Convert CSV to JSON in your browser

When headers are enabled, the first nonblank record supplies the object keys and is not included as a data row. Headers must be nonempty after checking surrounding whitespace, and must be unique with case distinguished. Their original spaces are preserved rather than trimmed. With headers disabled, every record becomes an array of strings. Every row must have the same number of cells; a trailing delimiter represents an empty final cell.

A quoted field can contain the selected delimiter, line breaks and doubled double-quotes. For example, a header row name,postcode followed by a record with the quoted name Smith, Ada and postcode 00123 produces one object whose name includes the comma and whose postcode is the string 00123. A quote must start at the beginning of a field; whitespace or other characters after a closing quote are rejected. Comma follows the common RFC 4180 form; other delimiters are explicit extensions, not support for every CSV dialect.

A value such as 00123 keeps its leading zeros; dates, large identifiers, empty cells and formula-looking text remain strings too. The converter does not infer numbers, dates or booleans, and does not execute spreadsheet formulas. Header names such as __proto__ and constructor are serialized as ordinary data properties. JSON still needs appropriate handling in whatever program consumes it.

A physically empty line outside a quoted field is ignored. A whitespace-only line or a quoted empty field is a real record and must have the expected width. An initial UTF-8 byte-order mark is removed; embedded marks and quoted line endings are retained. The preview includes at most 50 data records; Copy and Download JSON include all records within the input, aggregate cell and UTF-8 output-byte limits. Both header cells and data cells count toward the displayed cell budget.