Steve Carlin has posted comments on this change. ( 
http://gerrit.cloudera.org:8080/23562 )

Change subject: IMPALA-13902: Calcite planner: Implement is_spool_query_results
......................................................................


Patch Set 5:

(4 comments)

http://gerrit.cloudera.org:8080/#/c/23562/2/fe/src/main/java/org/apache/impala/analysis/AnalysisContext.java
File fe/src/main/java/org/apache/impala/analysis/AnalysisContext.java:

http://gerrit.cloudera.org:8080/#/c/23562/2/fe/src/main/java/org/apache/impala/analysis/AnalysisContext.java@a506
PS2, Line 506:
             :
             :
             :
> I don't quite get the logic of why createDataSink makes sense, but it looks
So basically, I guess I didn't like keeping the logic in here.

This logic is shared logic by both the original planner and the Calcite 
planner, but this only applies to the original planner.  By moving it to 
QueryStmt, it keeps AnalysisContext free of planner specific code.

The problem also comes from the fact that it's too early for the Calcite 
planner to know if it "returns at most one row".  The original planner has this 
information at Analysis time, but Calcite does not have this information at 
Analysis/validation time.


http://gerrit.cloudera.org:8080/#/c/23562/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/service/CalciteSingleNodePlanner.java
File 
java/calcite-planner/src/main/java/org/apache/impala/calcite/service/CalciteSingleNodePlanner.java:

http://gerrit.cloudera.org:8080/#/c/23562/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/service/CalciteSingleNodePlanner.java@130
PS5, Line 130:     while (relNode instanceof Filter || relNode instanceof 
Project) {
> Is there a more generalized version of this? Seems like it might be somethi
Changed this to call a method that handles this.


http://gerrit.cloudera.org:8080/#/c/23562/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/service/ExecRequestCreator.java
File 
java/calcite-planner/src/main/java/org/apache/impala/calcite/service/ExecRequestCreator.java:

http://gerrit.cloudera.org:8080/#/c/23562/5/java/calcite-planner/src/main/java/org/apache/impala/calcite/service/ExecRequestCreator.java@253
PS5, Line 253:     // This is a deprecated class, so the spool option is not 
supported here and set
> Is there somewhere else talking about why it's deprecated?
I just filed IMPALA-14541

The IMPALA-14408 patch was made so that queries go through JniFrontend instead 
of CalciteJniFrontend.  This class is only used by CalciteJniFrontend.


http://gerrit.cloudera.org:8080/#/c/23562/5/testdata/workloads/functional-query/queries/QueryTest/calcite.test
File testdata/workloads/functional-query/queries/QueryTest/calcite.test:

http://gerrit.cloudera.org:8080/#/c/23562/5/testdata/workloads/functional-query/queries/QueryTest/calcite.test@1127
PS5, Line 1127: row_regex: .*SPOOL_QUERY_RESULTS=0.*
> Should there be a SPOOL_QUERY_RESULTS=1 case?
I'm ok with adding one, but I didn't add it for two reasons:

1) spool_query_results is defaulted to true, so this is testing that new 
behavior works
2) There are e2e tests that will eventually test this when the e2e tests are 
turned on.

I always kinda struggle to find a balance of what to put in this file and what 
I should just wait for when e2e tests are eventually tested on a nightly basis. 
 If you think it should be here though, I can add it, I have no issue with that.



--
To view, visit http://gerrit.cloudera.org:8080/23562
To unsubscribe, visit http://gerrit.cloudera.org:8080/settings

Gerrit-Project: Impala-ASF
Gerrit-Branch: master
Gerrit-MessageType: comment
Gerrit-Change-Id: I5b9bf49e2874ee12de212b892bd898c296774c6f
Gerrit-Change-Number: 23562
Gerrit-PatchSet: 5
Gerrit-Owner: Steve Carlin <[email protected]>
Gerrit-Reviewer: Aman Sinha <[email protected]>
Gerrit-Reviewer: Fang-Yu Rao <[email protected]>
Gerrit-Reviewer: Impala Public Jenkins <[email protected]>
Gerrit-Reviewer: Joe McDonnell <[email protected]>
Gerrit-Reviewer: Michael Smith <[email protected]>
Gerrit-Reviewer: Riza Suminto <[email protected]>
Gerrit-Reviewer: Steve Carlin <[email protected]>
Gerrit-Comment-Date: Fri, 07 Nov 2025 22:11:03 +0000
Gerrit-HasComments: Yes

Reply via email to