- From: Daniel Barclay <Daniel.Barclay@digitalfocus.com>
- Date: Tue, 01 May 2001 18:42:29 -0400
- To: frystyk@microsoft.com
- CC: "'Doug Davis'" <dug@us.ibm.com>, xml-dist-app@w3.org
Henrik Frystyk Nielsen wrote: > .... > That would be bad as we then have no mechanism for identifying a SOAP > HTTP request. Do servers typically accept different types of requests/messages at a given URI? (Can someone enhance my understanding of how a SOAP server would typically be set up? I would think that a SOAP server would be set up at a given URL (or that URL and "child" URLs) within an HTTP server. Only SOAP messages should be posted to that URL. The SOAP server would handle only two kinds of messages: messages it recognized as valid SOAP messages those that it did not recognize. I don't see a need for such a SOAP server to differentiate SOAP requests from any other valid requests. So what am I not thinking of? Is the differentiation for a not-just-SOAP dispatcher that receives POST requests at some URI, and calls a SOAP handler for SOAP message and other handlers for other messages? Is the differentiation for the top-level HTTP server? What would it do differently based on whether a message was a SOAP message or some other message type? Is the differentiation for caches or firewalls or something similar? Is the differentiation for something else? Thanks.) Daniel -- Daniel Barclay Digital Focus Daniel.Barclay@digitalfocus.com
Received on Tuesday, 1 May 2001 18:42:15 UTC