Paste CSV or JSON in either pane — the other side updates instantly. RFC 4180-compliant, type-aware, nested-object flatten / unflatten, Excel-friendly. Runs entirely in your browser.
Paste CSV or JSON in either pane — the other side updates instantly. RFC 4180-compliant, type-aware, nested-object flatten / unflatten, Excel-friendly. Runs entirely in your browser.
Paste your CSV into the left pane and JSON appears on the right instantly. By default the first row becomes object keys, types are inferred (numbers / booleans / null), and dotted-path columns like address.city rebuild as nested objects. Toggle the options bar above to change any of those defaults.
Paste a JSON array of objects into the right pane and CSV appears on the left. Nested objects flatten to dotted-path columns ({user: {name: "…"}} → column user.name); arrays inside values become indexed columns (tags.0, tags.1). The output is RFC 4180-compliant — fields containing commas, quotes, or newlines are quoted automatically.
Customer lists, sales reports, employee data — sensitive datasets never leave your device. Open DevTools → Network and you'll see zero outbound requests during conversion.
RFC-compliant parsing, type inference, nested-object support, Excel-aware — every detail that breaks lesser tools.
Paste CSV in the left pane or JSON in the right — the other side updates as you type. Edit either side and the other re-renders within 300 ms.
Hand-rolled state-machine parser handles every edge case: quoted commas ("Hello, world"), embedded double-quotes (""), multi-line cells inside quotes, and Windows / Unix line endings.
Auto-detects comma, semicolon, tab, or pipe by scoring row-count consistency in the first 10 KB. Manual override available for European CSVs (semicolons) or TSV exports.
"42" becomes 42, "true" becomes true, empty becomes null. Dotted-path columns like address.city rebuild as nested JSON. Toggle off for strict-string mode.
Drag a .xlsx file onto the page — SheetJS reads it locally, no upload. Export back to CSV or JSON. Supports .csv, .tsv, .txt, .json, .xlsx, .xls.
Every byte of your data stays on your machine. The parser, type inferer, and download all run in JavaScript locally — open DevTools → Network and verify zero outbound requests.
Four ways to get clean data out the other side.
Paste CSV / TSV in the left pane, JSON in the right. Or drag a .csv, .tsv, .json, or .xlsx file onto the drop zone. The tool detects the format automatically and runs the conversion.
Set the delimiter (or leave on auto-detect), toggle whether the first row is a header, choose whether to infer numbers / booleans / null, and decide if dotted-path columns should rebuild as nested JSON objects. Each toggle re-runs the conversion immediately.
Both panes are fully editable. Tweak a value in JSON — CSV updates within 300 ms. Fix a column header in CSV — JSON updates the same way. Use the swap button to flip the panes.
Click the copy icon to send either side to the clipboard. Click the download icon to save as data.csv or data.json with proper UTF-8 BOM (so Excel opens the CSV correctly). For batch work, paste the next dataset and the tool re-runs.
Four common workflows where in-browser CSV ↔ JSON beats the alternatives.
Backend gives you a JSON array — paste it, get a CSV the product team can open in Excel. They edit, give you back a CSV — paste it, get JSON your API can ingest. No CLI, no Python script, no upload to a third-party service.
CRM / analytics tools love to export semicolon-separated CSVs with quoted multi-line cells. The auto-detect handles the delimiter; the RFC 4180 parser handles the quotes correctly (most regex-based tools mangle them). Re-export as JSON or clean CSV.
A REST API returns {user: {address: {city: "…"}}} per row. Paste it, get a flat CSV with user.address.city as a column header. Drop the CSV in Excel for a non-technical stakeholder and they can pivot, filter, and chart without ever seeing JSON.
You have a CSV of users / products / orders. Convert to JSON, drop into a Laravel / Django / Rails seeder, or paste straight into a NoSQL bulk-insert. The type inference saves the work of casting "42" back to 42 in your seed script.
Customer lists, sales reports, employee data — sensitive datasets never leave your device. Open DevTools → Network and you'll see zero outbound requests during conversion.
.xlsx files are parsed locally via SheetJS, which is loaded only when you drop an Excel file (zero network cost otherwise).
Hand-picked tutorials on CSV, JSON, and data conversion.
Quoted commas, embedded newlines, BOM headers, doubled quotes — the cases where a regex split breaks down and you need a real parser.
When CSV's row-orientation beats JSON's structure — and when nested objects make CSV a dead end. Worked examples, with conversion patterns for both directions.
Paste your CSV into the left pane and JSON appears on the right instantly. By default the first row becomes object keys, types are inferred (numbers / booleans / null), and dotted-path columns like address.city rebuild as nested objects. Toggle the options bar above to change any of those defaults.
Paste a JSON array of objects into the right pane and CSV appears on the left. Nested objects flatten to dotted-path columns ({user: {name: "…"}} → column user.name); arrays inside values become indexed columns (tags.0, tags.1). The output is RFC 4180-compliant — fields containing commas, quotes, or newlines are quoted automatically.
Yes. Auto-detect picks the most likely delimiter (comma / semicolon / tab / pipe) by scoring row-count consistency in the first 10 KB. You can also manually pick the delimiter from the dropdown — useful when your CSV has both commas and semicolons in the data and the heuristic guesses wrong.
Yes. Drag a .xlsx or .xls file onto the drop zone (or click "Choose file") and the first sheet is converted to CSV in your browser using SheetJS. The library is loaded only when you actually drop an Excel file — zero cost on cold load. After the conversion you can edit, copy, or download as JSON or CSV.
No. The CSV parser, JSON parser, type inference, nested-object flatten / unflatten, and file download all run as JavaScript in your browser. Even Excel files are parsed locally with SheetJS. Open DevTools → Network and you'll see zero outbound requests during conversion. Sensitive data — emails, sales reports, customer lists — never leaves your device.