How to select, reorder and drop CSV columns
Pick CSV columns by name or position, reorder them in one pass, and drop unwanted fields while keeping quoted values intact.
Addressing columns by name or number
A CSV column can be referenced two ways here: by the header name in the first row, or by a 1-based position where the leftmost column is 1. Names are convenient when a header exists and make your selection self documenting, while numbers are the only option when the file has no header. Because both styles resolve to the same underlying column, you can freely mix them, for example listing 'id, 4, status' in a single request.
Keep versus drop mode
Keep mode is a positive selection: only the columns you name survive, and they appear in the exact order you listed them. That makes it the tool for both slicing out a subset and reordering the file at the same time. Drop mode is the inverse. You name the columns to remove and everything else stays in its original order, which is faster when you only want to strip one or two fields from a wide export.
Why RFC 4180 quoting matters
Real world CSV is messier than comma separated values. A field can legitimately contain a comma, a double quote or even a line break as long as the whole field is wrapped in quotes, with internal quotes doubled. A parser that just splits on commas will corrupt those fields and shift every column after them. This extractor implements the RFC 4180 rules on the way in and re-quotes any output field that needs it, so an address like "221B Baker St, London" stays a single cell from input to output.
Reordering to match a target schema
A frequent chore is reshaping an export so its columns line up with what another system expects. Instead of opening a spreadsheet and dragging columns, you can list the target column order once in keep mode and copy the result. Since the tool preserves values exactly and only changes which columns appear and in what order, it is a safe, repeatable way to remap a file before importing it elsewhere.