June 2009 - jboss-dev-forums - Jboss List Archives

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Server-side browser refactoring

by jmesnil

sorry clebert, I pasted the wrong link. the patch is attached to https://jira.jboss.org/jira/browse/JBMESSAGING-1437 View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239127#4239127 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239127

16 years, 9 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Server-side browser refactoring

by clebert.suconic＠jboss.com

Jeff, I don' t see any files there? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239124#4239124 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239124

16 years, 9 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: Exception while deploying console from HEAD

by kukeltje

I've had this in eclipse and had to change a the 'Forbidden Reference' setting in the 'Deprecated and Restricted API' section of the Java Compiler settings from error to warning. But in my case it was only a compiletime error. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239121#4239121 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239121

16 years, 9 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: Why is the console a different repos

by kukeltje

http://blog.athico.com/2009/05/drools-flow-51-preview-1-web-console.html View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239120#4239120 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239120

16 years, 9 months

1
0
0 / 0

[Design of JBoss Build System] - Re: Incremental changes and building AS trunk

by pgier

The distribution build uses the Ant copy task to copy jar files from the local maven repository (~/.m2/repository). It's possible that if the file copied in the first build had a newer timestamp than the newer version jar in the local maven repo, that ant would skip the copy. I'm not sure if this is the cause of what you saw, but it seems possible. I'm not sure what is the best way to fix this. I could set the copy tasks to overwrite=true, to force a copy every time, but this might slow down incremental builds a little bit. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239103#4239103 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239103

16 years, 9 months

1
0
0 / 0

[Design of Messaging on JBoss (Messaging/JBoss)] - Re: Server-side browser refactoring

by jmesnil

I've added code to iterate on the queue when the consumer is a browser. Bulk of the patch is in ServerConsumerImpl.promptDelivery() However, there are now failures in LargeMessageTest when a browser is created to browse a queue with large messages. Clebert, I'm not familiar with this part of the code. Could you have a look at the patch attached to https://jira.jboss.org/jira/browse/JBMESSAGING-1432 to help me determine how to make the browser work for both regular and large messages? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239082#4239082 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239082

16 years, 9 months

1
0
0 / 0

[Design of JBoss Build System] - Incremental changes and building AS trunk

by jaikiran

I was recently trying a few things on AS trunk which now has the Maven builds. Initially i triggered a mvn clean install and generated the AS jboss6Alpha. Then while working on a few changes, i updated the versions of some components in the component-matrix pom.xml and triggered a mvn install However the AS jboss6Alpha still contained the old versions of the components. I had to delete the entire output/jboss6Alpha (or i think run a mvn clean install, don't remember which of these two fixed it) for the changes to be reflected. Is this expected? View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239073#4239073 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239073

16 years, 9 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: ClassPool for JBoss Reflection

by kabir.khan＠jboss.com

"flavia.rainone(a)jboss.com" wrote : | *** I mean, when you do toClass, the class is already in the cache by now. Plus, what if somebody is doing a toClass with a class that is in the softcache? The lockInCache call will force the class being added as a hard reference to the "classes" cache, thus ruining the soften feature. I'm not really sure what you mean here, but toClass() is called when creating a new class so does that not need to be in the hard cache? If only in the softcache, when it is removed and somebody needs to obtain the CtClass again we would need to load it from its resource. Although in AS that would probably be fine since dynamically created classes have their bytes written to a temporary location (tempURL in AS 4, vfsmemory url in AS 5) View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239064#4239064 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239064

16 years, 9 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: ClassPool for JBoss Reflection

by kabir.khan＠jboss.com

Historically, the ScopedClassPool is there to extract out the known behaviour at the time (early 2006 IIRC) of how the UnifiedClassLoaders (AS 4) worked. It used to all live in AOPClassPool, but since JBoss Retro also needed to do loadtime weaving stuff the common code was moved to ScopedClassPool. 1) The comment means that we should use the classloader associated with the pool rather than the default thread context classloader that would get passed in when somebody calls ClassPool.toClass(). 2) get0() is an internal implementation detail in the pools, so getLocally() is there to expose the functionality. Also, it ensures that we just search the pool itself, since at weaving time we know the exact classloader (passed in by the agent). 3) If I do a search for who is calling ClassPool.get() I get loads of callers, so we are using that in AOP. Also, when you load up a CtClass, get a CtField or a CtMethod parameter type I think javassist will call get() internally to obtain that. So as mentioned in 2 when weaving a class we use the pool for the classloader used to load the CtClass since we know it must come from there, but later in the weaving process we end up in get0() when fetching in the superclass, fields, methods, parameters etc. 4) Probably just to get the extra logging. javassist ScopedClassPool could be fixed I guess, but it does not use log4j so there is not really any way to add logging there. 6) I don't think it is a memory leak, since the CtClasses need to be cached, and in AS we create/drop the pools when the classloaders are deployed/undeployed. It might be redundant as you say though. 5&7) I agree with you here, but a bunch of stuff in the API is using the ScopedClassPool(Factory/Repository) stuff. I was in two minds at the time whether to get rid of it, but was worried about breaking backwards compatibility. But yes, for the new delegating classpools a lot of the stuff (the searching across classpools) in ScopedClassPool is redundant. However, the repository is still used when registering and unregistering classloaders. Also, in AS 4 we do not use the new delegating classpools, but the old ones returned by org.jboss.aop.asintegration.jboss4.JBossClassPoolFactory and org.jboss.aop.asintegration.jboss5.JBoss5ClassPoolFactory. 8) ClassPool.get() is being used, see 2& 3. View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239057#4239057 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239057

16 years, 9 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: New console snapshot published, Task Forms included

by heiko.braun＠jboss.com

anonymous wrote : | As the old saying goes, a layer of indirection solves every problem in computer science. | View the original post : http://www.jboss.org/index.html?module=bb&op=viewtopic&p=4239041#4239041 Reply to the post : http://www.jboss.org/index.html?module=bb&op=posting&mode=reply&p=4239041

16 years, 9 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 June 2009