The diff is not that pretty, but the main thing that I've changed is that I'm using an existing CSV library to parse the CSV (I had some nested commas in my descriptions).
Basically most of the actual processing CSV code is now nested within a csv.parse callback.
The diff is not that pretty, but the main thing that I've changed is that I'm using an existing CSV library to parse the CSV (I had some nested commas in my descriptions).
Basically most of the actual processing CSV code is now nested within a
csv.parse
callback.