CSV Column Extractor

Both. When the first row is a header you can list column names, and you can always use 1-based positions where 1 is the leftmost column. Mixing names and numbers in the same list works, so 'name, 3' is valid.

Yes. In keep mode the output columns follow the exact order you type, so listing 'email, id, name' reorders the file into that sequence. You can even repeat a column to duplicate it in the output.

Yes. The parser follows RFC 4180, so a quoted field such as "Smith, Jane" is read as one cell, embedded double quotes written as two quotes are decoded, and any field that needs quoting is re-quoted in the output.

Drop removes every column you list and keeps the remaining columns in their original left to right order. It is handy for stripping a few sensitive or unused fields without retyping the ones you want to keep.

Clear the 'First row is header' checkbox. With no header you must address columns by their 1-based position, and the first data row is treated as ordinary data rather than names.

The tool shows an inline error such as an unknown column name or an out of range index instead of guessing. Fix the name or number and the result reappears immediately.

Output rows are joined with CRLF, the line ending RFC 4180 specifies, regardless of whether your input used CRLF, CR or LF. Input in any of those styles is parsed correctly.

There is no hard row limit. Everything runs in your browser, so the practical ceiling is your device memory, and typical spreadsheets with thousands of rows process instantly.

No. Parsing and column selection run entirely in your browser, so your data never leaves your device and nothing is stored on a server.