To be compatible with Microsoft's Spark .Net solution, DataFrame.Collect() should return IEnumerable<Row> instead of List<object[]>.
The Row class should provide Get(string columnName) and Get(int index) methods that return object and that throw an exception when the given columnName or index are not available.
GoEddie
commented
4 months ago
To be compatible with Microsoft's Spark .Net solution,
DataFrame.Collect()should returnIEnumerable<Row>instead ofList<object[]>.TheRowclass should provideGet(string columnName)andGet(int index)methods that returnobjectand that throw an exception when the givencolumnNameorindexare not available.