Why does Excel mangle my CSV when this does not?

Spreadsheet software guesses types on the way in, which is what turns a leading-zero product code into a number and a date column into something else entirely. This reads the file as text and shows you what is actually in it.

What counts as a malformed row?

A row with more or fewer fields than the header - usually an unescaped comma or an unbalanced quote somewhere upstream. They are kept in the table rather than silently dropped, and flagged above it by row number so you can see what needs fixing. The total is always shown; if there are more than 20, only the first 20 are listed.

Can it open a very large export?

Yes, of any size. A file over 40 MB is never read into memory whole: the page finds where its rows start in one quick pass, then reads only the part you are looking at. Filtering, column summaries and exports each read the file through once more, a piece at a time, so a multi-gigabyte export takes as long as your disk takes to read it. Sorting and charts need the whole file in memory, so they are for files under 40 MB.

Can I convert CSV to JSON?

Yes. Switch to the JSON view and either copy it or download a .json file, with the header row used as the keys.

No. The file is parsed in the page by Papa Parse. A CSV export is often the most sensitive file a business owns - customers, salaries, orders - and none of it leaves your machine.

Stays on your computer How this works

Free CSV viewer and converter

Open, check and chart a CSV — entirely in your browser

Drag & drop a CSV here, or browse

CSV, TSV, semicolon or pipe separated, or a log file · any size, even several gigabytes · you can also paste rows · nothing is uploaded

About this CSV viewer

Open a CSV in a table you can sort by clicking any column heading, see any rows the parser could not read cleanly — the wrong number of fields is the usual culprit — chart a column, and export the result as tidied CSV, as JSON, or as a PNG of the chart.

Spreadsheet software struggles with very large exports and often mangles leading zeros and dates on the way in. Reading the file here avoids both. Excel stops at 1,048,576 rows; a file larger than 40 MB opens here in a large-file view that reads it a piece at a time straight from your disk, so a CSV or a log file of several gigabytes and hundreds of millions of rows opens in seconds, and you can scroll to any row, filter every row, summarise each column and save the rows you want. It is free to use, with no account and no upload limit, and it runs entirely in your browser: nothing is uploaded, and while the page stays open, a feature you have already used keeps working offline.

What you can do

Open CSV, TSV and plain .txt tables, with comma, semicolon, tab or pipe separators detected automatically
Paste rows straight into the page instead of opening a file
Sort by any column, with numbers in number order and blank cells last
See every row that does not match the header, listed by the row number the table shows
Convert CSV to JSON, with the header row as the keys, to copy or download
Chart a column as a bar, line, pie or doughnut chart and download it as a PNG; the chart plots the first 300 rows
Download a tidied CSV with comma separators, so a TSV or semicolon file becomes a standard CSV
Leading zeros and dates kept exactly as written, not reformatted the way a spreadsheet does
Open CSV and log files of several gigabytes: scroll to any row, filter with contains, exact, number or regular-expression matches, summarise every column, and save the matching rows as CSV or JSON Lines
Read log files a line at a time and filter them, for example to every ERROR or WARN line

How to open and check a CSV

Drop in a .csv or .tsv file. It opens in a table straight away; click a column heading to sort by it, again to reverse, and a third time to go back to the file's order.
Look at any rows flagged as malformed - the wrong number of fields is almost always the cause.
Switch to Chart to plot a column as a bar, line, pie or doughnut chart.
Export what you need: a tidied .csv, the data as .json, or the chart as a .png.

Questions

Why does Excel mangle my CSV when this does not?: Spreadsheet software guesses types on the way in, which is what turns a leading-zero product code into a number and a date column into something else entirely. This reads the file as text and shows you what is actually in it.
What counts as a malformed row?: A row with more or fewer fields than the header - usually an unescaped comma or an unbalanced quote somewhere upstream. They are kept in the table rather than silently dropped, and flagged above it by row number so you can see what needs fixing. The total is always shown; if there are more than 20, only the first 20 are listed.
Can it open a very large export?: Yes, of any size. A file over 40 MB is never read into memory whole: the page finds where its rows start in one quick pass, then reads only the part you are looking at. Filtering, column summaries and exports each read the file through once more, a piece at a time, so a multi-gigabyte export takes as long as your disk takes to read it. Sorting and charts need the whole file in memory, so they are for files under 40 MB.
Can I convert CSV to JSON?: Yes. Switch to the JSON view and either copy it or download a .json file, with the header row used as the keys.
Is my data uploaded?: No. The file is parsed in the page by Papa Parse. A CSV export is often the most sensitive file a business owns - customers, salaries, orders - and none of it leaves your machine.