According to the EJB 3.0 specs, there is and hierarchical model classlocal, methodlocal and global interception, which takes inheritance on all this levels into account. Unfortunately in JSR 299 it is not exactly stated, wether there can be multiple lifecycle/interception methods due to inheritance.
In 4.2 injection of inherated fields is required. In the next sentence the same phrases are used for @PostConstruct/@PreDestroy lifecycle methods, so IMHO inheritance must be taken into account for them too. wdyt? br, Sven 2009/11/20 Sven Linstaedt <[email protected]> > As far as I remember, exactly this is stated in 4.2. I am about of > finishing work for today, but I will have a look at the spec later this > evening. > > br, Sven > > > > 2009/11/20 Mark Struberg <[email protected]> > > gnah sorry forget it, I misread the example Sven gave us :/ >> >> the CreationalContext is for >> >> class A { >> @inject B b; >> } >> >> class B { >> @inject A a >> } >> >> >> but your sample is a derivation problem. >> Have to look at it, but as far as I remember most EE containers, for an >> instance of B only initB() must be called. It's up to initB() to >> subsequently call initA() itself (initA() must not be called at all for B >> instances). >> >> LieGrue, >> strub >> >> --- Mark Struberg <[email protected]> schrieb am Fr, 20.11.2009: >> >> > Von: Mark Struberg <[email protected]> >> > Betreff: Re: Inheritance of class local interception and lifecyle >> invocation? >> > An: [email protected] >> > Datum: Freitag, 20. November 2009, 18:24 >> > Hi folks! >> > >> > This should be working. At least that was exactly the >> > reason why we have a 'CreationalContext' in our interfaces >> > ;) It's only purpose is to make cyclic bean references >> > work. >> > >> > LieGrue, >> > strub >> > >> > --- Eric Covener <[email protected]> >> > schrieb am Fr, 20.11.2009: >> > >> > > Von: Eric Covener <[email protected]> >> > > Betreff: Re: Inheritance of class local interception >> > and lifecyle invocation? >> > > An: [email protected] >> > > Datum: Freitag, 20. November 2009, 17:32 >> > > On Fri, Nov 20, 2009 at 11:15 AM, >> > > Sven Linstaedt >> > > <[email protected]> >> > > wrote: >> > > > Hi, >> > > > >> > > > has someone tested the correct order of class >> > local >> > > interception and >> > > > lifecycle invocation in the case of inherited >> > beans, >> > > like it is specified in >> > > > 4.2.? >> > > > >> > > > Like: >> > > > >> > > > class A { >> > > > @PostConstruct >> > > > void initA() {} >> > > > } >> > > > >> > > > class B extends A { >> > > > @PostConstruct >> > > > void initB() {} >> > > > } >> > > >> > > Haven't here, but would be interested in your >> > findings. >> > > >> > > I believe even basic interceptor ordering via >> > beans.xml is >> > > currently >> > > non-deterministic so cross your fingers! >> > > >> > > -- >> > > Eric Covener >> > > [email protected] >> > > >> > >> > __________________________________________________ >> > Do You Yahoo!? >> > Sie sind Spam leid? Yahoo! Mail verfügt über einen >> > herausragenden Schutz gegen Massenmails. >> > http://mail.yahoo.com >> > >> >> __________________________________________________ >> Do You Yahoo!? >> Sie sind Spam leid? Yahoo! Mail verfügt über einen herausragenden Schutz >> gegen Massenmails. >> http://mail.yahoo.com >> > >
