Igor Vaynberg wrote:
> not formcomponents, which i think is a bit weird/wrong.
> what we want is to let the form control the versioning of
> formcomponents, so inside formcomponent constructor we set it to
> false. and then form overrides setversioned() and replicates that call
> to the formcomponents. the problem is that during page constructor
> that wont do anything because form components might be added later -
> the chain is not yet complete - and also defaulting to false is wrong
> i think. johan you got any thoughts on this?
In my opinion setting it to true by default is also more natural and
consistent with pages. This is unexpected behavior and one of the key
things I have with wicket is its predictability: Components usually just
behave as you expect, so making it consistent in wicket 2 would be a
good idea.
> in 2.0 we can fix this by overriding isversioned on formcomponent and
> polling the form directly instead of having the form push.
I think the polling behavior should be the default. Why shouldn't a form
be versioned?

Unfortunately, the book 'Pro Wicket' does not cover anything at all
about versioning and back button support (a search of the document for
'back button', 'modelChanging', and 'modelChanged' did not return any
results. I just hope that 'Wicket in Action' will cover versioning and
back button support in detail since it is such a major point of wicket
(one of the selling points).

> -Igor

Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
Wicket-user mailing list

Reply via email to