No. It is read and converted in your browser, which is why there is no size limit beyond your own memory and no copy anywhere else.

What happens to leading zeros?

They are kept. A field of digits is treated as text unless you ask otherwise, because postcodes, phone numbers and padded IDs are strings that happen to look like numbers.

Does it handle quoted fields and embedded commas?

Yes, including doubled quotes inside a quoted field and newlines within a field. Those are the cases that misalign columns partway down a file when a parser splits on lines first.

What about semicolon-separated files?

Choose the delimiter. Exports from spreadsheet software in locales that use a comma as the decimal separator are semicolon-separated, which is why a European CSV so often opens as one column.

A CSV converter that does not turn your IDs into numbers

CSV has no types. Every field in the file is text, and every converter has to decide whether a column of digits is a number or a string - a decision that silently destroys data when it goes the wrong way, in a file that still looks perfectly fine afterwards.

Open the CSV Converter →

ConvertCSV and Softland, side by side

	ConvertCSV	Softland
Leading zeros	Type inference is configurable	Preserved - a field of digits stays text unless you say otherwise
Quoted fields and embedded commas	Handled	Handled, including quotes and newlines inside a field
Semicolon and tab delimiters	Selectable	Selectable
JSON back to CSV	Yes	Yes
Where the file is read	On the server	In your browser - the file is not transmitted

The leading zero problem eats real data

A German postcode, a UK phone number, a US social security number, a bank sort code, a product SKU, an ISBN, a Stripe-style padded reference - all of these are strings of digits that are not numbers, and all of them begin with a zero often enough to matter. Convert one to an integer and 01234 becomes 1234, which is not a typo you can find later by looking at the file.

The same class of damage takes other forms. A long numeric ID crosses the point where a floating-point number stops representing integers exactly and the last digits change. A value like 1-2 becomes a date in some pipelines. A field containing only E and digits is read as scientific notation. Every one of these is a converter being helpful, and every one is silent.

Quoting is the other half of the specification

A field containing a comma must be quoted, a field containing a quote must escape it by doubling it, and a field can legally contain a newline as long as it is quoted. That last one breaks any parser written as split the file by lines, then split each line by commas - which is most parsers written in a hurry, and the reason an address column so often shifts every subsequent field one place to the left.

If a converted file comes out with the right number of rows but the columns misaligned partway down, that is nearly always an embedded newline or an unescaped quote. Proper parsing handles it; a regular expression does not, however tempting it looks at the time.

A CSV export is usually a copy of a real database

People convert CSVs because a spreadsheet export needs to go into an API, or an API response needs to go to someone who works in Excel. What is in the file is therefore customer records, order history, employee data, mailing lists - the contents of a production table, handed to whichever converter appeared first in the results.

Reading the file in the browser means it is never transmitted. There is no upload to wait for, no size ceiling other than your own memory, and no question about how long a copy of your customer list sits on somebody else disk. For a file of that kind the retention policy you want is the one where no copy was ever made.

When ConvertCSV is the better choice

You need a format this does not cover - fixed-width, SQL inserts, XML, Markdown tables - which is where a broad converter earns its place.
You want fine-grained control over type inference per column rather than a sensible default.
You are converting files large enough that a server should do the work instead of your laptop.

Frequently asked questions

Is my file uploaded?: No. It is read and converted in your browser, which is why there is no size limit beyond your own memory and no copy anywhere else.
What happens to leading zeros?: They are kept. A field of digits is treated as text unless you ask otherwise, because postcodes, phone numbers and padded IDs are strings that happen to look like numbers.
Does it handle quoted fields and embedded commas?: Yes, including doubled quotes inside a quoted field and newlines within a field. Those are the cases that misalign columns partway down a file when a parser splits on lines first.
What about semicolon-separated files?: Choose the delimiter. Exports from spreadsheet software in locales that use a comma as the decimal separator are semicolon-separated, which is why a European CSV so often opens as one column.

Try it yourself

CSV to JSON and back, with header detection.

Open the CSV Converter

ConvertCSV is a trademark of its respective owner. Softland is not affiliated with, endorsed by or sponsored by ConvertCSV. This comparison reflects how each product works rather than what either costs, because pricing and plan limits change; check ConvertCSV’s own site for its current terms.