Slice CSV Rows

first N · last N · row range · every Nth · skip from either end · header kept

Cut a CSV down to the rows you need — the first 100 for a quick look, the last 50 to see recent records, rows 500–600 to inspect the middle, or every tenth row to thin a huge export. It's head and tail for CSV, except the header row is always carried through so the result stays a valid file.

How to slice a CSV

1. Paste the CSV or drop a file.
2. Pick a mode and set N. For a range, N is the first row and to row is the last — both counted from 1, both inclusive, ignoring the header. Fields that don't apply to the current mode are hidden.
3. Read the status line to confirm how many rows were kept out of how many.
4. Copy or download the result.

What each mode is for

• First N — a sample to eyeball the shape of a file, or a small test fixture.
• Last N — the newest records in an append-ordered export, or checking whether a file was truncated.
• Rows N to M — inspecting the region around a row that failed to import.
• Every Nth — thinning a dense time series so a chart stays readable, keeping the spread of the whole file rather than one end of it.
• All but the first N — dropping preamble rows a report generator added above your data.
• All but the last N — removing a totals row, or a trailing summary block.

Slicing versus sampling

Slicing is positional and reproducible: the same input always gives the same rows. That's what you want for debugging a specific region or trimming known junk. If you want rows that represent the file rather than one part of it — because it's sorted by date, region, or ID and any slice would be biased — use Random sample, which can also take a seed so the selection is repeatable.

FAQ

Privacy

100% client-side. No upload. See the privacy policy.