> Basically, they want to point to a structure inside an WSDL instance, > and they want to use an XPointer-type thing to do. I wonder if the use of fragment ids is a requirement, or just an assumed form of solution? I've found that if something's worth identifying, it's worth identifying without a fragment id. My approach would be to treat this as a compound document serialization problem, and use rdf:about to identify the consituent documents within the compound document (the WSDL file). Using the stockquote example from the WSDL 1.1 spec, if I wanted to address the "GetLastTradePrice" operation, I'd just do this; <operation name="GetLastTradePrice" rdf:about="http://example.com/stockquote/operation/GetLastTradePrice"> <input message="tns:GetLastTradePriceInput"/> <output message="tns:GetLastTradePriceOutput"/> </operation> Then references to that operation could use that URI. I haven't used this approach (as I don't use WSDL), but I don't think there are any RDF/WSDL integration "gotchas" with it. I could be wrong though. MBReceived on Monday, 31 March 2003 11:32:26 GMT
This archive was generated by hypermail 2.2.0+W3C-0.50 : Monday, 7 December 2009 10:55:46 GMT