Github user cloud-fan commented on a diff in the pull request:
https://github.com/apache/spark/pull/22755#discussion_r229155463
--- Diff:
sql/core/src/test/scala/org/apache/spark/sql/DataFrameAggregateSuite.scala ---
@@ -669,22 +669,20 @@ class DataFrameAggregateSuite extends QueryTest with
SharedSQLContext {
}
}
- Seq(true, false).foreach { codegen =>
+ withWholeStageCodegenOnAndOff { codegenEnabled =>
test("SPARK-22951: dropDuplicates on empty dataFrames should produce
correct aggregate " +
- s"results when codegen is enabled: $codegen") {
- withSQLConf((SQLConf.WHOLESTAGE_CODEGEN_ENABLED.key,
codegen.toString)) {
- // explicit global aggregations
- val emptyAgg = Map.empty[String, String]
- checkAnswer(spark.emptyDataFrame.agg(emptyAgg), Seq(Row()))
- checkAnswer(spark.emptyDataFrame.groupBy().agg(emptyAgg),
Seq(Row()))
- checkAnswer(spark.emptyDataFrame.groupBy().agg(count("*")),
Seq(Row(0)))
- checkAnswer(spark.emptyDataFrame.dropDuplicates().agg(emptyAgg),
Seq(Row()))
-
checkAnswer(spark.emptyDataFrame.dropDuplicates().groupBy().agg(emptyAgg),
Seq(Row()))
-
checkAnswer(spark.emptyDataFrame.dropDuplicates().groupBy().agg(count("*")),
Seq(Row(0)))
-
- // global aggregation is converted to grouping aggregation:
- assert(spark.emptyDataFrame.dropDuplicates().count() == 0)
- }
+ s"results when codegen is enabled: $codegenEnabled") {
--- End diff --
we can add the test name postfix in `withWholeStageCodegenOnAndOff`, so
that the caller side only need to provide a base name. e.g.
```
def withWholeStageCodegenOnAndOff(testName: String)(f: String => Unit):
Unit = {
Seq("false", "true").foreach { enabled =>
test(s"$testName (whole-stage-codegen ${if enabled "on" else "off"})")
...
}
}
```
---
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
