Extract CSV Columns

RFC 4180 with permissive line endings. Fields can be wrapped in "…"; inside a quoted field, "" represents a literal "; quoted fields can contain the delimiter and newlines. Unquoted fields stop at the delimiter or end of line. CR, LF, and CRLF are all accepted as row terminators.

Keep “First row is a header” checked, then put header names in the selector field – comma-separated, case-insensitive. name,email picks those columns regardless of position. Unknown names are reported as a warning in the stats line (rather than failing the whole extraction).

Yes – the output column order matches the selector order. Type 3,1,2 and the output has column 3 first, then 1, then 2. Combine with names: email,name,age reorders even when those aren’t in that order in the source.

With a header row, JSON output is an array of objects keyed by the extracted header names – e.g. [{"name":"Smith","email":"jsmith@..."}, …]. Without a header row, it’s an array of arrays. Trailing empty rows are not deduplicated; each input row produces one output entry.

The parser is O(n) and runs entirely in your browser’s main thread. Typical limit before lag becomes noticeable: ~100 MB of CSV text. The 200 ms input debounce keeps typing-while-editing smooth even on large pasted inputs.

Quoted text round-trips cleanly. The output quoter wraps a field in "…" (and doubles internal quotes) only when the field contains the output delimiter, a quote, or a CR/LF. Otherwise it leaves the field unquoted. So plain stays plain, but he said "hi" becomes "he said ""hi""".

The parser uses the first row’s column count to define how many columns exist. If subsequent rows have fewer fields, missing selectors produce empty cells (''). If subsequent rows have extra fields, the extras are simply ignored – selectors are projected from the columns the parser saw on that row, with empties for any beyond.

No. The page loads three static files (HTML, CSS, JS) and then runs entirely in your browser. Your CSV bytes never leave the device – no fetch, no XHR, no analytics, no cookies. You can disconnect from the internet after the page loads and the tool still works.

Yes – free, unlimited, no signup, no watermark. The extracted CSV/TSV/JSON is yours to use anywhere. Attribution to is appreciated but not required.