On Fri, Nov 03, 2006 at 07:10:10PM +0200, Andrew W. Nosenko wrote:
> On 11/3/06, Daniel Veillard <[EMAIL PROTECTED]> wrote:
> >On Fri, Nov 03, 2006 at 05:41:50PM +0200, Andrew W. Nosenko wrote:
> >> >  Currently implemented only for Readers built on a document
> >>
> >> Yes, I saw it.
> >>
> >> >There might be semantic differences but I could not spot them 
> >immediately
> >> >the code is really different.
> >>
> >> Yes, I asked about functions semantic differences (may be using bad
> >> wording :-( ), not about implementation.
> >
> >  if you're not happy with my answers maybe I should just not reply.
> 
> I'm very sorry and please very much to excuse me if I affront you in
> the any way.  It was absolutely unintended! :-(

  okay, no worries :-)

> >> OK.  Let me rephrase: Is there any _intended_ semantic differences
> >> between these functions?
> >
> >  Not that I can think of. But 99% of the users use the reader to process
> >large document on limited memory, and clearly the first one won't work for
> >this so I *had to* point it out in my reply.
> 
> Thanks!  It's exactly what I wanted to know.

  Basically the first function after testing a few preconditions just uses
the ->next pointer from the tree. The second one does a series of Read() calls
until it gets back to a sibling. it should have the same semantic really,
but I may have missed some weird side effect, for example if you have an
entity reference which has not been replaced.

Daniel

-- 
Red Hat Virtualization group http://redhat.com/virtualization/
Daniel Veillard      | virtualization library  http://libvirt.org/
[EMAIL PROTECTED]  | libxml GNOME XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine  http://rpmfind.net/
_______________________________________________
xml mailing list, project page  http://xmlsoft.org/
[email protected]
http://mail.gnome.org/mailman/listinfo/xml

Reply via email to