Release 359 (1 Jul 2021)#

General#

  • Raise minimum required Java version for running Trino server to 11.0.11. (#8103)

  • Add support for row pattern recognition in window specification. (#8141)

  • Add support for SET TIME ZONE. (#8112)

  • Add geometry_nearest_points(). (#8280)

  • Add current_groups(). (#8446)

  • Add support for varchar, varbinary and date types to make_set_digest(). (#8295)

  • Add support for granting UPDATE privileges. (#8279)

  • List materialized view columns in the information_schema.columns table. (#8113)

  • Expose comments in views and materialized views in system.metadata.table_comments correctly. (#8327)

  • Fix query failure for certain queries with ORDER BY ... LIMIT on sorted data. (#8184)

  • Fix incorrect query results for certain queries using LIKE with pattern against char columns in the WHERE clause. (#8311)

  • Fix planning failure when using hash_counts(). (#8248)

  • Fix error message when grouping expressions in GROUP BY queries contain aggregations, window functions or grouping operations. (#8247)

Security#

  • Fix spurious impersonation check when applying user mapping for password authentication. (#7027)

  • Fix handling of multiple LDAP user bind patterns. (#8134)

Web UI#

  • Show session timezone in query details page. (#4196)

Docker image#

  • Add support for ARM64. (#8397)

CLI#

  • Add support for logging of network traffic via the --network-logging command line option. (#8329)

BigQuery connector#

  • Add bigquery.views-cache-ttl config property to allow configuring the cache expiration for BigQuery views. (#8236)

  • Fix incorrect results when accessing BigQuery records with wrong index. (#8183)

Elasticsearch connector#

  • Fix potential incorrect results when queries contain an IS NULL predicate. (#3605)

  • Fix failure when multiple indexes share the same alias. (#8158)

Hive connector#

  • Rename hive-hadoop2 connector to hive. (#8166)

  • Add support for Hive views which use GROUP BY over a subquery that also uses GROUP BY on matching columns. (#7635)

  • Add support for granting UPDATE privileges when hive.security=sql-standard is used. (#8279)

  • Add support for inserting data into CSV and TEXT tables with skip_header_line_count table property set to 1. The same applies to creating tables with data using CREATE TABLE ... AS SELECT syntax. (#8390)

  • Disallow creating CSV and TEXT tables with data if skip_header_line_count is set to a value greater than 0. (#8373)

  • Fix query failure when reading from a non-ORC insert-only transactional table. (#8259)

  • Fix incorrect results when reading ORC ACID tables containing deleted rows. (#8208)

  • Respect hive.metastore.glue.get-partition-threads configuration property. (#8320)

Iceberg connector#

  • Do not include Hive views in SHOW TABLES query results. (#8153)

MongoDB connector#

  • Skip creating an index for the _schema collection if it already exists. (#8264)

MySQL connector#

  • Support reading and writing timestamp values with precision higher than 3. (#6910)

  • Support predicate pushdown on timestamp columns. (#7413)

  • Handle timestamp values during forward offset changes (‘gaps’ in DST) correctly. (#5449)

SPI#

  • Introduce ConnectorMetadata#listMaterializedViews for listing materialized view names. (#8113)

  • Introduce ConnectorMetadata#getMaterializedViews for getting materialized view definitions. (#8113)

  • Enable connector to delegate materialized view refresh to itself. (#7960)

  • Allow computing HyperLogLog based approximate set summary as a column statistic during ConnectorMetadata driven statistics collection flow. (#8355)

  • Report output column types through EventListener. (#8405)

  • Report input column information for queries involving set operations (UNION, INTERSECT and EXCEPT). (#8371)