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

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: 1.4.1.GA ready

by clebert.suconic＠jboss.com

"timfox" wrote : "clebert.suconic(a)jboss.com" wrote : | | | | Anil... are you sure this won't affect performance? | | We must do a before and after performance check before applying those changes. If it's detrimental to performance it's a no go and we'll have to find some other way. IMO all we need is to make sure we aways test for System.getSecurityManager() != null before creating the Priviledged block. Then we know that users who care about performance won't be affected, and users who care about security on the VM will assume some little impact. It's aways about managing expectations, right? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190944#4190944 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190944

17 years, 4 months

1
0
0 / 0

[Design of POJO Server] - Re: VFS Permissions - JBMICROCONT-149

by david.lloyd＠jboss.com

No, I disagree entirely. Actually it's fine if it's a VFS URL. We've already gotten past that problem - we can specify VFS URLs in the security policy file now without any problems. The problem now is that vfsmemory URLs are not predictable. And in fact -any- VFS URL type might not have a corresponding "real" URL; I don't think this is a reliable solution. I think the best solution is to have predictable/repeatable names for VFS URLs. I don't think using a UUID is really useful in practice anyway. Why not allow a name to be assigned to vfsmemory URLs? Then one could simply put the appropriate vfsmemory URL in the security policy file. If security is a concern, one could simply introduce a permission that is necessary in order to create a vfsmemory URL with a given name. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190941#4190941 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190941

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by adrian＠jboss.org

"adrian(a)jboss.org" wrote : | But for more transient bean classes, e.g. those used only in the managed object layer | it might not be as good? | Also for the generic types, it will depend how the JDK's reflection framework holds references to those objects. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190940#4190940 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190940

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by adrian＠jboss.org

"alesj" wrote : "alesj" wrote : | | What about if I just change that the key is ClassInfo instead of String? | | (I've done that locally and all the tests pass) | Actually I think I can drop the whole ClassLoader as key, | since ClassInfo is per ClassLoader if I'm not mistaken. | But you still have the problem that ClassInfos can be transient. Let's go with that, but we need to do so extra work to see whether all this caching is (still) effective. I think in practice it will work for most of the ClassInfos we use to create beans since the WeakTypeCache will hold a reference to the TypeInfo until the bean's class is GCed. But for more transient bean classes, e.g. those used only in the managed object layer it might not be as good? If only there was a combination Weak/SoftReference mechanism which let you do SoftReference behaviour until the classloader gets GCed then it turns into WeakReference behaviour. ;-) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190936#4190936 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190936

17 years, 4 months

1
0
0 / 0

[Design of Security on JBoss] - Re: PrivilegedBlock location

by clebert.suconic＠jboss.com

Flavia, are you doing a check like? | if (System.getSecurityManager() == null) | { | cl = clazz.getClassLoader(); | } | else | { | ..... security block | } | If you are.. all we need on JBM is an AOP release. And this won't affect weaved code, right? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190932#4190932 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190932

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by adrian＠jboss.org

I don't know how to make the test fail, but the following change to the test shows the problem happening. | Index: src/test/java/org/jboss/test/metatype/values/factory/test/UnwrapValueUnitTestCase.java | =================================================================== | --- src/test/java/org/jboss/test/metatype/values/factory/test/UnwrapValueUnitTestCase.java (revision 81358) | +++ src/test/java/org/jboss/test/metatype/values/factory/test/UnwrapValueUnitTestCase.java (working copy) | @@ -35,7 +35,10 @@ | | import junit.framework.Test; | | +import org.jboss.beans.info.spi.BeanInfo; | +import org.jboss.config.plugins.property.PropertyConfiguration; | import org.jboss.metatype.api.values.MetaValue; | +import org.jboss.reflect.spi.ClassInfo; | import org.jboss.test.metatype.values.factory.support.SimpleCompositeInterface; | import org.jboss.test.metatype.values.factory.support.TestEnum; | import org.jboss.test.metatype.values.factory.support.TestGeneric; | @@ -103,9 +106,21 @@ | | checkSingle(new TestSimpleCompositeInterface("something"), SimpleCompositeInterface.class); | } | + PropertyConfiguration config = new PropertyConfiguration(); | | + private void printCollection(Type type) | + { | + ClassInfo typeInfo = (ClassInfo) config.getTypeInfo(type); | + System.out.println(typeInfo.getName() + " " + typeInfo.getComponentType()); | + BeanInfo beanInfo = config.getBeanInfo(typeInfo); | + typeInfo = beanInfo.getClassInfo(); | + System.out.println(typeInfo.getName() + " " + typeInfo.getComponentType()); | + } | + | public void testCollectionUnwrap() throws Exception | { | + printCollection(getType("Integer", Set.class)); | + printCollection(getType("Enum", Set.class)); | Integer i1 = 123; | Integer i2 = 123; | checkCollection(new ArrayList<Integer>(), i1, i2); | which produces output (the last line is wrong) | java.util.Set java.lang.Integer | java.util.Set java.lang.Integer | java.util.Set EnumInfoImpl(a)1df280b{name=org.jboss.test.metatype.values.factory.support.TestEnum} | java.util.Set java.lang.Integer | So it looks like the BeanInfo caching is the problem. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190928#4190928 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190928

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by alesj

"adrian(a)jboss.org" wrote : | Or if it is going to use that, then ParameterizedClassInfo needs to override getName() | to return the generic name (but I'm not sure what that would break?). | This | +++ src/main/java/org/jboss/reflect/plugins/introspection/ParameterizedClassInfo.java | @@ -77,6 +77,12 @@ | this.parameterizedType = parameterizedType; | } | | + @Override | + public String getName() | + { | + return parameterizedType.toString(); | + } | breaks this | testGenericSuperClassString(org.jboss.test.classinfo.test.IntrospectionGenericClassUnitTestCase) | testGenericSuperClassEmptyClass(org.jboss.test.classinfo.test.IntrospectionGenericClassUnitTestCase) | testGenericSuperInterfaceString(org.jboss.test.classinfo.test.IntrospectionGenericClassUnitTestCase) | testGenericSuperInterfaceEmptyClass(org.jboss.test.classinfo.test.IntrospectionGenericClassUnitTestCase) | :-) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190919#4190919 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190919

17 years, 4 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Why ServerSessionImpl::doHandleCreateQueue is not callin

by clebert.suconic＠jboss.com

I need to get a list of valid Addresses in order to start the PageStores. I was expecting Queues to be listed on the destinations, but they weren't. So, to get a list of Addresses, I needed: to interact over Destinations. Then interact over Bindings to find the core-queues that didn't have Destination. Then I had the list of valid queues. That's why I actually started this thread.. I needed to go over all the bindings to extra a list of valid addresses. IMO the CoreQueue should aways have a destination like the JMS queues. Then I would just need to interact with Destinations. If we conclude this is the way it should be, I will just remove the TODO and leave the code as it is. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190916#4190916 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190916

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by adrian＠jboss.org

"alesj" wrote : "adrian(a)jboss.org" wrote : | | Clearly, the caching in AbstractBeanInfoFactory shouldn't be using | | classInfo.getName() in its caching. | | | What about if I just change that the key is ClassInfo instead of String? | (I've done that locally and all the tests pass) But it wouldn't be much of a cache since the ClassInfo's can be pretty transient. Also I think it will cause a memory leak. You effectively have: WeakValueHashMap<ClassInfo, BeanInfo> but the ClassInfo could itself get a hard reference on the BeanInfo via one of its "attachments", e.g. the caching of SchemaBindings in jbossxb against the TypeInfo. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190912#4190912 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190912

17 years, 4 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: UnwrapValueUnitTestCase.testCollectionUnwrap failure on

by alesj

"alesj" wrote : | What about if I just change that the key is ClassInfo instead of String? | (I've done that locally and all the tests pass) Actually I think I can drop the whole ClassLoader as key, since ClassInfo is per ClassLoader if I'm not mistaken. | Index: src/main/java/org/jboss/beans/info/plugins/AbstractBeanInfoFactory.java | =================================================================== | --- src/main/java/org/jboss/beans/info/plugins/AbstractBeanInfoFactory.java (revision 79796) | +++ src/main/java/org/jboss/beans/info/plugins/AbstractBeanInfoFactory.java (working copy) | @@ -55,7 +55,7 @@ | public class AbstractBeanInfoFactory implements BeanInfoFactory | { | /** The cache */ | - protected Map<ClassLoader, Map<String, Map<BeanAccessMode, BeanInfo>>> cache = new WeakHashMap<ClassLoader, Map<String, Map<BeanAccessMode, BeanInfo>>>(); | + protected Map<ClassInfo, Map<BeanAccessMode, BeanInfo>> cache = new WeakHashMap<ClassInfo, Map<BeanAccessMode, BeanInfo>>(); | | protected static boolean isGetter(MethodInfo minfo) | { | @@ -133,20 +133,14 @@ | | synchronized (cache) | { | - ClassLoader cl = classAdapter.getClassLoader(); | ClassInfo classInfo = classAdapter.getClassInfo(); | - String className = classInfo.getName(); | - Map<String, Map<BeanAccessMode, BeanInfo>> map = cache.get(cl); | - Map<BeanAccessMode, BeanInfo> modeMap = null; | - if (map != null) | + //String className = classInfo.getName(); | + Map<BeanAccessMode, BeanInfo> modeMap = cache.get(classInfo); | + if (modeMap != null) | { | - modeMap = map.get(className); | - if (modeMap != null) | - { | - BeanInfo info = modeMap.get(accessMode); | - if (info != null) | - return info; | - } | + BeanInfo info = modeMap.get(accessMode); | + if (info != null) | + return info; | } | | Set<ConstructorInfo> constructors = getConstructors(classInfo); | @@ -159,16 +153,9 @@ | Set<EventInfo> events = getEvents(classInfo); | | BeanInfo result = createBeanInfo(classAdapter, accessMode, properties, constructors, methods, events); | - if (map == null) | - { | - map = new WeakValueHashMap<String, Map<BeanAccessMode, BeanInfo>>(); | - cache.put(cl, map); | - } | if (modeMap == null) | - { | modeMap = new WeakValueHashMap<BeanAccessMode, BeanInfo>(); | - map.put(className, modeMap); | - } | + | modeMap.put(accessMode, result); | return result; | } | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4190911#4190911 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4190911

17 years, 4 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 November 2008