Data Format Converter

The converter reads and writes JSON, JSON Lines, YAML, TOML, XML, CSV, TSV, INI, Java properties, Excel XLSX and OpenDocument ODS, and any of them can be converted into any other. Pick the source format in From and the target in To. For JSON and YAML, CSV to JSON, JSON to CSV, JSON to XML and XML to JSON the site also has dedicated tools with extra options.

Choose XLSX or ODS as the source format and upload the workbook instead of pasting text. The converter reads the first sheet, or the sheet you name in the Sheet name field, treats its first row as column headers and turns every following row into a record. Formulas are read as their last calculated values and date cells become ISO 8601 strings.

These formats cannot hold nested structures, so nested objects are flattened into dotted keys and array items get their index in brackets. For example {"user": {"name": "Ada", "tags": ["x"]}} becomes the keys user.name and user.tags[0]. Reading properties back nests the dotted keys again when they do not conflict, while CSV and TSV headers are kept exactly as written.

When the target is CSV, TSV, XLSX or ODS, a top-level array becomes the rows. An object that only wraps one array, such as {"users": [...]} or an XML document of repeated row elements, is unwrapped to that array, and anything else becomes a single row. The columns are the union of all keys in the order they first appear, and missing values stay empty.

JSON, JSON Lines, YAML and TOML keep their numbers, booleans and null values, and spreadsheet cells keep numbers and booleans. CSV, TSV, INI, properties and XML have no types, so their values are read as text and stay text in the result. TOML integers too large for a JavaScript number are kept as text so no digits are lost.

Each element becomes a key, repeated elements become arrays, attributes become keys starting with @ and text beside attributes is stored under #text. When writing XML, an object with a single top-level key uses that key as the root element, everything else is wrapped in a root element and array items are written as row elements. Characters that are not allowed in element names are replaced with underscores, and documents that declare entities are rejected for safety.

TOML has no null value, so keys holding null are left out and null items inside arrays become empty strings. TOML also needs a table at the top level, so a top-level array is written as an array of tables named rows and a single value as a key named value. Dates read from TOML are written to other formats as ISO 8601 strings.

INI sections become objects, key[] lines collect values into a list, lines starting with ; or # are comments and double-quoted values may use escapes such as \n. Properties follow the Java rules: =, : or whitespace separate key and value, lines starting with # or an exclamation mark are comments, a trailing backslash continues the line and \uXXXX escapes are decoded. Written properties are plain UTF-8, which Java 9 and later read in resource bundles.

JSON Lines input is read line by line, blank lines are skipped and the values form an array, while JSON Lines output writes each array item as one compact JSON value per line. YAML input with several documents separated by --- becomes an array with one entry per document. Invalid YAML, TOML and XML report the line and column where parsing stopped, and invalid JSON Lines report the line.

Text and files can be up to 5 MiB, with at most 100,000 rows, 1,000,000 cells, 64 levels of nesting, 50 MiB of unpacked workbook content and 16 MiB of result. Larger inputs are rejected with a message instead of being cut off. Your data is converted in memory for the request and is not stored.