venkata91 commented on a change in pull request #23803: [SPARK-26894][SQL]
Handle Alias as well in AggregateEstimation to propagate child stats
URL: https://github.com/apache/spark/pull/23803#discussion_r258651159
##########
File path:
sql/catalyst/src/test/scala/org/apache/spark/sql/catalyst/statsEstimation/AggregateEstimationSuite.scala
##########
@@ -45,6 +45,34 @@ class AggregateEstimationSuite extends
StatsEstimationTestBase with PlanTest {
private val nameToColInfo: Map[String, (Attribute, ColumnStat)] =
columnInfo.map(kv => kv._1.name -> kv)
+ test("aggregate with alias") {
+ val tableColumns = Seq("key11", "key12")
+ val groupByColumns = Seq("key11")
+ val attributes = groupByColumns.map(nameToAttr)
+
+ val rowCount = 2
+ val child = StatsTestPlan(
+ outputList = tableColumns.map(nameToAttr),
+ rowCount,
+ // rowCount * (overhead + column size)
+ size = Some(4 * (8 + 4)),
+ attributeStats = AttributeMap(tableColumns.map(nameToColInfo)))
+
+ val testAgg = Aggregate(
+ groupingExpressions = attributes,
+ aggregateExpressions = Seq(Alias(nameToAttr("key12"), "abc")()),
+ child)
+
+ val expectedColStats = Seq("abc" -> nameToColInfo("key12")._2)
+ val expectedAttrStats = toAttributeMap(expectedColStats, testAgg)
+
+ val expectedStats = Statistics(
+ sizeInBytes = getOutputSize(testAgg.output, rowCount, expectedAttrStats),
+ rowCount = Some(rowCount),
+ attributeStats = expectedAttrStats)
+ assert(testAgg.stats == expectedStats)
Review comment:
Makes sense. Done.
----------------------------------------------------------------
This is an automated message from the Apache Git Service.
To respond to the message, please log on GitHub and use the
URL above to go to the specific comment.
For queries about this service, please contact Infrastructure at:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
