fix the isseue with grouping match
added average aggregation column
changes to support avg for group by columns match
added the changes for average
removed the avg changes
average table scan optimization
fix UTs
fix UTs
add UTs
added UTs
fix UTs
added UT
added UTs
changed the UTs
edited UTs
edited UTs
added the session startree false in UTs
changes for group by match prior to cube scan node
avg aggregator source symbol changes
segregated the unit tests
checkstyle fixes
changes for Unit Tests, lineitem table replace the nation table
UTs are divided into smaller methods
added documentation comments for test method
changed the UT tests
changes for group match
updated hetu-docs
edits the create cubes of the UTs
fixes for avg aggregation
fix uts
checkstyle fix
fix the average aggregation function issue
added unit tests
unit test fixex
fix for symbol assignments
added ut tests, removed the conditions for checking duplicate symbol in AggregationRewriteWithCube
Added UT
code clean up
changed the UT with table scan aserts
edited the UTs
fixed cube name
removed duplicates
fix UTs
edited UTs
edited UTs
edit UTs
Hetu cube cli initial commit
Initial commit cube cli changes with memory efficiency changes
comparison expressions changes commit
presto-cli checkstyle changes
commits with expression conversions
commit for query string truncate
changes for literal type
code clean up
delete the presto-cube-cli module added previously.
client options changes
added sqlformatter for create cube query modification
clean up code
Added fallback mechanism and unit tests
Changes include running CLI with parameters
code clean up changes
Code clean up
updated output format
Reverting the server config changes
Code review changes
Code review changes
added hetu-docs description
updated the documentation in star-tree-cube
Code review changes
updated hetu docs
Changes done to the support literals
divided the methods
Added comments
Seprated the classes Console and CubeConsole
Code clean up changes
Added checkstyle changes
changes for single insert query if range is within processing limit
1. Block INSERT INTO CUBE if source table has been updated.
2. Block duplicate data insert into the cube
3. Insert entire data from source table into cube
4. Updated hetu docs
Support registry external function in connector and implement a example in the mysql jdbc connector;
store external function stored in memory;
Support push external function push down to jdbc connector and add mysql connector example.
Co-Author: bryawongsz <bryanwangsz@163.com>
Co-Author: Yize Li <liyizeyouxiang@live.cn>
Co-Author: chenpingzeng <chenpingzeng@huawei.com>
Co-Author: wyy566 <531938832@qq.com>
Partition column based partitioning for writes & parallel metastore operations using threadpool
Reduce redundant HMS calls
Fixed multiple writers with non-partitioned table
Adding VACUUM FULL MERGE Command
Using bulk APIs to update the partition column statistics
Fixing various issues
Using partition column for partitioning update/delete
Using the local sorting instead of Global orderby
Using threadpool for adding partitions, connecting in round-robin to multiple HMSs.
UT and Fixes for DM Optimzations for Insert Update Delete Vacuum
MERGE name conflict resolved
Skip Stats collection for delete and parallelize commit preparation operations
review comment fixes
- remove presto-cli's direct dependency on hindex
- add new CreateIndexOperator
- when a CREATE INDEX statement is received, rewrite the query as a TableScan and use the data from TableScan to create index
- hindex no longer has a direct dependency on hive packages, it now uses the hive connector
- support index creation at different levels, e.g. file level or partition level
- remove BitmapIndex support, the previous version had issues; a new version will be added later
- add metadata to Pages, currently only enabled for hindex usecase
Raghunandan
i-robot