a) further alignment of the "attribute element syntax"

ATFers:

[2] contains a syntax that does not align with the final proposal in [1],
but rather reflects an intermediate form.

The final proposed syntax in [1] eliminated the @body vs @element
distinction, concluding with a single, @body attribute.  Similarly the
proposal for /interface/attribute should also reflect this:

<attribute name="ncname" type=”qname” access=”get|set|both”>
…
</attribute >

This approach is similar to what Umit proposed in [3].

Several things to note:
/interface/attribute/@type is the replacement for @body or @element in the
previous proposal.  The qname value could reference an xsd:complexType or
an xsd:element.

In parallel to [1], the anonymous in-line xsd:complexType definition is
removed.

We MUST allow attribute and element extensibility to provide the hooks for
additional attribute meta-data to be specifiable.  Perhaps this goes
without stating, as having these extensibility points would be consistent
with the approach WSDL 1.2 is taking with other WSDL elements.

Diving into the meaning of @access, it is not clear to me what the value of
“set” really implies.  Do we want “write only attributes”?  Can someone
come up with a compelling use case for this?  Perhaps @access should become
@readOnly which is a Boolean with default=”true”. @readonly=”true” is
equivalent to @access=”get”, @readonly=”false” is equivalent to
@access=”both”.  There is no equivalent to @access=”set”.

This leaves us with a final model for the proposal which could look like:

<attribute name="ncname" type=”qname” readonly=”boolean” anyAttribute*>
   <anyElement/>*
…
</attribute >

[1] http://lists.w3.org/Archives/Public/www-ws-desc/2003Jul/0025.html
[2]
http://lists.w3.org/Archives/Public/public-ws-desc-state/2003Jul/0044.html
[3]
http://lists.w3.org/Archives/Public/public-ws-desc-state/2003Jul/0043.html

++++++++
Steve Graham
sggraham@us.ibm.com
(919)254-0615 (T/L 444)
STSM, On Demand Architecture
++++++++

Received on Saturday, 19 July 2003 13:20:23 UTC