dbtsai commented on a change in pull request #27728: [SPARK-17636][SQL] Nested
Column Predicate Pushdown for Parquet
URL: https://github.com/apache/spark/pull/27728#discussion_r385983948
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetFilters.scala
##########
@@ -49,15 +49,29 @@ class ParquetFilters(
pushDownInFilterThreshold: Int,
caseSensitive: Boolean) {
// A map which contains parquet field name and data type, if predicate push
down applies.
- private val nameToParquetField : Map[String, ParquetField] = {
- // Here we don't flatten the fields in the nested schema but just look up
through
- // root fields. Currently, accessing to nested fields does not push down
filters
- // and it does not support to create filters for them.
- val primitiveFields =
-
schema.getFields.asScala.filter(_.isPrimitive).map(_.asPrimitiveType()).map { f
=>
- f.getName -> ParquetField(f.getName,
- ParquetSchemaType(f.getOriginalType,
- f.getPrimitiveTypeName, f.getTypeLength, f.getDecimalMetadata))
+ private val nameToParquetField : Map[String, ParquetPrimitiveField] = {
Review comment:
Done.
----------------------------------------------------------------
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:
[email protected]
With regards,
Apache Git Services
---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]
