- From: Alex Milowski <alex@milowski.org>
- Date: Wed, 12 Jul 2006 20:08:22 -0700
- To: public-xml-processing-model-wg <public-xml-processing-model-wg@w3.org>
Alessandro Vernet wrote: > > Alex, > > On 7/12/06, Alex Milowski <alex@milowski.org> wrote: >> The component defines an output called 'result'. That element maps >> that output name to a name that can be referenced in the pipeline. > > OK, got it now. But still, as mentioned in my previous email, I would > find it easier to understand if it was: > > <p:output name="result" label="parsed"/> I've been consistently using 'name' to label things and 'from' to refer to things that have been labeled. Changing the 'name' attribute to 'label' in all places would be consistent. While I prefer 'name' here, I'm not against using 'label'. > Again, this is a purely syntactic consideration, but what would you > think of: > > <p:for-each select="description" over="feed" to="iteration" > replacement="parsed" label="final"> > <p:step kind="p:parse"> > <p:input name="document" from="iteration"/> > <p:output name="result" label="parsed"/> > <!-- Parameters here --> > </p:step> > </p:for-each> The 'label' attribute would be the the output? Yes, I had considered that and didn't rule it out. I just made one choice over the other. I'd be OK with such an attribute. Again, I'd like the choice of 'label' or 'name' attribute to be consistent. Here is a case for something more specific (e.g. label). > What is different: > > 1) Change as discussed earlier for the <p:output> inside the step. > 2) Label on <p:for-each> instead of <p:output> inside the > <p:for-each>. Benefits: lighter syntax, avoid the asymmetry where > there would be a <p:output> but no <p:input>. Yep. --Alex Milowski
Received on Thursday, 13 July 2006 03:08:41 UTC