Release 394 (29 Aug 2022)#
General#
Add support for JSON as an output format of
EXPLAIN
. (#12968)Improve performance of queries involving joins on a single
bigint
column. (#13432)Improve performance of
LIKE
expressions. (#13479)Ensure
UPDATE
queries cannot insert null values into columns with aNOT NULL
constraint. (#13435)Fix failure when an
UPDATE
query contains aWHERE
clause which always evaluates to false. (#12422)Fix potential failure for queries involving joins and implicit or explicit casts of
null
to a concrete type. (#13565)
Docker image#
Ensure Trino stops running with insufficient resources to avoid partial failures. (#13736)
BigQuery connector#
Delta Lake connector#
Druid connector#
Improve performance of queries that perform filtering on
varchar
columns that contain temporal data with the formatYYYY-MM-DD
. (#12925)
Elasticsearch connector#
Add support for multiple hosts in the
elasticsearch.host
configuration property. (#12530)
Hive connector#
Iceberg connector#
Improve query planning performance when a
varchar
partitioning column contains date values in theYYYY-MM-DD
format. (#12925)Fix query failure when using the
[VERSION | TIMESTAMP] AS OF
clause on a table created with Iceberg versions older than 0.12. (#13613)Fix failure when reading Parquet data that contains only null values. (#9424)
Oracle connector#
Improve performance of queries that perform filtering on
varchar
columns that contain temporal data with the formatYYYY-MM-DD
. (#12925)
Phoenix connector#
Improve performance of queries that perform filtering on
varchar
columns that contain temporal data with the formatYYYY-MM-DD
. (#12925)
Pinot connector#
PostgreSQL connector#
SQL Server connector#
Prevent renaming a column to a name which is longer than the maximum length supported by SQL Server. Previously, long names were truncated. (#13742)
SPI#
Add the query plan in JSON format to
QueryCompletedEvent
, and allow connectors to request anonymized query plans in theQueryCompletedEvent
. (#12968)