sunchao commented on a change in pull request #34199:
URL: https://github.com/apache/spark/pull/34199#discussion_r726599470
##########
File path:
sql/core/src/main/scala/org/apache/spark/sql/execution/datasources/parquet/ParquetSchemaConverter.scala
##########
@@ -60,40 +58,106 @@ class ParquetToSparkSchemaConverter(
/**
* Converts Parquet [[MessageType]] `parquetSchema` to a Spark SQL
[[StructType]].
*/
- def convert(parquetSchema: MessageType): StructType =
convert(parquetSchema.asGroupType())
+ def convert(parquetSchema: MessageType): StructType = {
+ val column = new ColumnIOFactory().getColumnIO(parquetSchema)
+ val converted = convertInternal(column)
+ converted.sparkType.asInstanceOf[StructType]
+ }
- private def convert(parquetSchema: GroupType): StructType = {
- val fields = parquetSchema.getFields.asScala.map { field =>
- field.getRepetition match {
- case OPTIONAL =>
- StructField(field.getName, convertField(field), nullable = true)
+ /**
+ * Convert `parquetSchema` into a [[ParquetType]] which contains its
corresponding Spark
Review comment:
If you compute the level info in a different place, then you'd still
need to repeat most of the logic in `ParquetSchemaConverter`, e.g., checking
legacy list/map formats. Therefore, I think it's better to reuse the existing
logic instead of splitting them into different places.
--
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]
