Quick notes for development

Libre Office Calc (or any other spreadsheet calculation software) is your friend! Use it to open the different CSV files, compare and check or modify them (e.g. to replace all commas in fields, you can easily use the built-in search and replace functionality).
All the temporary files of this script are saved in a directory in /tmp/crops-parser-... (see the TMPDIR constant in the script). You can check the files and steps there to debug the script.
Further manual adjustments made to the fao-country-db.csv:
- Due to the way YAML treats Norway (yes, search for that, it's funny), the NO 2-letter-key was quoted. See issue 48.
- Palestine was manually assigned the PS code as per this source, so it could be mapped without an alpha-3 code.
- That you get an error for "WARNING: No language code for China could be found. Skip." is to be expected, because the FAQ language mapping defines no 2-letter language for this entry, but for China, mainland instead. To not wrongly aggregate already aggregated data again, this should stay like it is. (I.e. we should not define the CN for "China" again, because it has been assigned to China, mainland according to the FAO data.)

Suggesting new OSM crops