Data converter

CSV/TSV — the delimiter is detected (comma, semicolon, tab, or pipe); quoted fields, embedded newlines and doubled quotes are handled. Numbers and booleans are detected (values with leading zeros stay text). Columns named a.b become nested objects, and nested objects are flattened back to dotted columns on the way out.

XML — attributes become @name keys (the prefix is configurable), repeated elements become arrays, and mixed text becomes #text.

Query strings — bracket notation (filter[size][]=small) becomes nested objects and arrays; a full URL can be pasted.

TOML — tables, arrays of tables, inline tables, dotted keys, and all string forms. Dates are kept as strings.

Code generation — types are inferred from every element of every array, so fields that are missing somewhere become optional and null makes a field nullable. Strings that look like dates, e-mail addresses, URIs, or UUIDs get a JSON Schema format. More type options… sends the input, as JSON, to JSON → types — Zod, Java, Kotlin and protobuf too, enums, tagged unions and naming settings.