Why another XML parsing library?

Some may have noted that in the December 2012 release of my MinGW Distribution I have included yet another XML parsing library: pugixml. Why this if it already has libxml2?

Well, while libxml2 is a complete, standard-compliant XML parsing library it lacks a DOM interface. With DOM it is easier to handle and to walk through the XML representation.

I have evaluated many simple to use, fast libraries: RapidXML, FastXML, AsmXML. And, of course pugixml. However pugixml is fast, has a good documentation and a DOM-like, modern interface that fits well with STL and C++11 ranged-for and lambdas. That was I was looking for.