On Tuesday 18 March 2008, Benson Margulies wrote:
> I've implemented a schema cache. I have some systest failures.
>
> I've assumed that once we pull schemas from a WSDL, we're done. Does
> anyone know of some other process in which we get more schema from
> somewhere else later?

Cases I know about:
1) Obviously all the code first cases where we don't use a wsdl....

2) WS-RM, if validation is turned on, does do something to pull in the 
rm/addressing schemas, but I don't think it uses the XmlSchema stuff.  
(I wrote it, but I honestly don't remember)   I think it just goes 
directly to StreamSource  and then gets the Schema object needed for the 
validating parsers.

What tests are failing?  Is it the type test failing?  Are you cacheing 
both the XmlSchema object and the DOM?   

Due to bugs in XmlSchema (fixed in 1.4 if they ever get it out), I know 
the type tests fail if you trying to validate by using the XmlSchema 
form to produce the Schema object used for validation.  It loses 
information that is needed.  That's why we hold onto the DOM right now.  


-- 
J. Daniel Kulp
Principal Engineer, IONA
[EMAIL PROTECTED]
http://www.dankulp.com/blog

Reply via email to