If you already have your own `FileFormat` implementation: just override the `supportBatch` method.
On Tue, Jun 16, 2020 at 5:39 AM Nasrulla Khan Haris <nasrulla.k...@microsoft.com.invalid> wrote: > HI Spark developers, > > > > FileSourceScanExec > <https://github.com/apache/spark/blob/807e0a484d1de767d1f02bd8a622da6450bdf940/sql/core/src/main/scala/org/apache/spark/sql/execution/DataSourceScanExec.scala#L159-L167> > extends ColumnarBatchScan which internal converts columnarbatch to > InternalRows, If I have a new Datasource/FileFormat which uses > customrelation instead of HadoopFsRelation, Driver uses > RowDataSourceScanExec > <https://github.com/apache/spark/blob/807e0a484d1de767d1f02bd8a622da6450bdf940/sql/core/src/main/scala/org/apache/spark/sql/execution/DataSourceScanExec.scala#L78-L86>, > this causes castexeption from internalRow to columnarBatch. Is there a way > to provide ColumnarBatchScan support to customrelation ? > > > > Appreciate your inputs. > > > > Thanks, > > NKH > > >