[
Lists Home |
Date Index |
Thread Index
]
james.anderson@setf.de (james anderson) writes:
>relative to the size of the parser, that is not an excessive burden.
It depends on the kind of parser you're writing.
If you're going to report an element and all of its attributes to an
application at once anyway, there's no really additional cost. That's
clearly the majority case at the moment. SAX2, DOM, etc.
If you're going to report an element and its attributes and attribute
values as separate events, it means jumping back and forth. In the work
I'm doing at the moment, that imposes genuinely annoying costs. The
bindings themselves aren't so difficult - it's just a jumble of
operations whose sequence is very different from the sequence of
characters in the document.
Still solvable, sure. Just extra work that leaves me alternately
laughing and crying, as mentioned before.
>"namespaces in xml" has a very clear semantics. having to do with the
>universal name denoted by a given (prefix x local-part) pair.
I guess I expect semantics to have something to do with signifieds
rather than the creation of signifiers. To me universal names are
labels, not meaning.
--
Simon St.Laurent
Ring around the content, a pocket full of brackets
Errors, errors, all fall down!
http://simonstl.com -- http://monasticxml.org
|
