Hi Sergey, Thanks for the comment. My rational is as follows:
- The spring config method should closely mirror the non-spring config method. - There may be many situations where your server is customised in some way e.g. injected interceptors - I think in these cases it wouldn't make sense to have multiple servers defined each with the same customisation. - While some scenarios may be covered by sub-resources, in many this is not applicable. - Finally this the patch allows the choice to inject multiple root resources or not, allowing one to choose their ideal approach. In short it's no show stopper but I think it is useful functionality. Barry On Feb 6, 2008 4:13 PM, Sergey Beryozkin <[EMAIL PROTECTED]> wrote: > Hi > > So what sort of problem this patch is trying to address ? > Why can't a user do : > > <jaxrs:server id="customerService" address="/"> > </jaxrs:server> > > <jaxrs:server id="restfulUserService" address="/user"> > </jaxrs:server> > > The proposed configuration makes me think that, in this case, a > restfulUserService is a subresource... > > Cheers, Sergey > > > Hi, > > > > I've created a patch for CXF-1416. It now enables multiple root > resources to > > be configured. > > > > https://issues.apache.org/jira/browse/CXF-1416 > > > > I've tested this manually with the following similar config to below: > > > > <jaxrs:server id="customerService" address="/"> > > <jaxrs:serviceBeans> > > <!--multiple services support --> > > <ref bean="restfulUserService" /> > > <ref bean="restfulLoginService" /> > > </jaxrs:serviceBeans> > > </jaxrs:server> > > > > There are currently no tests for the Spring bean definition parser that > I > > can see and I'm not sure exactly how to write a valid unit test for > this. > > Maybe someone can enlighten me! > > > > Thanks, > > > > Barry > > > > ---------------------------- > IONA Technologies PLC (registered in Ireland) > Registered Number: 171387 > Registered Address: The IONA Building, Shelbourne Road, Dublin 4, Ireland >
