- From: Alessandro Vernet <avernet@orbeon.com>
- Date: Tue, 15 May 2007 10:51:26 +0200
- To: public-xml-processing-model-wg <public-xml-processing-model-wg@w3.org>
On 5/14/07, Norman Walsh <ndw@nwalsh.com> wrote: > <p:matching-documents> > <p:input port="source"> > <p:inline> > <odd/> > </p:inline> > <p:inline> > <even/> > </p:inline> > <p:inline> > <odd/> > </p:inline> > <p:inline> > <even/> > </p:inline> > <p:inline> > <odd/> > </p:inline> > </p:input> > <p:option name="test" value="$p:position mod 2 = 1"/> > </p:matching-documents> > > We added $p:position for this very case. This is a good use case, and I agree: in this case you want that information to be made available the 'test' XPath expression. I only feel I should insist on two points: 1) That we clearly specify for each component what is exposed to XPath expression evaluated by the component. 2) That we use functions. Sorry, I can't help coming back to this once again: this example would IMHO be clearer if instead of $p:position we had document-position(). In this case a function does a better job than a variable at conveying that the value can be different depending on when that function is called by the step. And the name does a better job at describing what the function does. Alex -- Orbeon Forms - Web 2.0 Forms for the Enterprise http://www.orbeon.com/
Received on Tuesday, 15 May 2007 08:51:31 UTC