2009/5/22 Timothy Waxland <[email protected]>:

> with the above, all is well but i can not impose that on the server (client
> webmethods system). i fear i have missed something simple that is causing my
> client components to not be populated with respective values. i also
> understood from my reading that the above soap response is valid (dumb
> naming aside) and that even though the prefix is not used, the fact that it
> is declared is enough and it should also apply to child elements. from the
> w3c recommendation i read: (http://www.w3.org/TR/REC-xml-names)
>
> <x xmlns:edi='http://ecommerce.example.org/schema'>
>  <!-- the "edi" prefix is bound to http://ecommerce.example.org/schema
>      for the "x" element and contents -->
> </x>
>
> so in my similar case, even though x does not include the prefix, the
> namespace prefixed by edi applies to x and all its contents.

Unless I'm very much mistaken, the phrase 'the "edi" prefix is bound
to http://ecommerce.example.org/schema for the "x" element and
contents' doesn't mean 'the x element and all its children are in the
http://ecommerce.example.org/schema namespace'. It means, I thought,
'anywhere you use the edi prefix in the x element or any of its
children, it will be bound to http://ecommerce.example.org/schema'.
But AFAIK unprefixed elements will still be in the default namespace.

Apologies if that's what you meant and I'm misunderstanding...

> in my case,
> which i believe is the same, it seems to fail. using soapUI (or even in my
> own integration tests using a fake but running web service) the whole thing
> works fine. but as an example the default response that soapUI generates
> uses prefixes everywhere as does my fake web service. in any case, i believe
> the response to be valid and something on the cxf end is misinterpreting or
> not applying the namespace correctly.

Hard to tell much without your WSDL, but what happens when:

(a) you right-click on the problematic message in soapUI and choose 'Validate'?

(b) you turn on schema validation in CXF?

Andrew.



-- 
:: http://biotext.org.uk/ ::

Reply via email to