CSV Converter

Turn a CSV into the format the next tool actually wants

A CSV is fine right up until something downstream refuses to read it. Your accountant wants Excel. The API wants JSON. Some older system only speaks XML, and the import script needs SQL. This converter takes one CSV and hands you back whichever of those you need, plus HTML, TSV, and PDF. Because you can clean the data up on the page before exporting, what comes out the other side is already sorted, filtered, and free of the duplicate rows you didn't want. No account, nothing to install, and nothing left sitting on your computer afterwards.

What actually happens when you convert

The CSV is read in your browser. When you drop a file in or paste rows into the box, the parsing runs in JavaScript on your own machine, not on a server somewhere. It even works out the delimiter for you: it looks at the header line and picks whichever separator turns up most, so comma, semicolon, tab, and pipe files all load without you flipping a single setting. Quoted fields with commas inside them are handled too, including the doubled-quote escaping that Excel writes into its own exports.

Where the data goes next depends on the format you choose. Ask for CSV back and the file is built in the browser and never leaves your device. Ask for Excel, JSON, XML, HTML, TSV, SQL, or PDF, and the cleaned table is passed to a short-lived server job that assembles the file and streams its progress back while it works. You watch a real progress bar instead of a frozen tab, which starts to matter once a file runs to several thousand rows. Most exports finish in a few seconds.

Only need JSON? The single-purpose CSV to JSON converter does exactly this with one fewer choice to make.

Clean the data before you convert

You don't have to export the CSV exactly as it came in. Once it loads, the whole table shows up in an editable grid: double-click any cell to fix a typo, click a column header to sort, or open the filter bar to narrow thousands of rows down to the handful you actually care about. There are buttons to add a column, rename one, drop one you don't need, and strip out duplicate rows in a single click. The view pages through 50 rows at a time so a big file stays responsive. Whatever the table looks like when you hit convert is exactly what lands in the exported file.

The formats you can export to

Each output exists for a particular job, and they don't all behave the same way:

• Excel (XLSX) opens straight in Excel, Google Sheets, or LibreOffice Calc, with your headers on the first row of a single sheet.
• JSON comes back as an array of objects, one per row, using your column names as the keys, so it drops into an API call or a config file without reshaping.
• XML wraps each row in a record with a named field per column, for the older systems that still ask for it.
• HTML is a full styled table you can paste into a page or an email.
• TSV swaps the commas for tabs, which is the right call when your data already has commas sitting inside the cells.
• SQL builds a CREATE TABLE plus one INSERT per row, ready to pipe into MySQL.
• PDF lays the table out across paginated pages, repeating the header on each one.

Where it struggles, honestly

A couple of things are worth knowing before you trust the result. The big one is the Excel export: every cell is written as text, so a column of numbers comes through text-formatted. SUM() will sit there returning zero until you convert them, which takes about ten seconds with Data, then Text to Columns, or by wrapping a cell in VALUE(). I'd rather say that plainly than have you find it during a reconciliation.

If Excel is the only format you're ever after, the dedicated CSV to Excel converter takes you straight there without the format menu.

A few smaller quirks are worth a glance:

• The upload limit is 5 MB. That's still tens of thousands of rows in a typical CSV, but if you're over it, split the file and convert the halves separately.
• In the SQL output, column names are lowercased with underscores and every column is typed TEXT. That's fine for a throwaway load and wrong for a real schema, so tighten the types yourself once it's imported.
• The PDF trims any cell longer than 100 characters so the table still fits the page. Wide free-text columns read better as Excel.
• In TSV, stray tabs or line breaks inside a cell are replaced with spaces so the columns don't drift out of line.

A real example

Say you've pulled a 3,000-row product list out of a European system and it came out semicolon-separated, with prices written like 1.299,00. Paste it in and the tool spots the semicolons on its own, lines the columns up, and shows you the table. Had it been comma-separated instead, nothing would change; the detection quietly does the right thing either way. You sort by price, drop a duplicate SKU or two, and export to Excel. The numbers land as text, so you highlight the price column, run Text to Columns once, and the totals start adding up. Start to finish, under a minute, and no retyping.

When a different tool fits better

Coming the other way, from a JSON file back to a spreadsheet, the JSON to CSV converter handles that direction.

If you turn spreadsheets into database imports often, the focused CSV to SQL converter is the quicker bookmark.

One more note, on privacy. When you export back to CSV, the whole thing happens in your browser and nothing is uploaded at any point. For the other formats, the table you see is sent to the server purely to build the file, and that generated file is deleted automatically within roughly ten to twenty-five minutes. Nothing is kept around, read, or passed to anyone else.