GenericServer |
Edit this page |
The Wheat Server Engine
The Wheat engine is a multi-protocol server. In general, it accepts requests, each of which is mapped into a message call, and an execution environment. The call is made (in the environment), and the result is then mapped back into the requesting protocol and returned. Both mappings may involve further message calls.
The most common access is via HTTP: A request comes in for a URL. The URL is used to locate a wheat object. The headers (including the host, cookie, and any auth headers) are used to construct an environment (logged in user, session state, etc.). The URL also encodes an method call. The call is made, and then the result is converted back into one of the data types accepted by the requester. This usually means converting the object to XHTML. This usually involves a template application.
Other access types include WebDAV (which is really built on HTTP), a simple command line interface, and perhaps others.
What about SOAP? --MarkLentczner 2003-03-07
What about XMLRPC? --JimKingdon 2003-12-05
Since last March I've looked at the two specs and I much prefer XMLRPC. I have also read a number of things about SOAP that make me dislike it such as its sub-setting of XML syntax. I will have a look at the articles listed below and comment more later. Of course, Wheat could have it's own XML based RPC mechanism if it wanted... --MarkLentczner 2003-12-06
You might want to look at REST ∞, which at a glance seems to be a sane cousin of SOAP. It's the main protocol for Yahoo's Developer API∞, Amazon.com's API∞, and eBay's API∞. --PatrickTufts 2005-10-18
References
Two SOAP vs XMLRPC pages:http://ink.inkdroid.org/talks/soap/xmlrpc.html∞
http://weblog.masukomi.org/writings/xml-rpc_vs_soap.htm∞
There is no comment on this page.
[Display comments/form]
