Just to expand on this......

CXF DOES pass the JAX-WS TCK and that is definitely the primary target.   
However, WSI-BP compliance is also considered fairly important.  The JAX-WS 
TCK actually does run a slew of tests to test many of the WSI-BP assertions 
so some level of BP compliance is important for any JAX-WS implementation.

However, I don't think anyone has gone through every single assertion in the 
BP spec to make sure we're 100% compliant with every single one.   If an 
issue is found, we definitely would look into fixing it if at all possible 
(within the jaxws spec, of course), but there are a LOT of assertions in the 
spec so going through one by one is a huge undertaking.

Specifically regarding mtom...   CXF does support mtom and should be compliant 
with other mtom implementations and the BP specs.   

Dan


On Tuesday 12 August 2008 10:46:47 pm Glen Mazza wrote:
> bharath_t19 wrote:
> > Thanks Glen.MTOM is a part of the WS-I Basic Profile 1.2.Is CXF compliant
> > with this?
>
> Short answer: no.
>
> Long, mushy, politician's answer:  CXF is compliant with the JAX-WS spec,
> it can't serve two masters.  Where the JAX-WS spec has shaped by the WS-I
> Basic Profile (e.g., no rpc/encoded WSDL's allowed by the latter, so the
> former doesn't permit it), the answer is "yes".  But to the extent that
> WS-I is really an issue about how humans code their web services, the
> answer is "no"--CXF doesn't stop people from violating WS-I.  Sometimes,
> though, it *does* stop people from violating a WS-I rule even though it's
> permitted by JAX-WS, for example here:
> http://www.jroller.com/gmazza/date/20071019#notes (See note #1)--this is a
> case where CXF's internal implementation relies on people not violating a
> WS-I rule.
>
> Glen


-- 
Daniel Kulp
[EMAIL PROTECTED]
http://www.dankulp.com/blog

Reply via email to