godfreyhe opened a new pull request #8440: [FLINK-12509] [table-planner-blink] 
Introduce planner rules about non semi/anti join
URL: https://github.com/apache/flink/pull/8440
 
 
   
   ## What is the purpose of the change
   
   *Introduce planner rules about non semi/anti join*
   
   ## Brief change log
   
     - *added JoinConditionEqualityTransferRule that converts Join's conditions 
to the left or right table's own independent filter as much as possible, so 
that the rules of filter-push-down can push down the filter to below*
     - *added JoinConditionTypeCoerceRule that coerces the both sides of 
EQUALS(`=`) operator in Join condition to the same type while sans nullability*
     - *added JoinDependentConditionPushDownRule that extracts some 
sub-conditions in the Join OR condition that can be pushed into join inputs by 
FlinkFilterJoinRule*
     - *added JoinDeriveNullFilterRule that filters null values before join if 
the count null value from join input is greater than 
sql.optimizer.join.null.filter.threshold*
     - *added SimplifyJoinConditionRule that apply various simplifying 
transformations on join condition*
   
   
   ## Verifying this change
   
   This change added tests and can be verified as follows:
   
     - *added JoinConditionEqualityTransferRuleTest that validates 
JoinConditionEqualityTransferRule*
     - *added JoinConditionTypeCoerceRuleTest that validates 
JoinConditionTypeCoerceRule*
     - *added JoinDependentConditionPushDownRuleTest that validates 
JoinDependentConditionPushDownRule*
     - *added JoinDeriveNullFilterRuleTest that validates 
JoinDeriveNullFilterRule*
     - *added SimplifyJoinConditionRuleTest that validates 
SimplifyJoinConditionRule*
   
     - *Added integration tests for end-to-end deployment with large payloads 
(100MB)*
     - *Extended integration test for recovery after master (JobManager) 
failure*
     - *Added test that validates that TaskInfo is transferred only once across 
recoveries*
     - *Manually verified the change by running a 4 node cluser with 2 
JobManagers and 4 TaskManagers, a stateful streaming program, and killing one 
JobManager and two TaskManagers during the execution, verifying that recovery 
happens correctly.*
   
   ## Does this pull request potentially affect one of the following parts:
   
     - Dependencies (does it add or upgrade a dependency): (yes / **no**)
     - The public API, i.e., is any changed class annotated with 
`@Public(Evolving)`: (yes / **no)**
     - The serializers: (yes / **no** / don't know)
     - The runtime per-record code paths (performance sensitive): (yes / **no** 
/ don't know)
     - Anything that affects deployment or recovery: JobManager (and its 
components), Checkpointing, Yarn/Mesos, ZooKeeper: (yes / **no** / don't know)
     - The S3 file system connector: (yes / **no** / don't know)
   
   ## Documentation
   
     - Does this pull request introduce a new feature? (yes / **no)**
     - If yes, how is the feature documented? (not applicable / docs / JavaDocs 
/ **not documented**)
   

----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
 
For queries about this service, please contact Infrastructure at:
[email protected]


With regards,
Apache Git Services

Reply via email to