The class helps prepare data and metadata files for upload to the OpenEnergyPlatform.
Use with oedatamodel-parameter for:
- create dict with user-defined columns
- insert dict with user-defined columns into oedatamodel-parameter columns "method", "source", "bandwidth_type", "comment"
- make column header postgresql-conform
- update OEMetadata field names, with postgresql-conform column names
Methods
-------
_return_user_defined_columns(self) -> dict
return user defined columns that are neither columns of oedatamodel-parameter scalar or timeseries
create_json_dict_from_user_defined_columns(self) -> dict
read columns and return dict with column names as keys and empty value
insert_user_column_dict_in_csv_based_on_oedatamodel_parameter(self) -> None
insert each csv specific column dicts in respective csv
make_csv_columns_postgresql_conform(self) -> dict
correct columns from csv files to be postgresql-conform and save in csv
update_oemetadata_schema_fields_name_from_csv_using_similarity(self) -> None
update metadata information with actual csv column-header information and write into repective metadata json
"""