external parsed entites (was: A unique ID question ?)

[David Brownell]

Lauren Wood wrote:
> 
> On 10 Nov 99, at 14:48, David Brownell wrote:
> 
> > Having just been suffering through a DOM implementation
> > that actually exposes EntityReference nodes, with no way
> > to turn them off and no utilities to get rid of them,
> > I'd quite gladly abolish them from DOM ... ;-)
> 
> Well, you can always implement TreeWalker (part of the Level 2
> Traversal module), which has as its first use case how to navigate
> the tree without seeing entity reference nodes. Not that I disagree
> that tripping over them can be a pain, but there is now a DOM way
> of not doing so.

I was thinking of that as I wrote, in fact.  But the TreeWalker
is still a "look but don't touch" model ... the children of
an entity ref are still going to be readonly, even if the walker
masks the EntityReference from your sight.  Solutions addressing
the "I need to touch it!" aspect can't use just TreeWalker.

Perhaps the general issue is that whenever DOM starts to offer
APIs for populating a DOM Document, there need to be options
about what features should be exposed.  For my usage, things
like EntityReference and CDATASection (vs normal Text) would
hardly ever be used.

Or perhaps there should be variants of Element.normalize() that
give you other options for getting rid of extra tree nodes.
"No adjacent text nodes" is covered.  How about flags for cases
like "don't use EntityReference", "don't use CDATASection",
and "don't use Comment"?

- Dave

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev at ic.ac.uk
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/ and on CD-ROM/ISBN 981-02-3594-1
To unsubscribe, mailto:majordomo at ic.ac.uk the following message;
unsubscribe xml-dev
To subscribe to the digests, mailto:majordomo at ic.ac.uk the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa at ic.ac.uk)