>>>>> "Colin" == Colin Paul Adams <[EMAIL PROTECTED]> writes:
>>>>> "David" == David N Bertoni/Cambridge/IBM <[EMAIL PROTECTED]> writes:
David> However, we _can_ build Xerces DOMDocument instances as the
David> output of a transformation. In fact, I just checked in a
David> new sample and some support classes for that very purpose.
David> If you pull down the latest CVS, you'll get it. It's
David> called TransformToXercesDOM.
Colin> Thanks. This method works for me satisfactorily.
I understand I can also write to a DOMDocumentFragment (using a
slightly different constructor). But despite looking at the source
code:
FormatterToXercesDOM::FormatterToXercesDOM(
DOMDocument_Type* doc,
DOMDocumentFragmentType* docFrag,
DOMElementType* currentElement)
:
FormatterListener(OUTPUT_METHOD_DOM),
m_doc(doc),
m_docFrag(docFrag),
m_currentElem(currentElement),
m_elemStack(),
m_buffer(),
m_textBuffer()
{
assert(m_doc != 0 && m_docFrag != 0);
}
I can't fathom the semantics. It seems clear to me that docFrag will
be the document fragment to write to, and I assume I should set
currentElement to 0 as before. But what is the significance of the doc
parameter?
--
Colin Paul Adams
Preston Lancashire
