Error in reduced schedule CSV file

[dt-woods]

In the read_eia923_fuel_receipts method in coal_upstream.py, Page 5 of the EIA923 Excel workbook is saved to CSV. The header columns include a new line character "\n" for the following coal min columns:

• Coalmine\nType
• Coalmine\nState
• Coalmine\nCounty
• Coalmine\nMsha Id

When written to CSV, these headers are all truncated to "Coalmine," dropping the context after the newline. This results in a CSV file with four columns all of the same name and causes errors with merging.

ElectricityLCI/electricitylci/coal_upstream.py

Line 90 in e562681

eia_fuel_receipts_df.to_csv(csv_path, index=False)

To fix, consider running the data frame through the _clean_columns method before writing to CSV.

A symptom of this is a 'KeyError' on key 'fuel_group', accessed in generate_upstream_coal_map from the data frame returned by read_eia_fuel_receipts.

[dt-woods]

The worksheet:

And the reduced CSV:

[dt-woods]

Note: to implement this fix on your machine, you need to delete any CSV files in the f923_YEAR folders in your data directory.