Which goes back to your original statement in that you have to "identify"
ID attributes via a custom resolver, schema constraint, or DOM3 APIs calls?

On Mon, Aug 20, 2012 at 10:15 PM, Cantor, Scott <[email protected]> wrote:

> On 8/20/12 10:01 PM, "Michael Bishop" <[email protected]> wrote:
> >
> >Right, importing it and appending it are two different things.  So it
> >seems like a new requirement is that the content has to be part of the
> >document first.
>
> That isn't new, but I suppose it may be in the sense that perhaps before
> it wasn't making any DOM calls that cared whether the document was rooted.
>
> >I'm guessing it's related to the ID bug; I believe an xml:id has to be
> >unique within the entire document.
>
> It does, but that's not why, and that's not enforced (which is a major bug
> in Xerces IMHO). The problem is that it's calling getElementNSById() and
> that method starts at the doc root.
>
> Also, xml:id is a special attribute that happens to be an XML ID. Not all
> XML IDs are xml:id. Any attribute can be an ID.
>
> -- Scott
>
>

Reply via email to