singhpk234 commented on code in PR #36442:
URL: https://github.com/apache/spark/pull/36442#discussion_r864555565
##########
sql/core/src/test/scala/org/apache/spark/sql/ColumnExpressionSuite.scala:
##########
@@ -2988,4 +2988,16 @@ class ColumnExpressionSuite extends QueryTest with
SharedSparkSession {
checkAnswer(uncDf.filter($"src".ilike("ѐёђѻώề")), Seq("ЀЁЂѺΏỀ").toDF())
// scalastyle:on
}
+
+ test("SPARK-39093: divide period by integral expression") {
+ val df = Seq(((Period.ofDays(10)), 2)).toDF("pd", "num")
+ checkAnswer(df.select($"pd" / ($"num" + 3)),
+ Seq((Period.ofDays(2))).toDF)
+ }
+
+ test("SPARK-39093: divide duration by integral expression") {
+ val df = Seq(((Duration.ofDays(10)), 2)).toDF("dd", "num")
+ checkAnswer(df.select($"dd" / ($"num" + 3)),
+ Seq((Duration.ofDays(2))).toDF)
+ }
Review Comment:
[question] should we add this test with these SQL Confs :
(i) WHOLESTAGE_CODEGEN_ENABLED (true) , CODEGEN_FALLBACK (false)
(ii) WHOLESTAGE_CODEGEN_ENABLED (false)
As even when the codegen fails by default it will be retried in interpreted
mode (as you already called out) and hence checkAnswer would not be able to
catch this. And similar is the case with WSCG off as it may pass in Codegen but
fail in Interpreted mode, hence we might not be able to catch that. Your
thoughts ?
--
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]
