Computed Step Parameter and a Group Equivalence

I'd like to note that the following should be equivalent
in our language:

<p:step name="test" type="p:xslt">
   <p:parameter name="computed" step="somewhere" source="x"
                select="/doc/param"/>
   ...
</p:step>

and

<p:group name="test">
   <p:output name="result" step="inner" source="result"/>
   <p:parameter name="computed" step="somewhere" source="x"
                select="/doc/param"/>
   <p:step name="inner" type="p:xslt">
     <p:import-parameter name="computed"/>
     ...
   </p:step>
</p:group>

Now, I like to make sure everyone agrees that these have the
same effect in that the XSLT step gets the same parameter
named 'computed'.  Yes?

Second, I wonder if it makes sense to note this in the spec?
After writing this all out, it isn't necessarily compact
and certainly motivates the user as to why we have
such constructs.  It also shows how parameters can be
computed for a group of steps.

We could add a note in the draft demonstrating this
equivalence.



--Alex Milowski

Received on Friday, 20 October 2006 15:23:40 UTC