CSV to JSON Converter

An amount displayed as ₹1,23,456.78 is stored as the JSON number 123456.78, with "currency": "INR" recorded separately. Grouping commas and the rupee symbol are display text, not JSON numeric syntax. Formatting does not convert lakh/crore values automatically.

Indian digit grouping such as 1,23,456.78 belongs to display text. Machine-readable JSON uses 123456.78 without grouping commas. Record INR separately and check the expected syntax when moving data between systems.

Yes, completely free with no registration. All parsing happens locally in your browser.

Yes, as long as the field is quoted per RFC 4180. The parser correctly handles embedded commas, doubled quotes (""), and multi-line quoted fields - things that break naive split() approaches.

Yes. Turn off the "First row is the header" option and every row - including the first - is emitted as a plain array of strings.

Values that parse as finite numbers become JSON numbers, the exact strings true and false become booleans, and empty cells become null. Any other value stays a string. Disable the option to keep every value as a string.

Yes. Leave the delimiter on auto-detect (it counts candidate delimiters in the first line) or select semicolon or tab manually.

Each column is scanned: all-numeric → number, true/false → boolean, empty string → null, everything else → string. Type inference is per-column, not per-cell, so a column stays homogeneous - "007" in an ID column of strings remains a string, avoiding the classic spreadsheet leading-zero loss.

Yes - headerless mode names columns c1, c2, c3. This covers machine-generated exports where the schema lives elsewhere, and legacy files whose header row is actually the first data record.

RFC 4180 quoting is honored: a quoted field may contain commas, newlines, and doubled quotes, and the parser keeps them intact in the JSON string. Malformed rows (unbalanced quotes) are reported with line numbers instead of silently truncating the field.