Aman Sinha has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/15654 )

Change subject: IMPALA-9529: Fix multi-tuple predicates not assigned in column 
masking
......................................................................


Patch Set 3:

(2 comments)

http://gerrit.cloudera.org:8080/#/c/15654/3/fe/src/main/java/org/apache/impala/analysis/Analyzer.java
File fe/src/main/java/org/apache/impala/analysis/Analyzer.java:

http://gerrit.cloudera.org:8080/#/c/15654/3/fe/src/main/java/org/apache/impala/analysis/Analyzer.java@1493
PS3, Line 1493:     if (node instanceof JoinNode) {
It would be useful to tag the JoinNode with some special marker that it is 
somehow involved with the mask view and check it here .. just so we don't do 
unnecessary work for all the JoinNode instances in the plan that are not 
pertinent to the view.


http://gerrit.cloudera.org:8080/#/c/15654/3/fe/src/main/java/org/apache/impala/planner/JoinNode.java
File fe/src/main/java/org/apache/impala/planner/JoinNode.java:

http://gerrit.cloudera.org:8080/#/c/15654/3/fe/src/main/java/org/apache/impala/planner/JoinNode.java@224
PS3, Line 224:     for (TupleDescriptor tuple : 
analyzer.materializeSlots(conjuncts_)) {
Would this loop be penalizing joins in the normal case ? i.e if there is no 
column masking or even with column masking, there's no nested columns involved, 
wouldn't this try to recompute mem layout for the tuples involved in the join 
conjuncts ?  The comment above does mention that 'only tuples of the masked 
tables could be affected..' but is there a more explicit way to identify that ?



--
To view, visit http://gerrit.cloudera.org:8080/15654
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I12f1b59733db5a88324bb0c16085f565edc306b3
Gerrit-Change-Number: 15654
Gerrit-PatchSet: 3
Gerrit-Owner: Quanlong Huang <[email protected]>
Gerrit-Reviewer: Aman Sinha <[email protected]>
Gerrit-Reviewer: Csaba Ringhofer <[email protected]>
Gerrit-Reviewer: Fang-Yu Rao <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Quanlong Huang <[email protected]>
Gerrit-Reviewer: Vihang Karajgaonkar <[email protected]>
Gerrit-Comment-Date: Tue, 07 Apr 2020 01:11:53 +0000
Gerrit-HasComments: Yes

Reply via email to