From: Elliotte Harold <elharo@me...> - 2004-04-25 18:07:24
There's a case that isn't really dealt with in the SAX spec I'd love to
clear up if we could. Consider this document:
<!DOCTYPE root [
<!ELEMENT root (a)>
<!ELEMENT a EMPTY>
That was white space but this isn't!
This document is well-formed and invalid. Should the white space
preceding the non space PCDATA be reported to the ignorableWhiteSpace()
method or not? I can find nothing in the SAX specification or XML
specification that is clearly on point here. To make the question
tougher, assume there's 128 or so of white space before the first
non-whitespace charcter. That is, imagine the parser's buffer may fill
and one method or the other need to be called before it's known for
certain that all the content is ignorable. What should be done here?
Elliotte Rusty Harold