- From: Eve L. Maler <eve.maler@east.sun.com>
- Date: Fri, 01 Sep 2000 15:40:59 -0400
- To: Eric van der Vlist <vdv@dyomedea.com>
- Cc: www-xml-linking-comments@w3.org
At 07:46 PM 9/1/00 +0200, Eric van der Vlist wrote:
>Hi,
>
>I am trying to use XLink/XPointer local 'bare names' to specify links
>for nodes within the same document.
>...
>And I wonder how I can validate (to check the existence of the
>references) this kind of document.
In general, we have stayed away from solutions for checking link
validity/robustness in the XLink and XPointer specs. There is even a
section on the difficulty of the problem at the following location from
early 1999:
http://www.w3.org/TR/NOTE-xptr-req#Robustness
However, the "pseudo-IDREF" case is very nearly a syntactic validity issue,
rather than a robustness-over-time issue. Even so (personal opinion here),
I don't think it would be practical in the general case for XLink or
XPointer to mandate checking of the sort that an XML processor does,
because the ID might be in a totally separate document ("doc.xml#id"); the
availability of resources isn't stable enough to give a 100% answer as to
whether the ID'd thing is there, even if the document hasn't changed over time.
You wouldn't be able to use regular DTD validation to check the "#"
reference, obviously, because it's not an XML IDREF. But you could use
something like XSLT (Schematron?) to do this level of checking.
Eve
--
Eve Maler +1 781 442 3190
Sun Microsystems XML Technology Center eve.maler @ east.sun.com
Received on Friday, 1 September 2000 15:40:41 UTC