Skip to content

Choose your path

Built around the job you need to finish

Export strict JSON rows into escaped CSV while preserving number lexemes and disclosing nested, null, missing and spreadsheet-formula mapping.

Precision-sensitive export analyst

Avoid rounding large identifiers before spreadsheet import.

Export an unsafe integer and exponent, inspect raw CSV, review destination type-inference warning.

Original number tokens survive; receiving spreadsheet behavior is not guaranteed.

API integration engineer

Map nested metadata and heterogeneous object columns.

Compare compact JSON cells/reject, null policies, union columns and escaped headers.

No object-to-string corruption or missing columns; representation loss is explained.

Mobile Project export reviewer

Choose formula handling and save reverse conversion independently.

Reject formula-like text, opt into prefix only deliberately, export and save three reverse fields.

Visible policy matches output, and existing CSV-to-JSON Project fields remain intact.

Was this tool helpful?

Reference & details

How it works

Token-preserving JSON input

A bounded strict parser retains number tokens instead of passing them through JavaScript Number. Duplicate decoded member names, malformed strings and mixed object/array row shapes are refused.

Visible cell mapping

Objects use first-seen union columns, row arrays pad to the widest row and scalar arrays use one column. Nested data becomes compact JSON text or is rejected; missing and null cells have disclosed policies.

CSV and spreadsheet boundaries

Headers and cells share delimiter/quote/CR/LF escaping. Formula-like text is preserved with a warning, prefixed only by choice, or refused. Quoting alone is not spreadsheet protection. Download BOM is separately opt-in.

Updated: August 2026

Example Scenarios

An analyst exports a large identifier without numerical rounding and verifies that the receiving spreadsheet imports that column as text.

An engineer exports heterogeneous object keys, encodes nested metadata as JSON text cells and checks missing versus null handling before import.

A reviewer rejects formula-like text, corrects the source and saves JSON-to-CSV-specific fields without replacing the earlier CSV-to-JSON conversion.

FAQ

The converter copies JSON number tokens, including unsafe integers, exponents and negative zero, without numeric re-parsing. The receiving spreadsheet can still round or reinterpret text; configure and test its import.

Use a uniform array of objects, row arrays or scalars, or a single object/scalar. Nested objects and arrays become compact JSON cell text unless rejected. Mixed row shapes are refused; object columns follow first-seen keys.

No. Quoting is CSV syntax, not formula neutralization. Choose preserve with warnings, reject, or explicit apostrophe prefixing, which changes values. Behavior varies by spreadsheet and re-saving may change handling.

Only CSV output, without report headings. Choose delimiter, CRLF/LF and minimal/all quoting. A UTF-8 BOM is added only to the downloaded file if checked; copy and preview remain unchanged. The page reports a download request, not proof of an OS save.

CSV cannot preserve every JSON type distinction. Source is not automatically stored; an explicit Project save uses its own three local fields. Limits include 5,000 rows, 200 columns, 100,000 cells, 64 JSON levels, one/two million input/output characters and a ten-second timeout.

About JSON to CSV

Decide how JSON values become cells before exporting. Large number spelling is preserved, but spreadsheet import can still infer types or round values; review the consuming application.