So, the underlying question then is, what is the value proposition for migrating from SpringDM (or Gemini) to Aires blueprint for shops that rely on Spring annotations? I think the correct answer isn't that Aries cannot use Spring annotations, but rather an explanation of how Aries can use spring annotations. In the wild, there are far more shops that use spring than those that use aries. If moving to aries blueprint will break thier existing code because their spring annotations won't work, there is not cost-benefit to the transition.
Johan Edstrom-2 wrote: > > A bundle that is activated by blueprint would not activate springs > AnnotationProcessors, > nor do I think they would work as you'd have no application context. > > Sure they publish/consume services. > Spring relies heavily on proxies / cglib / aopalliance and whatnot-else, > the actual implementation of say a Dao may be quite far from what you > actually expect. > > Niether extender does more than try to publish / subscribe events from the > osgi service-registry. > > You'd run into the same thing expecting Spring annotations to work > magically in guice. > > /je > On Oct 20, 2011, at 9:48 AM, mikevan wrote: > >> Regarding question 2: I'm wondering why this would be the case. If you >> have >> a bundle that is activated by aries blueprint, and if that bundle then >> uses >> Spring annotations internally, wouldn't a spring context be created >> within >> the blueprint context? If this is true, then wouldn't the annotations be >> usable? If not, this seems like a strong use-case against migration to >> blueprint for bundles that use spring annotations. >> >> Regarding question 3: Doesn't the underlying OSGi framework manage >> wiring? >> I understand that both Aries and SpringDM have thier own extender bundles >> that appear to manage wiring, but don't they both leverage the wiring >> capability of the underlying framework? If so, wouldn't the extenders >> "share" wiring information based on the underlying framework's wiring >> mechanism? >> >> >> Johan Edstrom-2 wrote: >>> >>> It creates a bp context. >>> no. >>> Depends. - SpringProxy issues comes to mind as well as glib usage. >>> >>> On Oct 20, 2011, at 9:31 AM, mikevan wrote: >>> >>>> I have some questions about how Spring and Aries co-exist in Karaf. >>>> >>>> 1) If you have an application that uses aries blueprint, does aries >>>> blueprint create its own application context for bundles? >>>> 2) Can applications that rely heavily on spring annotations work inside >>>> of a >>>> blueprint environment? >>>> 3) Can a bundle using aries successfully wire to a bundle that uses >>>> springdm? >>>> >>>> ----- >>>> Mike Van (All links open in new tabs) >>>> Committer - Kalumet >>>> >>>> Atraxia Technologies >>>> >>>> NCI Inc >>>> >>>> Mike Van's Open Source Technologies Blog >>>> -- >>>> View this message in context: >>>> http://karaf.922171.n3.nabble.com/Aries-and-Spring-Co-Existance-in-Karaf-tp3438050p3438050.html >>>> Sent from the Karaf - User mailing list archive at Nabble.com. >>> >> >> >> ----- >> Mike Van (All links open in new tabs) >> Committer - Kalumet >> >> Atraxia Technologies >> >> NCI Inc >> >> Mike Van's Open Source Technologies Blog >> -- >> View this message in context: >> http://karaf.922171.n3.nabble.com/Aries-and-Spring-Co-Existance-in-Karaf-tp3438050p3438104.html >> Sent from the Karaf - User mailing list archive at Nabble.com. > ----- Mike Van (All links open in new tabs) Committer - Kalumet Atraxia Technologies NCI Inc Mike Van's Open Source Technologies Blog -- View this message in context: http://karaf.922171.n3.nabble.com/Aries-and-Spring-Co-Existance-in-Karaf-tp3438050p3438174.html Sent from the Karaf - User mailing list archive at Nabble.com.
