Release 0.179#
General#
- Fix issue which could cause incorrect results when processing dictionary encoded data. If the expression can fail on bad input, the results from filtered-out rows containing bad input may be included in the query output (#8262). 
- Fix planning failure when similar expressions appear in the - ORDER BYclause of a query that contains- ORDER BYand- LIMIT.
- Fix planning failure when - GROUPING()is used with the- legacy_order_bysession property set to- true.
- Fix parsing failure when - NFD,- NFC,- NFKDor- NFKCare used as identifiers.
- Fix a memory leak on the coordinator that manifests itself with canceled queries. 
- Fix excessive GC overhead caused by captured lambda expressions. 
- Reduce the memory usage of map/array aggregation functions. 
- Redact sensitive config property values in the server log. 
- Update timezone database to version 2017b. 
- Add - repeat()function.
- Add - crc32()function.
- Add file based global security, which can be configured with the - etc/access-control.propertiesand- security.config-fileconfig properties. See System access control for more details.
- Add support for configuring query runtime and queueing time limits to resource groups. 
Hive#
- Fail queries that access encrypted S3 objects that do not have their unencrypted content lengths set in their metadata. 
JDBC driver#
- Add support for setting query timeout through - Statement.setQueryTimeout().
SPI#
- Add grantee and revokee to - GRANTand- REVOKEsecurity checks.