CSV to JSON Online
Convert CSV ↔ JSON privately, in your browser. Custom delimiter, header toggle, type coercion.
How to convert CSV to JSON online
Paste your CSV into the Input box. JSON appears on the right instantly — an array of objects, one per CSV row, with the first row used as keys. Toggle the Direction switch to go the other way: JSON → CSV produces one row per object with a column for every unique key. The parser implements RFC 4180 quoting, so embedded commas, newlines, and doubled quotes work correctly.
Why a local CSV/JSON converter matters
Most CSVs people convert online are exports — customer lists, financial transactions, product inventories, analytics dumps. Web converters that POST your payload to a server log everything they receive. ToolChop runs the parser in your browser, so the data never leaves your device. We could not see it if we wanted to.
What you can do
- CSV → JSON with type coercion (numbers, booleans, null)
- JSON → CSV with auto-collected columns from any object in the array
- Comma, semicolon, tab, or pipe delimiters
- Toggle header row on or off (auto-numbered column names if off)
- Pretty-print or minify the JSON output
- Copy to clipboard or download as a file
Frequently asked questions
How do I convert CSV to JSON online for free?
Paste your CSV into the Input box on the left. JSON appears on the right instantly. Click Copy or Download. The conversion runs entirely in your browser — no account, no upload, no daily limit.
Does ToolChop upload my data?
No. The parser runs in your browser via JavaScript. Your data never leaves your device — there is no upload, no temporary storage, and no copy on our servers. That matters when the CSV contains customer lists, exports, PII, or financial data.
What delimiters are supported?
Comma, semicolon, tab, and pipe. Use Semicolon for European-locale CSVs (where the decimal separator is a comma). Use Tab for TSV files. Use Pipe for some internal data-warehouse exports.
Does the parser handle quoted fields and escaped quotes?
Yes. The parser implements RFC 4180 quoting: double-quoted fields, doubled quotes inside ("") as an escape, and embedded newlines and delimiters inside quotes. So a name like `O"Brien, Mary` parses correctly.
What does the Coerce types option do?
When enabled, the CSV → JSON converter looks at each value and converts plain numbers to numbers, the strings `true`/`false` to booleans, and `null` to JSON null. Strings stay strings. Turn it off if you need every value as a string (useful when downstream code expects exact source bytes, like phone numbers or IDs with leading zeros).
Why is a number coming through as a string?
Most likely because Coerce types is off, or the value contains a non-digit character (commas as thousands separators, currency symbols, leading or trailing whitespace). The parser only coerces clean numeric strings — `1234.56` becomes a number; `1,234.56` stays a string.
What happens if my CSV has no header row?
Uncheck `First row is header`. The output uses generated keys `col_1`, `col_2`, … instead of named keys.
Can I convert JSON to CSV?
Yes — toggle the Direction switch to JSON → CSV. The tool accepts an array of objects and produces a CSV with one row per object. Keys from any object contribute to the column set; missing keys become empty cells.
Does JSON → CSV handle nested objects or arrays?
Top-level scalars in each object map to CSV cells directly. Nested objects and arrays are serialized to their JSON string representation inside the cell so no data is lost. If you need flattened nested JSON (e.g., `user.email` as a column), pre-process with a tool like jq.
Is there a row limit?
Only your browser's memory. We have parsed 1 million-row CSVs (~100 MB) in modern Chrome. Above that, the textarea UI itself starts to lag — for very large data, run the same logic in a Node script or use a streaming parser like csv-parse.
How is this different from a server-based CSV converter?
Privacy. Server-based converters log your full payload — customer data, internal IDs, financial records. ToolChop never sees the data because it runs in your browser. That is the only durable difference for sensitive workloads.
Can I upload a CSV file instead of pasting?
Right now the tool is paste-only — it covers the common case and avoids any drag-and-drop ambiguity. File-upload (still client-side) is on the roadmap; for now, open the CSV in any text editor and copy its contents.
Why use ToolChop instead of running a script?
Convenience for one-off conversions. For a single ad-hoc job, pasting into a browser is faster than writing or reaching for a script. For repeated or large-scale work, a CLI like csvkit or jq is more appropriate.