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

[Design the new POJO MicroContainer] - Wildcards and JBossXb Builder

by alesj

Reading through Kabir's AOP work on jaxb-ing his metadata: "adrian(a)jboss.org" wrote : | You should let the MC schema define the {urn:jboss:bean-deployer:2.0}bean | and use it as a "wildcard". | I'm jaxb-ing our PolicyMetaData, having this in AbstractBindingMetaData: | @XmlTransient | public void setValue(ValueMetaData value) | { | this.value = value; | } | | @XmlAnyElement | public void setValueObject(Object value) | { | if (value == null) | setValue(null); | else if (value instanceof ValueMetaData) | setValue((ValueMetaData) value); | else | setValue(new AbstractValueMetaData(value)); | } | | @XmlValue | public void setValueString(String value) | { | if (value == null) | setValue(null); | else | { | ValueMetaData valueMetaData = getValue(); | if (valueMetaData instanceof StringValueMetaData) | { | ((StringValueMetaData) valueMetaData).setValue(value); | } | else | { | StringValueMetaData stringValue = new StringValueMetaData(value); | stringValue.setType(getType()); | setValue(stringValue); | } | } | } | and I'm getting this exception: | Caused by: org.jboss.xb.binding.JBossXBRuntimeException: {urn:jboss:bean-deployer:2.0}value not found as a child of {urn:jboss:policy:1.0}binding | I thought the @XmlAnyElement will handle this as a wildcard? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133637#4133637 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133637

17 years, 7 months

1
0
0 / 0

[Design of JBoss Build System] - Re: JBossAS CR1 build plan

by wolfc

You're forgetting jboss-as-server and the transitive dependencies. Which means we pull in almost all the AS components. I can send the dependency graph around again. :-) Aspects should be a library component. It has nothing to do with AS. They should be pojo aspects for use in any container environment. Maybe an extension component for MC? I'm not sure why we depend on the JCA implementation. We invoke the resource adapter through the MC. So the API should be enough (possibly some JCA API extentions). Iiop is an extension and could be separated out. It still requires an AS lib component though. Naming should be jnpserver (already fixed by Scott). I think the deployers and client launcher depend on jboss-as-server. Both should be moved from ejb3 to AS. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133627#4133627 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133627

17 years, 7 months

1
0
0 / 0

[Design of JBoss Remoting, Unified Invokers] - Re: A compilation error in Remoting 2

by trustin

doh! Thanks for the information. Then what do you think about replacing trunk with brahches/2.x? Is there any reason that we are keeping something dead? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133620#4133620 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133620

17 years, 7 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Restricting scanning for ejbs

by wolfc

The bug is that we're scanning the root of a war. We should only scan WEB-INF/classes (and lib jars). In JavaEE 6 this will become mandatory. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133616#4133616 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133616

17 years, 7 months

1
0
0 / 0

[Design of JBoss Remoting, Unified Invokers] - Re: A compilation error in Remoting 2

by david.lloyd＠jboss.com

You should be working from the "2.x" branch. The "trunk" branch is sort of dead at this point. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133614#4133614 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133614

17 years, 7 months

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: WARNING: Container project move

by adrian＠jboss.org

You can find a tag for the container project before this move here: https://svn.jboss.org/repos/jbossas/projects/microcontainer/tags/OLD_CONT... or anonymous https://anonsvn.jboss.org/repos/jbossas/projects/microcontainer/tags/OLD_... View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133611#4133611 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133611

17 years, 7 months

1
0
0 / 0

[Design the new POJO MicroContainer] - WARNING: Container project move

by adrian＠jboss.org

I've going to be moving the container project into two seperate subprojects (reflection and metadata repository) and releasing CR1s for them. Until those releases are done, I'll leave a copy of the container project in the MC trunk. Don't modify it, unless you want your changes to be lost. ;-) View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133609#4133609 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133609

17 years, 7 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Eclipse project settings in the ejb3 trunk

by wolfc

The .classpath files must not contain a reference to a workspace project, because a different version might be in use. So generating them with the aggregator is a no-go. I somewhat like the idea of using M2 Eclipse plugin but there are two cons: 1. slow startup 2. dependency resolution is different versus Maven builds Especially the second item makes it unusable. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133607#4133607 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133607

17 years, 7 months

1
0
0 / 0

[Design of POJO Server] - Re: Another startup time regression

by stale.pedersen＠jboss.org

i tested this on feb 27 builded from trunk. i updated head and rebuilded as5 now and tested it again (with and without the annotation changes). without annotation changes: 27s:162ms 27s:659ms 28s:46ms with annotation changes: 21s:835ms 22s:460ms 22s:751ms i profiled the bootup after the annotation changes and our primary problem atm is that for every new binding we rebuild the whole chain. flavia is working on a different implementation that would hopefully elimnate this or atleast reduce it significantly. this will be added to trunk within the next week. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133602#4133602 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133602

17 years, 7 months

1
0
0 / 0

[Design of POJO Server] - Re: Another startup time regression

by dimitris＠jboss.org

Stale, are you using the latest AS trunk to get those numbers? I'm asking because that includes the enabling of the jboss.vfs.forceNoCopy=true flag. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4133596#4133596 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4133596

17 years, 7 months

1
0
0 / 0

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums March 2008