Joe McDonnell has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/23017 )

Change subject: IMPALA-13945: Change hash trace to show each node's individual 
contribution
......................................................................


Patch Set 6:

(8 comments)

http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java
File fe/src/main/java/org/apache/impala/planner/PlanNode.java:

http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java@464
PS6, Line 464: final int keyFormatWidth = 100;
> Maybe we can define it as "private static final int" in the class for bette
A private static final int would move a few hundred lines away. This is only 
used here, so I would rather have it close to this location.


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java@468
PS6, Line 468: detailPrefix + "  " + elem.getComment() + ": " +
             :                 hashTrace + "\n"
> Maybe using String.format is clearer?
Done


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java@471
PS6, Line 471: detailPrefix + "  " + elem.getComment() + ":\n"
> String.format can be clearer here too
Done


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java@473
PS6, Line 473: stop_idx
> nit. better to match java conventional naming, stopIdx
Done


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/PlanNode.java@475
PS6, Line 475: detailPrefix + "    [" + hashTrace.substring(idx, stop_idx) + 
"]\n"
> Same as above for String.format
Done


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/TupleCacheInfo.java
File fe/src/main/java/org/apache/impala/planner/TupleCacheInfo.java:

http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/TupleCacheInfo.java@287
PS6, Line 287: comment
> The comment seems to always be non null in our current use cases, how about
I added a precondition in the constructor for HashTraceElement. I think we can 
enforce that in a single location.


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/TupleCacheInfo.java@329
PS6, Line 329: comment
> do we need Preconditions checkNotNull for comment?
Enforced in HashTraceElement's constructor


http://gerrit.cloudera.org:8080/#/c/23017/6/fe/src/main/java/org/apache/impala/planner/TupleCacheInfo.java@351
PS6, Line 351: comment
> do we need Preconditions checkNotNull for comment like s?
Enforced in HashTraceElement's constructor



--
To view, visit http://gerrit.cloudera.org:8080/23017
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: If53eda24e7eba264bc2d2f212b63eab9dc97a74c
Gerrit-Change-Number: 23017
Gerrit-PatchSet: 6
Gerrit-Owner: Joe McDonnell <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Joe McDonnell <[email protected]>
Gerrit-Reviewer: Kurt Deschler <[email protected]>
Gerrit-Reviewer: Michael Smith <[email protected]>
Gerrit-Reviewer: Yida Wu <[email protected]>
Gerrit-Comment-Date: Tue, 22 Jul 2025 20:43:41 +0000
Gerrit-HasComments: Yes

Reply via email to