Add release notes for 0.243

[caithagoras]

Missing Release Notes

Daniel Ohayon

• Allow qualified names in types #15302 Allow qualified names in types (Merged by: Rongrong Zhong)

George Wang

• Fix #14693: support TRUNCATE function for DOUBLE and REAL #14783 Fix double/real truncation incorrect #14693: support TRUNCATE function for DOUBLE and REAL (Merged by: Rongrong Zhong)

Ke

• Revert "GC issue fixes in SliceDictionarySelectiveReader" #15289 Revert "GC issue fixes in SliceDictionarySelectiveReader" (Merged by: Leiqing Cai)

ankit0811

• Adding approxMostFrequent aggregate from prestoSQL #15216 Adding approxMostFrequent aggregate from prestoSQL (Merged by: Rongrong Zhong)

Extracted Release Notes

• Fix partitioned spatial joins with small spatial index #14485 (Author: James A. Gill): Fix partitioned spatial joins with small spatial index
  • Allow geometries outside of the spatial partitioning to match in a partitioned spatial join.
• Support CREATE VIEW and CREATE TABLE verification #15195 (Author: Leiqing Cai): Support CREATE VIEW and CREATE TABLE verification
  • Add support to verify CREATE VIEW and CREATE TABLE queries.
• Optimize MetadataQueryOptimizer with filter pushdown enabled #15227 (Author: Shixuan Fan): Optimize MetadataQueryOptimizer with filter pushdown enabled
  • Optimize metadata query optimizer so that it does not fetch all partitions from metastore when hive filter pushdown is enabled.
• Support pushdown approx_distinct(x, e) into pinot #15231 (Author: Xiang Fu): Support pushdown approx_distinct(x, e) into pinot
  • Support pushing down aggregation function approx_distinct(x, e) to Pinot connector.
• Adding array functions pushdown to pinot #15260 (Author: Xiang Fu): Adding array functions pushdown to pinot
  • Support pushing down array functions array_sum, array_min, array_max, array_average, and contains to Pinot connector.
• Add option to do more precise ACL checks #15269 (Author: prithvip): Add option to do more precise ACL checks
  • Add session property check_access_control_on_utilized_columns_only, which, when enabled, only performs access control checks on columns that would actually be required to produce the query output, ignoring columns that are referenced in the query, but are not required to compute the query results.
• Remove unnecessary config and session property #15271 (Author: Rongrong Zhong): Remove unnecessary config and session property
  • Remove config property optimizer.optimize-full-outer-join-with-coalesce and the corresponding session property optimize_full_outer_join_with_coalesce. The feature will always be enabled.
• Add local cache for Parquet Metadata #15276 (Author: Vivek): Add local cache for Parquet Metadata
  • Add caching module for Parquet metadata.
• Ignore broadcast memory limit in verifier #15281 (Author: Ariel Weisberg): Ignore broadcast memory limit in verifier
  • Improve verifier handling of broadcast memory limit by removing it if it is set via session property to avoid CBO broadcast induced failures on smaller verifier clusters.
• Support formatting Use statement #15293 (Author: Leiqing Cai): Support formatting Use statement
  • Fix query failure when explaining Use statement.
• Add session property remote_functions_enabled #15305 (Author: Rongrong Zhong): Add session property remote_functions_enabled
  • Add session property remote_functions_enabled to configure whether remote functions are allowed.
• Remove deprecated grouped execution properties #15317 (Author: Rebecca Schlussel): Remove deprecated grouped execution properties
  • Remove deprecated configuration properties grouped-execution-for-aggregation-enabled and grouped-execution-for-join-enabled and their corresponding session properties grouped_execution_for_aggregation and grouped_execution_for_join. These have been replaced by a single configuration property grouped-execution-enabled and session property grouped-execution.
  • Remove deprecated configuration property dynamic-schedule-for-grouped-execution and session property dynamic_schedule_for_grouped_execution. No replacement is needed.

All Commits

• d577306 Move PageDataOutput to presto-spi (Wenlei Xie)
• d879d7f Move DataSink/DataOutput to presto-common (Wenlei Xie)
• bef17e6 Add option to do more precise ACL checks (prithvip)
• 4479fb8 Remove deprecated grouped execution properties (Rebecca Schlussel)
• 061f432 Allow qualified names in types parsing (Daniel Ohayon)
• 4fdfb89 Add sealedPartition flag to HivePartition (Nikhil Collooru)
• 029ae8d Add a test for adding a new column (Mayank Garg)
• 84ade36 Refactor ParquetPageSource (Zhenxiao Luo)
• 2e0c44f Add scale_tdigest function (leonpanokarren)
• d3bd4ec Bug fixes for file_renaming feature (Nikhil Collooru)
• 7a122bb Handle cases when flat map doesn't have dummy encoding for sequence 0 (Zhenyuan Zhao)
• ee74e4b Use getApproximateLogicalSizeInBytes() in PageProcessor (Ying Su)
• 7bffeb5 Add getApproximateRegionLogicalSizeInBytes() to Block (Ying Su)
• 465644c Create DictionaryBlock for identity projections (Ying Su)
• 57a1b20 Update drift-api module's scope to provided (Ajay George)
• 118b48a Add thrift serde support for TaskStatus (Ajay George)
• 213660b Add local cache for Parquet Metadata (Vivek)
• f2965f6 Add support for TRUNCATE function for DOUBLE and REAL (George Wang)
• 4631b6d Add session property remote_functions_enabled (Rongrong Zhong)
• 321c8b6 Adding approxMostFrequent aggregate from prestoSQL (ankit0811)
• f1fb2e0 Remove setFunctionAndTypeManager from TypeRegistry (Rongrong Zhong)
• ac735db Use FunctionAndTypeManager instead of TypeRegistry (Rongrong Zhong)
• 7fffbd8 Remove TypeManager from BuiltInFunctionNamespaceManager (Rongrong Zhong)
• ca07a42 Remove TypeManager from APIs when FunctionAndTypeManager is available (Rongrong Zhong)
• 63af935 Change FunctionManager to FunctionAndTypeManager (Rongrong Zhong)
• 4720be3 Allow PrestoS3FileSystem and GlueHiveMetastore to assume an aws role (Ashish Tadose)
• 8cbf330 Allow PrestoS3FileSystem and GlueHiveMetastore to assume an aws role (Ashish Tadose)
• 3ec373e bump aws-sdk to 1.11.697 (Ashish Tadose)
• ac60c75 Support formatting Use statement (Leiqing Cai)
• ce96d88 Fix a potential NPE by allowing nulllable cache base directory (Bin Fan)
• 4de1f4d Optimize MetadataQueryOptimizer with filter pushdown enabled (Shixuan Fan)
• 05f3875 Fix ValuesMatcher (Shixuan Fan)
• 8ab725f Revert "Defer the creation of dictionary in SliceDictionarySelectiveReader" (Ke)
• 460a4d0 Revert "Remove stripeDictionaryData buffer in SliceDictionarySelectiveReader" (Ke)
• a8825ab Revert "Introduce large dictionary mode in SliceDictionarySelectiveReader" (Ke)
• 83e4afc Add a new deployment example (Bin Fan)
• 03fb2f6 Ignore broadcast memory limit in verifier (Ariel Weisberg)
• 9870fb8 Rename PartitionVersionFetcher to PartitionMutator (James Sun)
• 55508df Remove unnecessary config and session property (Rongrong Zhong)
• 63565a9 Improve spatial join tests (James A. Gill)
• 4cd357c Assign partition index to geometries outside of KdbTree (James A. Gill)
• 89c3996 Add eligibleToIgnore flag to Hive partitions (James Sun)
• ca06ed5 Allow partial filter failures in BenchmarkPageProcessor (Ying Su)
• c0455e9 Add identity projection to BenchmarkPageProcessor (Ying Su)
• 883e883 Support CREATE TABLE verification (Leiqing Cai)
• a50fb09 Support CREATE VIEW verification (Leiqing Cai)
• 2f6965d Fix injection error when failure resolvers are disabled (Leiqing Cai)
• facde22 Rename DataQueryBundle to QueryObjectBundle (Leiqing Cai)
• a81016c Remove unsupported query types (Leiqing Cai)
• 79a8c8f Enforce broadcast memory limits in Presto on Spark (Andrii Rosa)
• 13a8f64 Compress broadcast pages in Presto on Spark to save memory (Andrii Rosa)
• d87cee5 Make sure PagesSerde is not used in a non thread safe manner (Andrii Rosa)
• 322321d Revert "Enable nullifying iterator for broadcast join" (Andrii Rosa)
• 299f751 Support for CSV format in hive (Ashish Tadose)
• 9f4203e Support pushdown approx_distinct(x, e) into pinot (Xiang Fu)
• 5c29aa2 Adding array functions pushdown to pinot (Xiang Fu)
• a930829 Add unit tests on async cache restore (Bin Fan)
• 6ff3052 Add BIGINT benchmark for projection (Ying Su)
• 3a44824 Add GCProfilter to BenchmarkPageProcessor (Ying Su)
• 186ee27 Add verify methods for BenchmarkPageProcessor (Ying Su)
• 632c403 Allow 100% filter pass rate in BenchmarkPageProcessor (Ying Su)
• 1d3ef17 Move data preparation to BenchmarkData for BenchmarkPageProcessor (Ying Su)

[rongrong]

General Changes
* Add function ``truncate(x, n)``.
* Add aggregation function ``approxMostFrequent``.

[mayankgarg1990]

I highly recommend reading https://github.com/prestodb/presto/wiki/Release-Notes-Guidelines

[mayankgarg1990]

last few nits - but overall good to go