Re: [Scrollkeeper-devel] structure of extracted index page

[Daniel Veillard <veillard redhat com>]

On Thu, Apr 26, 2001 at 01:47:28PM +0100, Mary Dwyer wrote:
> hi Daniel
> 
> I am somewhat confused on this issue too. I sent a mail earlier today, that you 
> may have not received, which relates to this (see attached).
> 
> The main point is the problem of having scrollkeeper generating ids relating to 
> the original document. Specifically:
> from what I understand the convertor at run time generates ids where required 
> (using generate-id() ).  However,  I don't believe scrollkeeper 
> can predict what these ids will be because:
> 1) scrollkeeper may use a different convertor to the 'run-time' convertor.
> 2) even if the same convertor is used the w3.org description indicates that 
>   "An implementation is under no obligation to generate the
>    same identifiers each time a document is transformed."
>    see http://www.w3.org/TR/xslt#function-generate-id

  Right I know.
  I could change libxslt to actually always generate the same ID
based on the trick described in my last mail to László . But it would
be a bit slower (not noticeable IMHO) and a feature dependant on libxslt
specific implementation.

Daniel

-- 
Daniel Veillard      | Red Hat Network http://redhat.com/products/network/
veillard redhat com  | libxml Gnome XML XSLT toolkit  http://xmlsoft.org/
http://veillard.com/ | Rpmfind RPM search engine http://rpmfind.net/