Re: [xml] redicting parts of trees

[Daniel Veillard <veillard redhat com>]

On Thu, May 19, 2005 at 11:00:54PM +0200, Kasimier Buchcik wrote:
> Hi,
>
> On Thu, 2005-05-19 at 16:52 -0400, Daniel Veillard wrote:
> > On Thu, May 19, 2005 at 10:28:23PM +0200, Kasimier Buchcik wrote:
> > > Is slowly boils down to using just the xmlReconciliateNs mechanism
> > > to declare all missing ns-declarations on the given node, as Daniel
> > > first suggested - although you have been warned ;-)
> >
> >   I'm not the one this API is designed for, so I will refrain from
> > getting in the choices unless I see something which would be too
> > dangerous to put in the API. I'm happy to see the 3 main consumers
> > of that API firmly discussing the options and fully understanding
> > the scope of the problem. You just need to not forgot ID/REF in the
>
> Ah, right ID/IDREF: maybe we need an additional traversal before any
> attempt to adopt can be done; so reject the node if this constraint
> is violated. Would be a mess to stop somewhere in the branch.
> Any ideas here?

  No just check if an attribute being moved is an ID in the ID list of
the source document and then unregister it from the ID table. The more 
painful part if that when adding back in the new document you would
need to update the target ID table too. The good point is that those
operations are NOP if you don't have a DTD (and they are not xml:id),
checking the presence/emptyness of the id table on the source and the
presence of a DTD on the target should make big shortcut in most cases.

Daniel
-- 
Daniel Veillard      | Red Hat Desktop team http://redhat.com/
veillard redhat com  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/