viirya commented on a change in pull request #24158: [SPARK-26847][SQL] Pruning
nested serializers from object serializers: MapType support
URL: https://github.com/apache/spark/pull/24158#discussion_r268043072
##########
File path:
sql/catalyst/src/main/scala/org/apache/spark/sql/catalyst/optimizer/objects.scala
##########
@@ -132,49 +130,74 @@ object ObjectSerializerPruning extends Rule[LogicalPlan]
{
fields.map(f => collectStructType(f.dataType, structs))
case ArrayType(elementType, _) =>
collectStructType(elementType, structs)
+ case MapType(_, valueType, _) =>
+ // Because we can't select a field from struct in key, so we skip key
type.
+ collectStructType(valueType, structs)
case _ =>
}
structs
}
+ /**
+ * This method returns pruned `CreateNamedStruct` expression given an
original `CreateNamedStruct`
+ * and a pruned `StructType`.
+ */
+ private def pruneNamedStruct(struct: CreateNamedStruct, prunedType:
StructType) = {
+ // Filters out the pruned fields.
+ val prunedFields = struct.nameExprs.zip(struct.valExprs).filter { case
(nameExpr, _) =>
+ val name = nameExpr.eval(EmptyRow).toString
+ prunedType.fieldNames.exists { fieldName =>
+ if (SQLConf.get.caseSensitiveAnalysis) {
+ fieldName.equals(name)
+ } else {
+ fieldName.equalsIgnoreCase(name)
+ }
+ }
+ }.flatMap(pair => Seq(pair._1, pair._2))
+
+ CreateNamedStruct(prunedFields)
Review comment:
`prunedType` contains at least one field in the struct.
----------------------------------------------------------------
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.
For queries about this service, please contact Infrastructure at:
us...@infra.apache.org
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org
For additional commands, e-mail: reviews-h...@spark.apache.org
