Java: What is the difference between LSParser and DocumentBuilder

Question

I can't find any information on org.w3c.dom.ls.LSParser. I know it is an interface but there is an on only way to get a concrete object afaik.

DOMImplementationLS factory = (DOMImplementationLS) myXMLDocument.getImplementation();
LSParser parser = factory.createLSParser(DOMImplementationLS.MODE_ASYNCHRONOUS, null);

How is LSParser different from javax.xml.parsers.DocumentBuilder (or SAXParser)

Answer 1

First, a SAXParser is different from the javax.xml.parsers.DocumentBuilder and LSParser, in that it streams through the XML, instead of building a Document Object Model (DOM).

That leaves us with explaining javax.xml.parsers.DocumentBuilder and LSParser. The DOM spec has levels to define the structure and behavior of the DOM. There are three levels. (1, 2, and 3). LSParser stands for "Load and Save Parser". This Parser provides support for DOM Level 3. It implements the behavior defined here - https://www.w3.org/TR/DOM-Level-3-LS/load-save.html . Since LSParser supports the DOM Level 3 standard, it allows event handling (when DOM is loaded), saving and filtering. Note that this is a spec, so it can be implemented in any language.

In contrast, javax.xml.parsers.DocumentBuilder is a Java based API (JAXP). It is a project, not a spec.

LSParser spec has been influenced by both the JAXP and SAX projects. It just standardized the parsing of XML and using the DOM.

References: https://www.amazon.com/Processing-XML-documents-Oracle-JDeveloper/dp/1847196667 (Refer Chapter 7 - relevant to this topic).

https://www.w3.org/TR/DOM-Level-3-LS/load-save.html (this is the full spec).