January 2007 - jboss-dev-forums - Jboss List Archives

[Design of JBoss jBPM] - Re: releasing 3.2.Beta1

by david.lloyd＠jboss.com

"david.lloyd(a)jboss.com" wrote : Is there a reliable way to tell whether a file exists within a FileDefinition, short of catching the exception? ...and if not, can we add one? e.g. fileDefinition.containsFile(file) or something like that? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997262#3997262 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997262

17 years, 4 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: releasing 3.2.Beta1

by david.lloyd＠jboss.com

"koen.aers(a)jboss.com" wrote : A related problem remains however. This is when the webapp is used with a processdefinition with tasks for which there is no assiciated taskform. In this case the webapp should not try to render an unexisting form to show the task information. Is there a reliable way to tell whether a file exists within a FileDefinition, short of catching the exception? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997257#3997257 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997257

17 years, 4 months

1
0
0 / 0

[Design of AOP on JBoss (Aspects/JBoss)] - Re: Update AOPTestDelegate to better integrate into eclipse

by kabir.khan＠jboss.com

To set up for debug, you need to 1) set the jboss.aop.path system property to point to the jboss-aop.xml file 2) set -javaagent:%OUTPUT_LIB_FOLDER%\jboss-aop-jdk50.jar While the requirement for 1) could be optimized out by the mechanism you describe, it means rewriting all the tests since things like field, constructor and array interception need weaving of the testcase class, which in turn depends on the aop.xml file being deployed before the testcase class is loaded. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997239#3997239 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997239

17 years, 4 months

1
0
0 / 0

[Design of AOP on JBoss (Aspects/JBoss)] - Re: Array interception needed for jboss cache

by kabir.khan＠jboss.com

I have added the ability to have different advice chains when reading and writing arrays. The following snippet from the tests should explain it | <arraybind type="READ_WRITE"> | <advice aspect="org.jboss.test.aop.array.AspectForPrecedence" name="advice"/> | <interceptor-ref name="org.jboss.test.aop.array.TestArrayElementInterceptor"/> | </arraybind> | <arraybind type="READ_ONLY"> | <interceptor-ref name="org.jboss.test.aop.array.TestArrayElementReadInterceptor"/> | </arraybind> | <arraybind type="WRITE_ONLY"> | <interceptor-ref name="org.jboss.test.aop.array.TestArrayElementWriteInterceptor"/> | </arraybind> | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997221#3997221 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997221

17 years, 4 months

1
0
0 / 0

[Design of JBoss jBPM] - Re: releasing 3.2.Beta1

by koen.aers＠jboss.com

We have fixed the FileDefinition class so that it does not throw an error when looking for a file that does not exist. This solved part of the stack trace problem when using the webapp with a processdefinition without a forms.xml file. A related problem remains however. This is when the webapp is used with a processdefinition with tasks for which there is no assiciated taskform. In this case the webapp should not try to render an unexisting form to show the task information. We will nevertheless proceed with the beta1 release. Regards, Koen View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997210#3997210 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997210

17 years, 4 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Parse jar to detect ejb3 beans

by wolfc

With the EJB3 deployer rewrite you should fit snuggly in between the annation scanner deployer and the component deployer. Something like: - EJB3ParsingDeployer - EJB3JBossParsingDeployer - EJB3ScanningDeployer (after class loader) - JACCDeployer - EJB3Deployer http://jira.jboss.com/jira/browse/EJBTHREE-811 View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997205#3997205 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997205

17 years, 4 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Component Proposal - Configurable Service Locator

by wolfc

Please take a look at: http://anonsvn.jboss.org/repos/jbossas/projects/ejb3/trunk/injection/ I'm rewriting the injection framework there. Note that the JavaEE 5 client specifications already solve most of the problems you describe. With the new clients you can do: @EJB MySession session; to get a reference without JNDI calls. I don't know about the caching part. It's easier to have every injection do a JNDI call, then try to differentiate which calls won't create a new object and which calls do. The usual client application will most likely cache SLSB references anyway. A public ServiceLocater might be nice to get references to EJB's based on interface. With the last item you mean: having the client use services located on difference hosts? That would be cool. It's in violation with JavaEE 5 spec though so we need to clearly separate it from a regular JavaEE 5 client. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997203#3997203 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997203

17 years, 4 months

1
0
0 / 0

[Design of JBoss Internal Benchmarking] - Re: JBoss 4.0.4 + MySQL 5.0.17c Pro + SpecJ2004

by mahendra_kutare

I have also encountered the similar program and somewhere in the jboss code it checks whether its a single phase or two phase commit and if its two phase commit then it set some STATUS on prepare and commit. I believe may be the reason for this is JBOSS is not able to handle the 2PC due to - a) MySQL version of j-connector which do not support 2PC. Anything earlier than 5.0 would not be have 2PC commit capabilities. I will be performing tests runs to cross check if this is the case. If anyone has any other view or explaination of this error, please let me know. - Mahendra View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997188#3997188 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997188

17 years, 4 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Parse jar to detect ejb3 beans

by anil.saldhana＠jboss.com

As long as I am able to register the policy configuration for the EJB3 deployment before getting active, it should be fine. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997153#3997153 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997153

17 years, 4 months

1
0
0 / 0

[Design of EJB 3.0] - Re: Parse jar to detect ejb3 beans

by bill.burke＠jboss.com

I wouldn't suggest doing it this way.... What you should do is have your JACC deployer come after the EJB3 deployers and look for: | | public void deploy(DeploymentUnit unit) throws DeploymentException | { | Ejb3Deployment deployment = unit.getAttachment(Ejb3Deployment.class); | if (deployment == null) return; // there is no EJB3 deployment in this file | Scanning is time consuming. We don't want 50 deployers doing scanning. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=3997140#3997140 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=3997140

17 years, 4 months

1
0
0 / 0

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

jboss-dev-forums January 2007