[XML-DEV Mailing List Archive Home] [By Thread] [By Date] [Recent Entries] [Reply To This Message]

Re: James Anderson's table

• From: james anderson <James.Anderson@m...>
• To: XML Dev <xml-dev@i...>
• Date: Thu, 13 Aug 1998 10:59:20 +0200

John Cowan wrote:
> 
> [... yes, thank you.]
> 
> But anyway, here are the correct values, based on the principle
> that a value defaulted from the DTD is equivalent in every way

I am not certain that this is "correct". As I had noted in the "cover note", I
had followed one of the alternatives:

  IV: for the moment, i've chosen the precedence
       tag-content,
       bindings-from-containing-elements,
       attribute-defaults
  where the bindings-from-containing-elements takes its initial value from the xml-decl.

The justification is that lexically apparent bindings (modulo the answer on
entities) take precedence over declared defaults. One could reasonable take
that a step further, and argue that the distinction fixed v/s simply default
on a declared attribute should figure here as well, with the effect that fixed
defaults take precedence over the bindings in the lexical context, while
simple defaults do not. I didn't do that, but the information is there to make
the distinction, and I don't see anything "wrong" with it.

> to one explicitly present in the instance (except of course that
> if both are available, the explicit one takes precedence):
> 
> ...
>
> The cell "cb" vs. "cd" can come out either way, depending on which
> is the nearer ancestor.

This is also not self-evident. The presence of the lexically apparent binding
could quite reasonably be said to take precedence. I would hestitate to say
proximity is the right criteria for precedence until either I've seen some
'real' instances or the standard describes the situation and specifies how to
resolve it.
[Even then I might differ on whether it was "right", but at least I'd be
certain about what to implement...]

xml-dev: A list for W3C XML Developers. To post, mailto:xml-dev@i...
Archived as: http://www.lists.ic.ac.uk/hypermail/xml-dev/
To (un)subscribe, mailto:majordomo@i... the following message;
(un)subscribe xml-dev
To subscribe to the digests, mailto:majordomo@i... the following message;
subscribe xml-dev-digest
List coordinator, Henry Rzepa (mailto:rzepa@i...)

• Follow-Ups:
  • Re: James Anderson's table
    • From: John Cowan <cowan@l...>

• References:
  • James Anderson's table
    • From: John Cowan <cowan@l...>

• Prev by Date: Helping with specific problems (was Re: [help]xml4j (Linux))
• Next by Date: Re: John Cowan's view on SAX extensions for ns drafts
• Previous by thread: James Anderson's table
• Next by thread: Re: James Anderson's table
• Index(es):
  • Date
  • Thread