- From: christopher ferris <chris.ferris@east.sun.com>
- Date: Tue, 05 Jun 2001 07:48:41 -0400
- To: Noah_Mendelsohn@lotus.com
- CC: Marc Hadley <marc.hadley@Sun.COM>, xml-dist-app@w3.org
- Message-ID: <3B1CC719.D9DCE050@east.sun.com>
Noah_Mendelsohn@lotus.com wrote: <snip/> > XMLP/SOAP namespaces. It MUST generate a fault (see section 4.4) on > receipt of messages using >>SOAP/XMLP-defined elements and attributes<< > that have incorrect namespaces." How is any processor (SOAP, XML, otherwise) supposed to arrive at the conclusion that an incorrect namespace has been used for >>SOAP/XMLP-defined elements and attributes<< ? This seems to suggest that an element which has a local name of "Envelope" and a namespace of "http://www.example.com/mynamespace" is somehow invalid? In fact, it is a wholly distinct and perfectly valid element that has a qualified name of http://www.example.com/mynamespace:Envelope quite possibly with distinct semantic meaning from the SOAP/XMLP Envelope. The only case that could be made for a fault would be the presence of an element or attribute that is assigned to the SOAP/XMLP namespace that also has a local name that is NOT in the set of names reserved for SOAP/XMLP element and attribute local names. e.g. a qname of: http://schemas.xmlsoap.org/soap/envelope/:Frob would result in a generated Fault, but a qname of: http://www.example.com/mynamespace/:Envelope would not. I would also point out that permitting a SOAP/XMLP processor to process localnames only (ignoring the namespace) should be considered harmful. How is that same processor supposed to distinguish between like-element/attribute names from different namespaces??? > > What do you think? > > ------------------------------------------------------------------------ > Noah Mendelsohn Voice: 1-617-693-4036 > Lotus Development Corp. Fax: 1-617-693-8676 > One Rogers Street > Cambridge, MA 02142 > ------------------------------------------------------------------------
Received on Tuesday, 5 June 2001 08:32:23 UTC