CSV ↔ JSON Converter

How to use

1. Choose CSV → JSON for a spreadsheet export, or JSON → CSV for a JSON object or array of objects.
2. Paste text or open a local CSV, TSV, or JSON file; the result changes immediately.
3. For CSV input, confirm the detected delimiter and choose whether the first row contains column headers.
4. Use type inference only when values such as 007 are not identifiers that must remain text.
5. Inspect the first converted rows in Table preview, then copy, download, or send the result back to the input for the reverse conversion.

CSV quoting is not the same as splitting at commas

RFC 4180 describes the familiar CSV convention: commas separate fields, but a field that contains a comma, a line break, or a double quote can be wrapped in double quotes. A literal double quote inside that wrapper is written twice. For example, `"Seoul, KR"` is one field, and `"She said ""yes"""` contains one pair of quote marks. This converter reads those quoted commas, quotes, and line breaks instead of using a simple comma split.

It also recognizes CRLF, LF, and CR line endings, and removes a leading UTF-8 BOM often added by spreadsheet exports. A quote in the middle of an unquoted field, an unclosed quote, or a row with a different number of cells is an error here: reporting it is safer than shifting a value into the wrong column without notice.

Headers, identifiers, and type inference

With “First row is headers” on, `id,name` becomes the property names for every JSON object below it. Turn it off when every row is data; the JSON result will be an array of arrays. Duplicate headers are rejected because JSON object keys must be unique for a reliable conversion.

CSV has no native number, boolean, or null type. Type inference is therefore off by default: `007`, `00042`, and a large integer remain strings exactly as exported. When you turn it on, unambiguous values such as `42`, `-3.5`, `true`, `false`, and `null` become JSON values; integer values outside JavaScript’s safe-integer range remain text so their digits are not rounded.

What JSON can fit in a CSV cell

JSON can nest objects and arrays; CSV is a flat row-and-column format. For JSON → CSV, this tool accepts one object or an array of objects, gathers every top-level key in first-seen order, and creates one row per object. A missing key becomes an empty cell. A nested object or array is preserved as compact JSON text in one cell rather than silently discarding it or inventing columns.

The CSV result uses CRLF line endings and quotes any cell containing the selected delimiter, a double quote, or a line break. The table preview shows the first 10 converted rows, so a later-only key, a blank field, or a quoted value is visible before you send the result to a spreadsheet or API.

Frequently asked questions