comphead commented on code in PR #4818:
URL: https://github.com/apache/datafusion-comet/pull/4818#discussion_r4608075178


##########
spark/src/test/scala/org/apache/comet/exec/CometAggregateSuite.scala:
##########
@@ -3844,6 +3844,44 @@ class CometAggregateSuite extends CometTestBase with 
AdaptiveSparkPlanHelper {
     }
   }
 
+  test("kurtosis merges an empty partial like Spark") {

Review Comment:
   This repeats the two-file setup of the test above it. 
`percentile_partials.sql` gets the same multi-partition shape in a SQL fixture: 
`-- Config: spark.sql.files.openCostInBytes=134217728` plus one `INSERT ... 
SELECT /*+ COALESCE(1) */` per file keeps each file in its own partition. Could 
a `kurtosis_partials.sql` hold this case, with the second INSERT carrying the 
NULL row?
   
   The same fixture could also merge ordinary values from two files. Today no 
test compares a multi-partition `kurtosis` result on ordinary data against 
Spark. The Rust merge test uses a `1e-9` tolerance and the SQL fixtures are 
single-partition, so a merge-order change in `moments4_merge` would not be 
caught bit-exactly.



-- 
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.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to