Curiouser and curiouser ... same problem in 5.2.4 as in 5.2.5 (so, no, I don't think the code has changed between releases). I only started seeing this problem in my client project after the upgrade to 5.2.5, but perhaps there's some other recent change (to AbstractFoo, I suppose) that explains its emergence as a problem.
But ... I think the PropertyConduitSource or PropertyAccess is broken for this case; the type of property "bar" should be Bar not AbstractBar. Could this be a JVM issue ... some difference in how the JVM identifies covariant return types in subclasses? On Fri, Apr 1, 2011 at 11:49 AM, Howard Lewis Ship <[email protected]> wrote: > I'm seeing what could be a serious problem in 5.2.5. Was anyone > working on the PropertyConduitSource code in 5.2.5? > > There's been a subtle shift in how the code deals with implementation > of abstract methods. > > public abstract class AbstractFoo > { > public abstract AbstractBar getBar(); > } > > public class Foo extends AbstractFoo > { > public Bar getBar(); > } > > > In this circumstance, the PropertyConduitSource sees the type of > property "bar" of class Foo as AbstractBar, not Bar. > > I'm going to check if this is a change from 5.2.4 to 5.2.5. > > -- > Howard M. Lewis Ship > > Creator of Apache Tapestry > > The source for Tapestry training, mentoring and support. Contact me to > learn how I can get you up and productive in Tapestry fast! > > (971) 678-5210 > http://howardlewisship.com > -- Howard M. Lewis Ship Creator of Apache Tapestry The source for Tapestry training, mentoring and support. Contact me to learn how I can get you up and productive in Tapestry fast! (971) 678-5210 http://howardlewisship.com --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
