JavaFX ships with a PullParser class for processing XML and JSON documents node by node. This linear approach is cumbersome, typically demanding the maintaining of all manner of data outside of the event handling function to track the current position within the document as it is parsed.

Rather than a single event handling function, Arkles uses JavaFX Script's declarative syntax to create trees of event handling nodes that mirror the XML/JSON structure being processed. These event trees may be bound to parts of the input document using a subset of the XPath notation, determining when they will run.

The result is a much easier and more natural way of working with XML/JSON data (the current beta release of the project supports only XML).