Re: XSLT Component

On 2/1/07, Alessandro Vernet <avernet@orbeon.com> wrote:
>
> On 1/31/07, Alex Milowski <alex@milowski.org> wrote:
> > A smart implementation can do the right thing a choose the right
> > processor.  I did that in smallx in that XSLT 1.0 was handled by my version
> > of  XT and XSLT 2.0 was handled by saxon.
>
> I agree.
>
> I just have some doubts about the warning mechanism. I am referring
> to: "An implementations are allowed to use and XSLT 2.0 processor to
> run an XSLT 1.0 transformation but a warning must be issued". I would
> prefer, like Norm suggested, to have an optional parameter that the
> pipeline author can set to say that he really wants to use an 1.0
> engine or a 2.0 engine, whatever the version attribute in the
> stylesheet says. In most cases pipeline authors won't need to worry
> about this parameter and everything will work for them as expected.

Totally agreed

The content sniffing must be a fallback. The user may specifies a
version. If not the implementation should try to apply the most
appropriate transformer available, by, for example, looking at the
version attribute.

One other rationale if necessary, is that if attribute is positionned,
depending on what is available as transformer, the warning may even be
issued statically. If not, it must be issued dynamically.

>
> Alex
> --
> Orbeon Forms - Web Forms for the Enterprise, Done the Right Way
> http://www.orbeon.com/
>
>


-- 
Innovimax SARL
Consulting, Training & XML Development
9, impasse des Orteaux
75020 Paris
Tel : +33 8 72 475787
Fax : +33 1 4356 1746
http://www.innovimax.fr
RCS Paris 488.018.631
SARL au capital de 10.000 €

Received on Thursday, 1 February 2007 07:01:54 UTC