Github user cloud-fan commented on a diff in the pull request: https://github.com/apache/spark/pull/22750#discussion_r226819049 --- Diff: sql/core/src/main/scala/org/apache/spark/sql/execution/DataSourceScanExec.scala --- @@ -168,10 +168,11 @@ case class FileSourceScanExec( // Note that some vals referring the file-based relation are lazy intentionally // so that this plan can be canonicalized on executor side too. See SPARK-23731. - override lazy val supportsBatch: Boolean = relation.fileFormat.supportBatch( - relation.sparkSession, StructType.fromAttributes(output)) + override lazy val supportsBatch: Boolean = { + relation.fileFormat.supportBatch(relation.sparkSession, schema) + } - override lazy val needsUnsafeRowConversion: Boolean = { + private lazy val needsUnsafeRowConversion: Boolean = { if (relation.fileFormat.isInstanceOf[ParquetSource]) { --- End diff -- Our Parquet reader has one more feature than ORC: if vectorized reader is on but whole-stage-codegen is off, we can still read parquet with batches, and return `ColumnarRow`s. In ORC, if whole-stage-codegen is off, we also turn off vectorized reader, so ORC never return `ColumnarRow`. However, I just found that this parquet reader feature is gone. Let's send a new PR to fix it or completely remove it.
--- --------------------------------------------------------------------- To unsubscribe, e-mail: reviews-unsubscr...@spark.apache.org For additional commands, e-mail: reviews-h...@spark.apache.org