Lists Home |
Date Index |
- From: Marcus Carr <email@example.com>
- To: XML-Dev Mailing list <firstname.lastname@example.org>
- Date: Thu, 01 Apr 1999 15:31:09 +1000
Simon St Laurent moved this discussion to this list from XSL, so I followed.
Didier PH Martin wrote:
> you are right it parse the DTD from a syntactic point of view but do not
> enforce the structural integrity of the document. It is faster to just parse
> the DTD syntax than to enforce structural integrity.
To clarify, it's faster to process the DTD but not the instance than it is to
process both, but it may be a marginal difference. Imagine using the DocBook DTD
for the following instance:
<!DOCTYPE para PUBLIC "-//whatever the DocBook//DTD goes by//EN">
Surely validation of the instance is a fairly minor issue, after the processor
has had to plow through a large DTD? <InBigFatLetters>I'm not suggesting that
XML documents should be validated client side</InBigFatLetters>, I'm questioning
the wisdom of looking at the DTD at all if you're not putting it to any use.
Surely the overhead of looking at the DTD outweighs the benefit (none) obtained
in rendering the above example?
> It takes the principle
> that it will try to render the document even if a structural error is
> present. So, rendition takes over integrity of the structure. This is
> because the browser main purpose is to render.
No it doesn't, it determines that it will try to render the document regardless
of whether it contains errors. I agree that for a browser, rendition should take
precedence over structure.
> However, when the same parser is used in a different context, structural
> integrity may becomes a main constraint.
Agreed - that's when I'd have the processor examine the DTD.
Marcus Carr email: email@example.com
Allette Systems (Australia) www: http://www.allette.com.au
"Everything should be made as simple as possible, but not simpler."
xml-dev: A list for W3C XML Developers. To post, mailto:firstname.lastname@example.org
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To (un)subscribe, mailto:email@example.com the following message;
To subscribe to the digests, mailto:firstname.lastname@example.org the following message;
List coordinator, Henry Rzepa (mailto:email@example.com)