pub fn materialize_empty_df( projection: Option<&[usize]>, reader_schema: &ArrowSchema, hive_partition_columns: Option<&[Series]>, row_index: Option<&RowIndex>, ) -> DataFrame
parquet