ rn ✅ docs
• Add warning for query passthrough #15562 ✅ rn ✅ docs
• Add query planning CPU time stat #15318 ✅ rn ✅ docs
• Update Coral #16994 ✅ rn ✅ docs
• Cleanup Hadoop file systems on connector shutdown #17110 ✅ rn ✅ docs
• Simplify TableFunctionSplitProcessor interface #17032 ✅ rn ✅ docs
• Simplify TestingConnectorBehavior declarations #16825 ✅ rn ✅ docs
• Remove flaky race-prone assertions in TestDriver #17144 ✅ rn ✅ docs
• Fix comparison documentation for BETWEEN and NULL #17141 ✅ rn ✅ docs
• Minor improvements to HiveType and HiveUtil #17132 ✅ rn ✅ docs
• Implement adaptive partitioning for FTE based on input #17024 ✅ rn ✅ docs
• Include exception when failed to parse filter in MongoDB #17152 ✅ rn ✅ docs

21 Apr 2023

• Extract a test to verify Databricks and OSS Delta TPCH tables and merge connector tests into one #17137 ✅ rn ✅ docs
• Fix SQL formatting in TestDynamicTable #17109 ✅ rn ✅ docs
• Remove apparent correctness issue with calculatePrefixesWithTableName #17121 ✅ rn ✅ docs
• Use Page#getColumns in TableWriterOperator #17147 ✅ rn ✅ docs
• A couple of small scheduler improvements #17149 ✅ rn ✅ docs
• Type primitives #17136 ✅ rn ✅ docs
• Remove function name from TableFunctionHandle #17117 ❌ rn ✅ docs
• Refactor IrPathNode implementations #17125 ✅ rn ✅ docs
• Document unsigned numeric type MySQL/MariaDB to Trino mapping #17181 ✅ rn ✅ docs
• Add Vale setup for writing style checks #17108 ✅ rn ✅ docs
• Remove trino-main tests against JDK 19 since we already test against JDK 20 #17163 ✅ rn ✅ docs

22 Apr 2023

• Move planning CPU time to resource utilization summary #17155 ✅ rn ✅ docs
• Use standard error code for non-Hudi table in Hudi connector #17168 ✅ rn ✅ docs
• Fix spurious failure for Raptor system.shards table #17184 ✅ rn ✅ docs

23 Apr 2023

• Add oneof support for Protobuf via Confluent Schema Registry #16836 ✅ rn ✅ docs
• Remove AbstractDeltaLakePageSink.addSpecialColumns method #17169 ✅ rn ✅ docs
• Fix broken Iceberg Glue access count test #17185 ✅ rn ✅ docs
• Showcase dereference pushdown limited effectiveness in Iceberg #17145 ✅ rn ✅ docs

24 Apr 2023

• Remove unused codepath in AvroColumnDecoder #17188 ✅ rn ✅ docs
• Allow Delta flush_metadata_cache after table creation #17174 ✅ rn ✅ docs
• Shutdown BigQuery, Kinesis, Redis and Memory connectors #17175 ✅ rn ✅ docs
• Use bulk merging of TDigestHistogram #17186 ✅ rn ✅ docs
• Support Iceberg's DROP TABLE for corrupted tables #16674 ✅ rn ✅ docs
• Suppress existent namespace error in AccumuloClient constructor #17204 ✅ rn ✅ docs
• Document changes in default access control #17153 ✅ rn ✅ docs
• Add documentation for adaptive partitioning and adaptive task sizing of FTE #17179 ✅ rn ✅ docs
• Improve code readability with QueryTracker#hasQuery #17206 ✅ rn ✅ docs
• Report dynamic filters as part of Test*CostBasePlans #17207 ✅ rn ✅ docs
• Use enhanced routine when dropping Delta tables backed by AWS Glue on Databricks #17197 ✅ rn ✅ docs
• Add file pruning test for predicate using primitive fields in ROW column in Delta Lake Connector #17160 ✅ rn ✅ docs
• Use MetadataEntry#getSchemaString in DeltaLakeMetadata#setTableComment #17176 ✅ rn ✅ docs
• Fail the build if a non-existent Maven profile is used #17198 ✅ rn ✅ docs
• Combine smaller related jobs in CI to reduce overhead #17167 ✅ rn ✅ docs

25 Apr 2023

• Test JDBC flush_metadata_cache sequentially #17211 ✅ rn ✅ docs
• Allow disabling caching missing tables in CachingMetastore #17177 ✅ rn ✅ docs
• Enable reading Parquet's bloomfilter statistics for Iceberg connector #17192 ✅ rn ✅ docs
• Make functionNullability property consistent in ResolvedFunction #13777 ✅ rn ✅ docs
• Add merge behavior check in delete with varchar predicate tests #17070 ✅ rn ✅ docs
• Be more conservative when turning off partial aggregation #17143 ✅ rn ✅ docs
• Update Azure & AWS SDKs in exchange filesystem #17100 ✅ rn ✅ docs
• Add property type links to fault-tolerant execution properties #17218 ✅ rn ✅ docs
• Implement predicate push down for parquet dereference column #15163 ✅ rn ✅ docs
• Support commenting on table and column with id and name column mapping mode in Delta Lake #17139 ✅ rn ✅ docs
• Increase default dynamic filter collection limits #17130 ✅ rn ✅ docs

26 Apr 2023

• Support BETWEEN in Delta check constraints #17120 ✅ rn ✅ docs
• Bump calcite-core #16929 ✅ rn ✅ docs
• Refactor internal http client usage #17187 ✅ rn ✅ docs
• Test Iceberg stats on write to unanalyzed table #17227 ✅ rn ✅ docs
• Expose planning CPU time for MYSQL event listener #16841 ✅ rn ✅ docs
• Avoid unnecessary DeterminePartitionCount work #17150 ✅ rn ✅ docs
• OAuth refresh token clarification #16551 ✅ rn ✅ docs
• Documentation copy and style edits Delta Lake #16920 ✅ rn ✅ docs
• Remove deprecated getStartTime method from ConnectorSession #17193 ✅ rn ✅ docs
• Add missing discovery client TLS configuration #17253 ✅ rn ✅ docs
• Upgrade Ignite test container version to 2.9 #17240 ✅ rn ✅ docs
• Allow accessing external environments in Delta Lake query runner #17202 ✅ rn ✅ docs
• Remove doc reference to $data metadata table #17258 ✅ rn ✅ docs
• Avoid starting thread if no file systems were created #17257 ✅ rn ✅ docs
• Update to Airbase 137 and Airlift 230 #17255 ✅ rn ✅ docs
• Small improvements on HiveColumnProjectionInfo #17224 ✅ rn ✅ docs

27 Apr 2023

• Add missing Confluent repo in record decoder module #17243 ✅ rn ✅ docs
• Improve DistributedQueryRunner.Builder #17104 ✅ rn ✅ docs

[martint]

I have no idea what this means :)

[colebow]

It's that performance is more stable / less variable? But not necessarily better? How about just "improve consistency"

[martint]

What do file sizes have anything to do with this, though? cc @sopel39

[mosabua]

I think its so that performance is equally good for smaller and larger files and any mix of them. Specifically this probably also means that performance does not vary as much if some of the data is compacted in object stores and partitions or so are not.