What this tool does
Prepare a small data file for a service that expects different field names. Select only the columns you need, check missing values and choose explicit types before exporting the full result.
How to use it
- Paste data or import one UTF-8 file. Select CSV or a JSON object array and confirm the CSV delimiter.
- Inspect columns, then select fields, output names, types and whether missing values may become null.
- Apply the mapping. Review up to 20 records, then copy or download the complete JSON or CSV file.
Formula and method
CSV uses its first record as unique, non-empty column names and supports quoted commas, line breaks and doubled quotes. JSON must be an array of flat objects. Strings preserve whitespace. Numbers must be finite, within the safe range and free of currency symbols or grouping commas. Booleans accept only true/false. Allowing missing values converts empty strings, absent keys and null into null.
Worked example
For fictional stock data, rename quantity to count and choose Number to turn "12" into 12. Keep identifiers such as "0012" as String. A blank quantity causes an error unless missing values are explicitly allowed. No prices or quantities are inferred.
Useful for
- Match field names between services
- Check CSV types and missing values
- Export only selected columns
Limits and notes
Input: UTF-8, up to 2 MB, 1–20,000 data records and 100 columns. Each JSON/CSV output is limited to 8 MB. CSV requires unique, non-empty headers. Nested JSON, Excel workbooks and other encodings are unsupported. Standard JSON parsing does not preserve numeric spelling, key order or duplicate keys. Quote long IDs as strings in the original input. Converting to Number can round decimal values. CSV apostrophe protection can turn negative numbers into text and does not guarantee safety in every spreadsheet or after saving again.
Frequently asked questions
Is my file uploaded or stored?
Source and results are processed in this page’s memory. They are not uploaded or saved in browser storage. Download the result before reloading or leaving the page.
Are different kinds of missing values preserved?
No. If allowed, empty strings, absent keys and null become JSON null and empty CSV cells. Keep the original file if the distinction matters.
Why does the preview show fewer records?
Source previews show five records and result previews show up to twenty. Copy and download use every mapped record, not the preview table.