Well, baisically, you have three options:
1. Returned escaped XML as xsd:string. This is usually very easy to do.
However this denies you from checking against a schema - your client
will have to have "blind faith" in the xml you send instead of being
able to verify it against the web service schema.
2. Change your schema to include the returned xml, then return the xml
by returning it's root. This is usually much harder to do with
automatic tools, but will enable the client to know if the xml you sent
is valid.
3. Use a seperate cache and send the client the key. This should be
done only if the web service is asynchronous and then it's usually the
only option then. This can be done by either using an internal cache
and another web service request (then you're back to 1 and 2), or by
external cache (ftp/http).
|
Similar Threads
