May 2008 - jboss-dev-forums - Jboss List Archives

[Design the new POJO MicroContainer] - Re: Composite Map key and weak reference

by jason.greene＠jboss.com

"alesj" wrote : "jason.greene(a)jboss.com" wrote : So I take it the annotation refs point to annotations on the class ref? | | | Not really. | But some of class's annotations might be equal to that annotations. | Ah, ok, then you don't want the first option. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152046#4152046 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152046

17 years, 10 months

1
0
0 / 0

[Design of POJO Server] - Re: JBMICROCONT-288 - Invalid feature request and invalid pr

by adrian＠jboss.org

"wolfc" wrote : | (I think this translates to being able to have a serialized reference to a MC bean itself. But to facilitate that the MC bean would need a serialized reference to the Kernel.) No, that's the wrong place like I said above. If the problem is passivation then it is the job of the container to remove implementation details that are not serializable and re-instate them later. i.e. its EJB3's job Equally, you can't put non-serializable stuff in a web session and expect it to be serializable without extra work. I don't see what the issue is really. These must be internal implementation details so you know how to implement this "externalization", you're just being lazy. i.e. you're expecting the MC to do something that is not its job and will likely lead to other problems that don't want, expect this behaviour or want to do something else. DON'T MIX IMPLEMENTATION AND POLICY View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152044#4152044 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152044

17 years, 10 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Bean validation

by alesj

Regarding this JIRA issue: - http://jira.jboss.com/jira/browse/JBMICROCONT-277 I've added this | /** | * Get bean validator bridge. | * | * @param context the owner context | * @return bean validator instance if exists, null otherwise | */ | static BeanValidatorBridge getBeanValidatorBridge(KernelControllerContext context) | { | Controller controller = context.getController(); | ControllerContext validator = controller.getInstalledContext(BeanValidatorBridge.class); | return validator != null ? BeanValidatorBridge.class.cast(validator.getTarget()) : null; | } | a way to get the bridge between jsr303 and our MC code. I've then added hooks into - InstantiateAction (constructor and instantiation validation) - PropertyDispatchWrapper (property injection) - AbstractKernelControllerContext (method invocation) Once I'm (+ hopefully getting Emmanuel to chip in) done with jsr303, I'll implement a proper bridge, which we can simply drop into Controller. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152042#4152042 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152042

17 years, 10 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Composite Map key and weak reference

by alesj

"jason.greene(a)jboss.com" wrote : So I take it the annotation refs point to annotations on the class ref? | Not really. But some of class's annotations might be equal to that annotations. I'm using this with WebBeans kind of matching: | @Red | @Blue | @Inject | public void setFoo(Foo foo) | where class might have | @Red @Blue @Green | public class RGBFoo implement Foo | The code is here - but the caching is currently wrong :-) - http://anonsvn.jboss.org/repos/jbossas/projects/microcontainer/trunk/kern... View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152036#4152036 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152036

17 years, 10 months

1
0
0 / 0

[Design of POJO Server] - Modularising the appserver bootstrap

by adrian＠jboss.org

I've made a start on modularising the appserver bootstrap, as expected there are some subtle problems laying in wait. CHANGE First, what I've done is replaced bootstrap-beans.xml with a bootstrap.xml that confirms to the this simple schema: https://svn.jboss.org/repos/jbossas/trunk/bootstrap/src/resources/schema/... The idea being that the other modularised files are loaded in the order given. This makes the changes to the profile service to use a repository rather a lot easier. See bootstrap-repo.xml which swaps profile-service.xml for profile-service-repository.xml ISSUES Obviously there's some circular references that need to be setup in the correct order to avoid "chicken and egg" problems. This includes making the sure the main "repositories" * ClassLoading (classloading modules) * AspectManager (aop config) * MainDeployer (structure/normal deployers) are configured first so they can accept other services that get registered later (the things in brakets). A good example of this, is that the @JMX aspect won't work if something annotated with it is deployed before the AspectManager. But then the AspectManager itself needs to be registered with JMX also which is a sort of circular reference. Less obvious is that AOP has some dependencies in the classloading (ServiceMBean* and deployers) that need fixing before I can proceed with further modularisation. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152035#4152035 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152035

17 years, 10 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Composite Map key and weak reference

by bstansberry＠jboss.com

Oh, but my class above is intended to be passed as a key to WeakHashMap. The ReferenceQueue polling will remove the only strong ref to the Key, at which point the WeakHashMap will do its normal thing and make the Key/value pair eligible for removal. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152032#4152032 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152032

17 years, 10 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Composite Map key and weak reference

by bstansberry＠jboss.com

I believe Jason's #2 is what I was thinking about in our chat discussion. A rough impl: | package hacks; | | import java.lang.annotation.Annotation; | import java.lang.ref.Reference; | import java.lang.ref.ReferenceQueue; | import java.lang.ref.WeakReference; | | public class Key | { | private static final ReferenceQueue refQueue = new ReferenceQueue(); | | private final KeyRef keyRef; | private final WeakReference<Annotation>[] annotations; | | public Key(Class clazz, Annotation[] annotations) | { | this.keyRef = new KeyRef(this, clazz, refQueue); | this.annotations = new WeakReference[annotations.length]; | for (int j = 0; j < annotations.length; j++) | { | this.annotations[j]= new WeakReference<Annotation>(annotations[j]); | } | } | | @Override | public boolean equals(Object obj) | { | if (obj instanceof Key) | { | Key other = (Key) obj; | Class clazz = keyRef.get(); | Class otherClass = other.keyRef.get(); | if (clazz == null || otherClass == null) | { | cleanKeyRefs(); | return false; | } | else | { | if (shouldCleanKeyRefs()) | cleanKeyRefs(); | return clazz.equals(otherClass); // need to add in annotations | } | } | | return false; | } | | /** Allows optimization; i.e. change to only poll the queue every X invocations */ | private boolean shouldCleanKeyRefs() | { | return true; | } | | /** | * Poll the reference queue and clear the strong ref to Key from all | * reference objects. This allows the Key to get removed from | * WeakHashMap. | */ | private void cleanKeyRefs() | { | Reference ref = null; | while ((ref = refQueue.poll()) != null) | { | ((KeyRef) ref).key = null; | } | } | | private static class KeyRef extends WeakReference<Class> | { | private Key key; | | KeyRef(Key key, Class clazz, ReferenceQueue queue) | { | super(clazz, queue); | this.key = key; | } | } | } View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152029#4152029 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152029

17 years, 10 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: Composite Map key and weak reference

by jason.greene＠jboss.com

So I take it the annotation refs point to annotations on the class ref? I could see a couple of options: * You could use a WeakIdentityHM using a Class ref as the key, and a nested normal HM using a key which has an array of weak annotation refs. This would then be wrapped in a delegate which emulates the composite key behavior. | | * Use a normal map, of any kind, and extend WeakReference to have an additional strong reference field, that points to your composite key class. Use this special reference for both the Class ref and all of the annotation refs. These are then registered in a ref queue, which is processed to remove gc'd entries periodically when accessing the map (by using the key reference). For this to work the key must do an identity check before comparing its contents (since the contents are gone during cleanup). If the map is concurrent you could do this cleanup in a separate thread. | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152028#4152028 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152028

17 years, 10 months

1
0
0 / 0

[Design of AOP on JBoss (Aspects/JBoss)] - Re: Optimizing Pointcut Matching

by flavia.rainone＠jboss.com

anonymous wrote : If you have time, then I think this will be a good thing to have. In this case, I'm going to implement it this week. Let's hope it improves the performance somewhat. anonymous wrote : I have expanded the PointcutStats class, which already did some classification, to classify things more finely. Thanks! That's the most difficult part of it. Plugging it in should be easy now. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152027#4152027 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152027

17 years, 10 months

1
0
0 / 0

[Design of JBoss ESB] - Re: Newbie problem

by ama1

Okay I found the problem. The maven packager build a file *.jboss-esb. This file gets interpretet by the esb-server but apparently in a wrong way. After changing the name to simply *.esb the stuff works. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4152022#4152022 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4152022

17 years, 10 months

1
0
0 / 0

2026

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums May 2008