Closed jnnr closed 1 year ago
Quoting @FelixMau: build_datapackage
tries to pass the dictionary but the return from get_process_df
is a dictionary with process names and csv file names.
expected input:
{process_name: data (dict)}
actual input:
{process_name: {csv_name: data(pd.DataFrame)}}
Maybe the Mapper
class or build_datapackage
function could be extended to take the dictionary of (multiple) Dataframes and parametrize
with all Data corresponding to the passed process
Adds a test that aims to reproduce the minimal example datapackage in tests/_files/tabular_datapackage_minimal_example.
Problems: