Github
来源:Presto
浏览 337
扫码
分享
2020-04-20 22:54:31
12.110. Release 0.137
General Changes
- Fix
current_date
to return correct results for all time zones. - Fix invalid plans when scalar subqueries use
GROUP BY
, DISTINCT
or JOIN
. - Do not allow creating views with a column type of
UNKNOWN
. - Improve expression optimizer to remove some redundant operations.
- Add
bit_count()
, bitwise_not()
, bitwise_and()
,bitwise_or()
, and bitwise_xor()
functions. - Add
approx_distinct()
aggregation support for VARBINARY
input. - Add create time to query detail page in UI.
- Add support for
VARCHAR(length)
type. - Track per-stage peak memory usage.
- Allow using double input for
approx_percentile()
with an array ofpercentiles. - Add API to JDBC driver to track query progress.
Hive Changes
- Do not allow inserting into tables when the Hive type does not matchthe Presto type. Previously, Presto would insert data that did notmatch the table or partition type and that data could not be read byHive. For example, Presto would write files containing
BIGINT
data for a Hive column type of INT
. - Add validation to CREATE TABLE and CREATE TABLE ASto check that partition keys are the last columns in the table and in the sameorder as the table properties.
- Remove
retention_days
table property. This property is not used by Hive. - Fix Parquet decoding of
MAP
containing a null value. - Add support for accessing ORC columns by name. By default, columns in ORCfiles are accessed by their ordinal position in the Hive table definition.To access columns based on the names recorded in the ORC file, set
hive.orc.use-column-names=true
in your Hive catalog properties file.