XML to TONL Converter

Example: XML input and real TONL output

The output below was generated by the same conversion code the converter uses, run on this sample when the page was built.

<catalog>
  <book id="bk101">
    <title>Designing Data-Intensive Applications</title>
    <author>Martin Kleppmann</author>
    <year>2017</year>
  </book>
  <book id="bk102">
    <title>The Pragmatic Programmer</title>
    <author>Andrew Hunt</author>
    <year>1999</year>
  </book>
</catalog>

#version 1.0
root:
  catalog:
    book[2]{@_id,author,title,year}:
      bk101,Martin Kleppmann,Designing Data-Intensive Applications,2017
      bk102,Andrew Hunt,The Pragmatic Programmer,1999

TONL output is 35% smaller than the XML input. The same data is 214 characters as minified JSON and 341 as pretty-printed JSON. Sizes are character counts, not tokens: token counts depend on the model's tokenizer (see the tokenizer comparison).

XML to TONL: what changes

• Reading XML: Attributes are kept with an @_ prefix (id="bk101" becomes "@_id"), repeated child elements become arrays, and numeric text becomes numbers. A single child element becomes an object, not a one-item array, so the shape depends on how many elements appear.
• Writing TONL: TONL output starts with #version 1.0 and a root key, lists field names in braces, and sorts keys alphabetically. On small payloads the header and per-object field lists make it larger than TOON; TONL earns its place through its query, schema, and streaming tooling.

Why Convert XML to TONL?

Use TONL when you want a compact text format plus tooling around it: querying, schema validation, or streaming. For plain prompt compression, TOON is usually the simpler choice.