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

[Design of OSGi Integration] - Re: Facade Questions

by johnbailey

"alesj" wrote : | Can you post javadoc? | Not that I don't trust you, but just lazy to do own lookup. ;-) | | public static final int INSTALLED | | The bundle is installed but not yet resolved. | | A bundle is in the INSTALLED state when it has been installed in the Framework but is not or cannot be resolved. | | This state is visible if the bundle's code dependencies are not resolved. The Framework may attempt to resolve an INSTALLED bundle's code dependencies and move the bundle to the RESOLVED state. | Should we thrown an Exception instead? From the possible states (UNINSTALLED,INSTALLED,RESOLVED, STARTING, STOPPING, ACTIVE), INSTALLED seems to be the default. I was origionally thinking UNINSTALLED, but it is valid only after the Bundle has been uninstalled. | public static final int UNINSTALLED | | The bundle is uninstalled and may not be used. | | The UNINSTALLED state is only visible after a bundle is uninstalled; the bundle is in an unusable state but references to the Bundle object may still be available and used for introspection. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131908#4131908 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131908

18 years, 1 month

1
0
0 / 0

[Design of JBoss Web Services] - Re: Signatures and CRs

by richard.opalka＠jboss.com

"alessio.soldano(a)jboss.com" wrote : | We might strip out CR from messages before signing but that would be a poor fix since thirdparty implementations could still produce signatures over messages containing CRs. | Stripping out the CR character from such elements violates the message integrity. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131907#4131907 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131907

18 years, 1 month

1
0
0 / 0

[Design of JBoss Web Services] - Re: Signatures and CRs

by richard.opalka＠jboss.com

I'm not 100 % sure but IMHO such enclosed 0xD character won't be converted to new line by the parser. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131906#4131906 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131906

18 years, 1 month

1
0
0 / 0

[Design of JBoss Web Services] - Re: Signatures and CRs

by richard.opalka＠jboss.com

"alessio.soldano(a)jboss.com" wrote : As you can read here http://www.w3.org/TR/REC-xml/#sec-line-ends CRs must be converted in line feeds when parsing. Xerces does this of course, thus the message bytes change and signature (which is obtained using the message containing the CR) is not valid anymore when processing the request message at server side. | >From my understanding this have nothing to do with our problem. The above specification says what to do with new-lines only. Our use-case is different. We have element containing sequence of characters enclosing 0xD character: | <some_element>asdf\rjklo</some_element> | View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131905#4131905 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131905

18 years, 1 month

1
0
0 / 0

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

by dimitris＠jboss.org

I had an interesting discussion with Brian, after the call. He made the very simple observation that until all EJB3 dependencies are identified and externalized (by introducing proper SPIs) what we are having now is a fake independent ejb3 project that complicates thing and doesn't add any real value. 1) Break ejb3 into multiple modules inside the AS codebase (i.e. as per their new ejb3 subprojects). That is, let ejb3 refactor itself into components the way they want. 2) For every other AS module EJB3 depends on, break out the portion that is API/SPI into a separate project like I did with the cluster module. Once that's done, ejb3's dependency on that module is gone. 3) Once each module dependency is gone, *then* ejb3 is really independent; that's when you pull the ejb3 modules out of the AS source tree into their own independent project. Or else, until the dependency graph for EJB3 is properly mavenized, the project should not move outside AS, causing all this trouble and jumping through hoops. It really boils down to priorities here (AS5 CR1 vs EJB3 externalization) in which case that looks like a sensible plan C to me. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131903#4131903 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131903

18 years, 1 month

1
0
0 / 0

[Design of JBoss ESB] - Timeboxed development schedule

by Kevin.Conner＠jboss.com

We are now entering a timeboxed release process. Starting from next week we will be adhering to the following schedule 3rd March - 14th March (2 weeks) JIRA issues will be assigned to the 4.3 release of the project. At this stage we will consider requests for new functionality. 17th March - 25th April (6 weeks) Development work continues through this period. We will only consider additional JIRA tasks if they are definitely bug fixes, new functionality will be assigned to the next release 28th April - 9 May (2 weeks) This period covers the final testing of the project, in readiness for the release. If there is anything you wish to see in the next version of the project then please let us know *before* March 14th. Thanks, Kev View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131902#4131902 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131902

18 years, 1 month

1
0
0 / 0

[Design of OSGi Integration] - Re: Facade Questions

by alesj

"johnbailey" wrote : Are the following resonable? | | | ControllerState controllerState = getControllerContext().getState(); | | if (ControllerState.ERROR.equals(controllerState)) | | { | | bundleState = Bundle.INSTALLED; // Seems strange, but see javadoc | | } | | | Can you post javadoc? Not that I don't trust you, but just lazy to do own lookup. ;-) Other look OK. And since the controller states to bundle states mapping is 'growing function', you can do a simple function which will determine bundle state according to the controller state index value. Hence not needing that List impl that I proposed earlier. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131901#4131901 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131901

18 years, 1 month

1
0
0 / 0

[Design the new POJO MicroContainer] - Re: SpecialValueFactoryUnitTestCase failure

by alesj

"adrian(a)jboss.org" wrote : | I get the same result, so it must be something at your end? | Uf, yup, it must be something here. :-( When I ran in IDEA it works. Running local 'mvn install' fails on VFSClassLoaderFactoryUnitTestCase (with the error I posted before). Running on my TeamCity build server again works OK. Nuking whole maven repo? Or just jbossxb? View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131899#4131899 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131899

18 years, 1 month

1
0
0 / 0

[Design of OSGi Integration] - Re: Javadocs and OO design

by alesj

"johnbailey" wrote : | Again, thanks for the guidance. I will put more effort into getting things cleaner before check-in. | NP. ;-) You can always hide all the small numerous classes into one public/protected one. OO + type safety, that's an important part, so that you can always refactor, while knowing it will still work. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131890#4131890 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131890

18 years, 1 month

1
0
0 / 0

[Design of OSGi Integration] - Re: Javadocs and OO design

by johnbailey

Ales- No problem. Bother me all you want. :). I need to get more comfortable with the project and all guidance is great. I will JAVADOC what has been done as well as all new work. I generally refactor a lot and normally that is done when I have flushed out the majority of the work. I think I committed this a bit early as I normally would flush a bit more out before putting it in. As for the if/else business, in the Bundle.getState, that is just a place holder until I get confirmation on the correct mappings. I have started to clean it up, but wanted to get confirmation first. Again, thanks for the guidance. I will put more effort into getting things cleaner before check-in. View the original post : http://www.jboss.com/index.html?module=bb&op=viewtopic&p=4131887#4131887 Reply to the post : http://www.jboss.com/index.html?module=bb&op=posting&mode=reply&p=4131887

18 years, 1 month

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 February 2008