I thought so ;) Ok, I'm going to add a subform-component to my approach. I still need the attribute on the command-components, though, cause in my use-case (and probably many out there) the command component may be _outside_ the subform-component. Additionally, I might have commands which don't trigger validation and update for _any_ subform at all, even if they are children of a subform.
Or do you have an idea for handling this as well? It's a PITA that I need the behaviour right now and can't switch in neither Tobago (they have something like this as well, but the components don't work together with normal Tomahawk components) nor ADF - it not being in incubator so far. We should get the process of merging together the component sets finished ASAP. regards, Martin On 12/29/05, Adam Winer <[EMAIL PROTECTED]> wrote: > I don't think tweaking the API classes is legit - even if the signature > didn't change. The components would break if you swapped in the > JSF RI, and that's not OK. Behavioral portability is just as important > as successful compilation! > > ADF Faces addresses this with a "subform" component: > > http://tinyurl.com/b2n2t > > -- Adam > > > On 12/28/05, Martin Marinschek <[EMAIL PROTECTED]> wrote: > > Hi *, > > > > I've implemented a proposal for "partial validation and model update". > > > > Essentially what I've done is I've given the commandLink and > > commandButton a new attribute, "actionFor". This attribute is a > > comma-separated lists of container-component-ids for which validation > > and model update should happen. > > > > So if you have a > > > > <h:panelGroup id="xxx"> > > <h:inputText .../> > > </h:panelGroup> > > > > and a > > > > <t:commandButton actionFor="xxx"/> > > > > the validation and model-update phases will only be triggered for > > components which are descendants of the panelGroup with id="xxx". > > > > Sounds good? > > > > Fellow developers, I had to tweak the API-classes a little to make > > this happen - I didn't change any signature or so, but I store stuff > > in the externalContext.getRequestMap(). Maybe someone of you can come > > up with a better, less intrusive solution to the problem? > > > > The only other solution I could think of would have been to implement > > processValidations() and processUpdates() for every extended > > component, not a very interesting outlook! > > > > regards, > > > > Martin > > > > -- > > > > http://www.irian.at > > > > Your JSF powerhouse - > > JSF Consulting, Development and > > Courses in English and German > > > > Professional Support for Apache MyFaces > > > -- http://www.irian.at Your JSF powerhouse - JSF Consulting, Development and Courses in English and German Professional Support for Apache MyFaces
