12.146. Release 0.101
- Add support for CREATE TABLE (in addition to ).
- Add support to DROP TABLE and .
- Add
array_agg()
function. - Add function.
- Add
array_position()
function. - Add function.
- Add support for
millisecond
todate_diff()
and . - Fix excessive memory usage in
map_agg()
. - Fix excessive memory usage in queries that perform partitioned top-N operationswith .
- Optimize ARRAY comparison operators.
- Fix
JOIN
associativity to be left-associative instead of right-associative. - Add
source
column to table in . - Add
coordinator
column toruntime.nodes
table in System Connector. - Add
errorCode
,errorName
anderrorType
toerror
object in REST API(errorCode
previously existed but was always zero). - Fix
DatabaseMetaData.getIdentifierQuoteString()
in JDBC driver. - Handle thread interruption in JDBC driver
ResultSet
. - Add
history
command and support for running previous commands via!n
to the CLI. - Change Driver to make as much progress as possible before blocking. This improvesresponsiveness of some limit queries.
- Add predicate push down support to JMX connector.
- Add support for unary
PLUS
operator. - Improve scheduling speed by reducing lock contention.
- Extend optimizer to understand physical properties such as local grouping and sorting.
- Add support for streaming execution of window functions.
- Make run partitioned, if underlying plan is partitioned.
- Add support for connecting to S3 using EC2 instance credentials.This feature is enabled by default. To disable it, set
hive.s3.use-instance-credentials=false
in your Hive catalog properties file. - Treat ORC files as splittable.
- Change PrestoS3FileSystem to use lazy seeks, which improves ORC performance.
- Fix ORC
DOUBLE
statistic for columns containingNaN
. - Lower the Hive metadata refresh interval from two minutes to one second.
- Invalidate Hive metadata cache for failed operations.
- Support
s3a
file system scheme. - Fix discovery of splits to correctly backoff when the queue is full.
- Add support for non-canonical Parquet structs.
- Add support for accessing Parquet columns by name. By default, columns in Parquetfiles are accessed by their ordinal position in the Hive table definition. To accesscolumns based on the names recorded in the Parquet file, set
hive.parquet.use-column-names=true
in your Hive catalog properties file. - Add JMX stats to PrestoS3FileSystem.
- Add
hive.recursive-directories
config option to recursively scanpartition directories for data.
- Add connector callback for rollback of
INSERT
andCREATE TABLE AS
. - Introduce an abstraction for representing physical organizations of a tableand describing properties such as partitioning, grouping, predicate and columns.
ConnectorPartition
and related interfaces are deprecated and will be removedin a future version.