The conversion everyone needs eventually
CSV is what non-technical systems produce: exports from spreadsheets, CRMs, billing platforms and analytics dashboards. JSON is what applications consume. Almost every data-import task starts by moving between the two, and doing it by hand for anything past a few dozen rows is unrealistic.
The reverse direction matters just as much. When a colleague asks for data from an API, sending them a JSON array is unhelpful - they want something that opens in a spreadsheet. Converting to CSV turns a developer answer into a usable one.
Why CSV parsing is harder than splitting on commas
CSV has no formal standard, which is why so many home-grown parsers are subtly broken. A field containing a comma must be quoted. A quoted field containing a quote must escape it by doubling. A field can contain a line break, meaning one record legitimately spans several lines - and any parser that reads line by line corrupts that record silently.
Delimiters vary too. Much of Europe uses semicolons because the comma is the decimal separator, and tab-separated files are common in exports. Header detection is a further guess: the first row usually holds column names, but not always, and getting it wrong shifts every record by one.
Types survive one direction and not the other
CSV has no types. Every value is text, so a conversion to JSON has to decide whether 007 is a number or a string, whether TRUE is boolean, and what an empty cell means - null, empty string, or absent key. These guesses are where leading zeros vanish from postcodes and product codes.
Going from JSON to CSV loses structure instead. Nested objects and arrays have no natural representation in a flat grid, so they must be flattened into dotted column names or serialised into a cell. Neither round-trips perfectly, which is worth knowing before you treat CSV as a storage format.
Frequently asked questions
- Is my file uploaded?
- No. Conversion runs entirely in your browser, so a customer export or a financial report never leaves your machine.
- Does it handle quoted fields and embedded commas?
- Yes. Quoted fields, escaped quotes and line breaks inside a quoted field are all parsed correctly rather than being split naively on commas.
- Can it convert JSON back to CSV?
- Yes, both directions. Note that nested objects have to be flattened into columns, since a CSV grid has no way to represent nesting.
- What happens to leading zeros?
- Values that look numeric may be converted to numbers, dropping leading zeros. If you are converting postcodes or product codes, check the output rather than assuming they survived.
- What about semicolon-separated files?
- Semicolon and tab delimiters are common in European and export-generated files, and the delimiter is detected rather than assumed to be a comma.