Open pedrobernardina opened 6 years ago
When we have a big integer on the dataset, schema is inferred as integer. As a result, on the workflow, all rows are set to null. Changing manually the data type to long solves the problem.
When we have a big integer on the dataset, schema is inferred as integer. As a result, on the workflow, all rows are set to null. Changing manually the data type to long solves the problem.