- From: Innovimax W3C <innovimax+w3c@gmail.com>
- Date: Tue, 10 Nov 2009 14:58:39 +0100
- To: Toman_Vojtech@emc.com
- Cc: public-xml-processing-model-wg@w3.org
Thanks Vojtech,
That was exaclty the point why I didn't want to have @use-when when we
proposed it at the WG
I'm afraid the context have to be explicitly clear to avoid such strange case
On the other hand, we shouldn't change the spec on the purpose on one
particular implementation strategy
But as I already pointed this out, use-when is a nightmare for XProc
Designers especially because of that !
Cheers,
Mohamed
On Tue, Nov 10, 2009 at 1:29 PM, <Toman_Vojtech@emc.com> wrote:
> I now actually think that in its present form, use-when is
> underspecified. A the moment, the spec says that use-when "must contain
> an XPath expression that can be evaluated statically". But what does
> that really mean? How does that affect XPath extension functions like
> p:step-available(), p:value-available() etc.?
>
> For instance, does p:step-available() in use-when see only standard
> XProc steps, or also other nested or imported steps? What about cases
> like this:
>
> <p:declare-step type="ex:step1"
> use-when="p:step-available('ex:step2')">
> ...
> </p:declare-step>
>
> <p:declare-step type="ex:step2"
> use-when="p:step-available('ex:step1')">
> ...
> </p:declare-step>
>
> Personally, I think we will have to be more specific about what you can
> do in use-when, and what limitations are there if you use the XPath
> extension functions in use-when expressions.
>
> Regards,
> Vojtech
>
>
--
Innovimax SARL
Consulting, Training & XML Development
9, impasse des Orteaux
75020 Paris
Tel : +33 9 52 475787
Fax : +33 1 4356 1746
http://www.innovimax.fr
RCS Paris 488.018.631
SARL au capital de 10.000 €
Received on Tuesday, 10 November 2009 13:59:15 UTC