In reponse to Andy Heninger's Welcome message, I'd like to list a few things that I would like to see incorporated into future versions of Xerces-C and Xerces-J. I haven't gotten a chance to review the changes since XML4C v2, so these may be more achievable with the modifications since then.
1. Ability to validate against arbitrary schema/DTD It would be useful if an application can state that an XML document should be validated against an arbitrary DTD or schema in addition to any DTD or schema declared in the document. This would allow an application to use an application specific schema to filter out documents that use features it doesn't support or inappropriate documents. 2. Support for a type-aware DOM For applications that deal with a large amount of numeric data, it should reduce required memory if a "type-aware" DOM mode was supplied where DOM nodes that were numeric could be stored in the DOM tree as numeric types instead of text strings. The validator would probably be involved (through virtual functions) in distinguishing what type is appropriate for a value to anticipate datatype support in XML schema. In the interim a user should be able to provide a descriminator that could do something like determine to store anything that successfully converts to a double as a double.
