Count Rows, Columns and Cells

records vs lines · columns · cells · non-empty · size · per-column fill

How many rows does this CSV actually have? Not the same question as how many lines the file has — a quoted field can contain newlines, so wc -l lies. This counts real CSV records, columns, cells, and how many of those cells are actually populated, and tells you when the two line counts disagree.

How to count rows in a CSV

1. Paste the CSV or drop a file — dropping is better for large files and gives you the true byte size.
2. Read data rows for the answer most people want: records excluding the header.
3. Compare records against physical lines. If they differ, some fields contain line breaks.
4. Check the per-column table for how many values each column actually has.

Why wc -l disagrees with your row count

CSV allows a newline inside a quoted field, so one record can span several lines in the file:

id,address
1,"12 Main St
Kyiv"

That's two records (header plus one row) across three lines. Line-counting tools report three, importers report one data row, and the disagreement is where "my import lost rows" tickets come from. This tool shows both numbers and flags multi-line fields explicitly.

What each number means

• records — parsed CSV rows, including the header.
• data rows — records minus the header, unless you tick the no-header box.
• cells — records × columns, i.e. the size of the grid.
• non-empty cells — cells with something other than whitespace. The gap between this and cells is your sparseness.
• ragged rows — rows whose field count differs from the widest row. Any non-zero value here means the file will misalign somewhere; Repair CSV fixes it.
• size — the real file size when you drop a file, or the size of the pasted text.

FAQ

Privacy

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