Join two CSV files
Join two CSV files on a key column and download one joined CSV.
Working…
Questions people ask
- How are the key columns matched?
- As plain text, exactly. The cell "3" matches the cell "3" and nothing else: no numeric coercion ("3.0" and " 3" match nothing), no trimming, no case folding. Tidy the key columns first if their values must be comparable in a stricter form.
- What happens with duplicate keys?
- Every left row holding a key combines with every right row holding it: two left rows and three right rows on the same key produce six output rows. A single key pairing may not exceed 10,000 combined rows and the whole plan may not exceed 1,000,000 rows; both refusals name the counts and the first affected left row, never the key value.
- Why does the formula policy default to prefix here?
- The other csv tools re-emit your own rows, so they keep your values verbatim by default. A join manufactures a new document, whose cells a spreadsheet may re-interpret as formulas, so the tool defanges risky lead characters (= + - @) by default. Keep passes cells through untouched; refuse stops the run naming the output row and column.
- What are the size limits?
- Each file up to 50 MB, at most 200,000 data rows per side, a 10,000-row cross-product per key, 1,000,000 joined rows and a 256 MB download. The shape caps are checked before anything is written, so an oversized join fails fast with a named refusal instead of a partial file.
- What happens to my files?
- Both uploads are deleted as soon as processing finishes; the joined CSV stays downloadable for about 10 minutes, then is deleted automatically.