Hi,
On Wed, May 21, 2014 at 11:15 AM, Stefan Seefeld
I have always strongly argued against the idea that an "XML API" was only about parsing XML data, as there are many useful features that involve manipulation of XML data (including transformations between documents, xpath-based search, etc.). I have also argued that re-implementing all that functionality from scratch is foolish with so many existing implementations, so any boost.xml project should focus on wrapping such implementations, rather than reinventing them. In fact, I believe such an API should be robust enough to be able to wrap different backends, rather than depending on a particular implementation choice.
I agree with that sentiment completely. Different implementations have different strengths and weaknesses. I generally choose pugixml, but I understand Xerces is also very popular. Boost.Multiprecision is a good example of this kind of library, but, like Boost.Multiprecision, I'd say there should be a "default" implementation included with the Boost.XML library in the case where the user doesn't want to add another dependency besides Boost. I guess there are also wrapper type libraries like Boost.MPI, which don't really lend themselves to a default implementation, but I'd argue that XML is standardized well enough that a default implementation is appropriate. Just thought I'd throw in my 0.02. -g