[00:06:44] *** ldimaggi has quit IRC
[00:14:15] <aslak> very odd.. it's being reset somewhere..  trying to figure it out
[00:18:54] <aslak> mhuniewicz, aa, it's look for it on BeanDeploymentArchive level, not Deployment
[00:19:36] <mhuniewicz> aslak, what is?
[00:19:42] <aslak> mhuniewicz, for(BeanDeploymentArchive beanArchive: container.getDeployment().getBeanDeploymentArchives()) { beanArchive.getServices().add(...) }
[00:20:45] <aslak> mhuniewicz, weld internally during startContainer
[00:27:09] <mhuniewicz> aslak, yep, this works.
[00:29:26] <mhuniewicz> aslak, how do I configure my tests now? :)
[00:30:44] <aslak> mhuniewicz, what do you mean ?
[00:30:57] <mhuniewicz> I need to pass true somehow.
[00:32:37] <aslak> you have that in the WeldEEMockConfiguration
[00:32:46] <aslak> coming in from arquillian.xml
[00:36:30] <mhuniewicz> aslak, yeah, it works, thanks. Pushing onto my branch.
[00:37:42] *** nilian has quit IRC
[00:40:21] <mhuniewicz> aslak, I guess what is left is making sure that persistence.xml doesn't leak inside?
[00:41:47] *** jharting has quit IRC
[00:50:19] *** alesj has quit IRC
[00:57:03] <aslak> mhuniewicz, yea
[00:57:09] <aslak> bed time.. :)
[00:57:11] *** aslak has quit IRC
[00:58:08] *** mhuniewicz has quit IRC
[01:25:47] *** kevinpollet has quit IRC
[01:47:03] *** rruss has quit IRC
[02:15:49] *** rruss has joined #jbosstesting
[02:33:26] *** jamezp is now known as jamezp_afk
[03:01:46] *** ianbrandt has quit IRC
[03:10:26] *** rruss has quit IRC
[03:43:54] *** ldimaggi has joined #jbosstesting
[05:53:15] *** ldimaggi has quit IRC
[05:57:49] *** jamezp_afk has quit IRC
[05:58:27] *** jamezp_afk has joined #jbosstesting
[06:21:37] *** rruss has joined #jbosstesting
[06:57:26] *** tombee has quit IRC
[07:51:48] *** timte has joined #jbosstesting
[08:12:31] *** ge0ffrey has joined #jbosstesting
[08:17:45] *** kpiwko has joined #jbosstesting
[08:21:59] <kpiwko> Hi ALR, are you still there?
[08:22:08] <ALR> kpiwko: Sure.
[08:22:20] <ALR> Having a late night on some of your issues :)  ARQ + AS7.
[08:22:23] <kpiwko> ALR: wow, what's the time at your place?
[08:22:30] <ALR> kpiwko: 2:20A
[08:22:52] <kpiwko> ALR: uff...I have a quick question about SHRINKWRAP-244
[08:22:53] <jbossbot> jira [SHRINKWRAP-244] When calling JavaArchiveImpl.addClass(String, ClassLoader) the TCCL gets used instead [Closed (Done) Bug, Major, Davide D'Alto] https://issues.jboss.org/browse/SHRINKWRAP-244
[08:23:03] <ALR> kpiwko: Shoot
[08:23:12] <kpiwko> ALR: it seems to me that this patch had made a regression
[08:23:29] <kpiwko> now I'm not able to include classes load by Bootstrap classloader
[08:23:50] <kpiwko> because class.getClassloader() will return null
[08:23:53] <ALR> kpiwko: Hehe, something untested isn't really a regression :)
[08:24:05] <ALR> kpiwko: Got a test case?
[08:24:13] <kpiwko> I used it to generate SWR test environment
[08:24:26] <kpiwko> sure, let me find the link
[08:24:30] <ALR> kpiwko: Fantastic
[08:24:41] <ALR> Let's see if we can't fix this up without affecting anything else.
[08:25:25] <kpiwko> ALR: this one will do, for example
[08:25:26] <kpiwko> https://github.com/kpiwko/resolver/blob/master/impl-maven/src/test/java/org/jboss/shrinkwrap/resolver/util/JarGenerator.java
[08:25:38] <kpiwko> WarGenerator.java having likely the same issue
[08:26:54] <ALR> kpiwko: addClasses(content) ... failing?
[08:27:00] <kpiwko> yep
[08:27:15] <ALR> ie. Integer.class can't be found because it's going to Bootstrap CL?
[08:28:03] <kpiwko> ALR: exactly
[08:28:07] <kpiwko> here's the problem https://github.com/shrinkwrap/shrinkwrap/blob/master/impl-base/src/main/java/org/jboss/shrinkwrap/impl/base/container/ContainerBase.java#L1183
[08:28:16] <ALR> kpiwko: Give me a few moments.
[08:28:21] <kpiwko> this method can return null to signalize bootstrap classloader
[08:41:40] <ALR> I know :)
[08:43:43] *** aslak has joined #jbosstesting
[08:54:10] <ALR> kpiwko: https://github.com/shrinkwrap/shrinkwrap/commit/437ab0e41d3d69ff9a1ebb5e9d1ed7bdff3cdbc1
[08:54:11] <jbossbot> git [shrinkwrap] 437ab0e.. Andrew Lee Rubinger [SHRINKWRAP-335] Allow ClassContainer.addClass(Class) to accept classes loaded from the Bootstrap CL
[08:54:13] <jbossbot> jira [SHRINKWRAP-335] Classes from the Bootstrap CL can not be added to ClassContainers [Resolved (Done) Bug, Critical, Andrew Rubinger] https://issues.jboss.org/browse/SHRINKWRAP-335
[08:55:06] <kpiwko> ALR: awesome, thanks!
[08:55:13] <ALR> kpiwko: My pleasure.  Good find.
[08:56:07] <lightguard_jp> aslak: ping
[08:57:24] <ALR> kpiwko: Holy shit, we haven't done a SW release since July.
[08:57:28] *** mgoldmann has joined #jbosstesting
[08:57:29] <ALR> kpiwko: I'm going to do one now.
[08:57:36] <lightguard_jp> Anyone familiar with using arquillian-jacoco?
[08:58:19] <kpiwko> ALR: this is rather good, isn't it? it means it's pretty stable :)
[08:58:43] <ALR> kpiwko: Oh, it's excellent.  We rarely get real bug reports.
[08:59:57] *** maschmid has joined #jbosstesting
[09:00:30] <lightguard_jp> maschmid: Have you tried arquillian-jacoco?
[09:01:25] <maschmid> lightguard_jp: yes
[09:01:33] <lightguard_jp> How do I set it up?
[09:03:03] <aslak> ALR, does it include the fixes the maschmid did for jacoco ?
[09:03:26] <ALR> aslak: Does what?  New SW release?
[09:03:29] <maschmid> lightguard_jp: There is still ARQ-599 blocking us, waiting for new SW release
[09:03:29] <aslak> yea
[09:03:30] <jbossbot> jira [ARQ-599] ApplicationArchiveInstrumenter should also instrument classes inside JARs [Open (Unresolved) Feature Request, Major, Unassigned] https://issues.jboss.org/browse/ARQ-599
[09:03:59] <lightguard_jp> Okay, I'll wait for it then.
[09:04:09] <ALR> aslak: https://issues.jboss.org/secure/ReleaseNote.jspa?projectId=12310884&version=12317450
[09:05:07] <maschmid> lightguard_jp: I have found Jose's blog quite useful http://techblog.joserodolfo.com/2011/07/arquillian-coverage-tests-reported-with-sonar-howto/  to make it work for our testsuite structure, the only additional thing you have to do is copy the /classes from all the sub-modules, so the jacoco report knows for what classes it should generate the report
[09:05:13] <ALR> maschmid: I don't see a link on 599 pointing to any SW issue?
[09:05:31] <ALR> What fixes are we talking about?
[09:05:53] <maschmid> ALR: SHRINKWRAP-331
[09:05:54] <jbossbot> jira [SHRINKWRAP-331] Add ArchiveAsset to the API [Resolved (Done) Feature Request, Major, Marek Schmidt] https://issues.jboss.org/browse/SHRINKWRAP-331
[09:05:55] <aslak> SHRINKWRAP-332 was one of them atleast
[09:05:56] <jbossbot> jira [SHRINKWRAP-332] Add ClassLoaderAsset and ClassAsset to the API [Closed (Done) Feature Request, Major, Marek Schmidt] https://issues.jboss.org/browse/SHRINKWRAP-332
[09:06:08] <aslak> 331 the other..
[09:06:10] <aslak> :)
[09:06:18] <ALR> Those are both released in 1.0.0-beta-6
[09:06:33] <aslak> 331 is missing fixed for.. :)
[09:06:36] <ALR> Wait
[09:06:38] <ALR> One sec
[09:07:42] <ALR> OK, now reflected in the release notes.
[09:07:55] <ALR> SHRINKWRAP-331
[09:07:56] <jbossbot> jira [SHRINKWRAP-331] Add ArchiveAsset to the API [Closed (Done) Feature Request, Major, Marek Schmidt] https://issues.jboss.org/browse/SHRINKWRAP-331
[09:08:02] <ALR> Much better.  Closed. :)
[09:08:09] <aslak> excellent
[09:08:14] <maschmid> good, I'll try to update the patch then...
[09:08:20] <ALR> Anything else I can help you boys with, then?
[09:08:32] <aslak> maschmid, great, i'll pull upstream and do Alpha2 when done.. :)
[09:08:53] <ALR> aslak, maschmid: I didn't know you were waiting on a release.  Next time let me know.
[09:09:16] *** lfryc has joined #jbosstesting
[09:13:21] <lightguard_jp> maschmid: I also have catch updated to the new testing structure
[09:13:35] <lightguard_jp> maschmid: The versions of arquillian are CR4 instead of CR2
[09:16:12] <aslak> lightguard_jp, what about CR5 ? :)
[09:16:21] <aslak> that's the future
[09:16:22] <aslak> hehe
[09:16:52] <lightguard_jp> Didn't know about CR5
[09:16:56] <lightguard_jp> breaking changes?
[09:17:37] <aslak> https://issues.jboss.org/secure/ConfigureReport.jspa?atl_token=AQZJ-FV3A-N91S-UDEU%7C666347bb8f6ddf669aca4b407151300148050a13%7Clin&versions=12318105&sections=all&style=html&selectedProjectId=12310885&reportKey=org.jboss.labs.jira.plugin.release-notes-report-plugin%3Areleasenotes&Next=Next
[09:18:40] *** jharting has joined #jbosstesting
[09:18:43] <aslak> lightguard_jp, maybe some stuff around descriptors.. depends on what you use in your code i guess (upgraded to a stable spi version)
[09:19:17] <aslak> new features are "allow a group to be defined as default in arq.xml" and System properties replacement in arq.xml
[09:24:13] <maschmid> lightguard_jp: Don't we already have CR4 in current seam parent? (well, 15-SNAPSHOT)
[09:25:21] *** oskutka has joined #jbosstesting
[09:25:58] *** maeste has joined #jbosstesting
[09:26:00] <lightguard_jp> 15-SNAPSHOT is, as you stated
[09:26:16] <lightguard_jp> But I'm using 14 in catch until 15 is released
[09:26:25] <lightguard_jp> That should probably be something to discuss
[09:26:35] <lightguard_jp> There are many moving parts :(
[09:26:53] *** rruss has quit IRC
[09:33:03] <aslak> ALR, cool, ARQ-607 and ARQ-49 upstream
[09:33:05] <jbossbot> jira [ARQ-607] Upgrade to ShrinkWrap 1.0.0-beta-6 [Resolved (Done) Component  Upgrade, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-607
[09:33:06] <jbossbot> jira [ARQ-49] Arquillian depends upon ShrinkWrap internals [Resolved (Done) Task, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-49
[09:34:24] <ALR> aslak: Looks like no POM change to put shrinkwrap-impl-base in "test" or "provided" scope though? https://github.com/arquillian/arquillian-core/commit/f33410e426037e0121742f9fa09b330416d2c377
[09:34:25] <jbossbot> git [arquillian-core] f33410e.. Aslak Knutsen ARQ-607 ARQ-49 Upgrade to ShrinkWrap 1.0.0-beta-6, this release moves more of the Asset impls to API and fixes Arquillian dependencies on ShrinkWrap impl
[09:34:25] <jbossbot> jira [ARQ-607] Upgrade to ShrinkWrap 1.0.0-beta-6 [Resolved (Done) Component  Upgrade, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-607
[09:34:26] <jbossbot> jira [ARQ-49] Arquillian depends upon ShrinkWrap internals [Resolved (Done) Task, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-49
[09:34:41] <aslak> maschmid, you didn't sign a CLA already did you ?
[09:34:58] <maschmid> aslak: I am a RedHat employee
[09:35:30] <aslak> maschmid, aa yea, look at that.. :)
[09:36:37] <aslak> maschmid, we already own your ass then..  hehe
[09:37:55] <aslak> aa, no sorry, mixing you with mhuniewicz
[09:38:32] <aslak> seem i need to expand my nick memory block with more then one byte
[09:38:37] *** lightguard_jp has quit IRC
[09:39:04] <ALR> pnwed
[09:39:26] *** lightguard_jp has joined #jbosstesting
[09:40:14] <maschmid> aslak: yeah, well, not everybody can have such an easily recognized name as yours... :)
[09:43:13] <aslak> ALR, t's already in provided scope
[09:43:30] <ALR> aslak: Hmm, should be "test"
[09:43:40] <aslak> yea
[09:43:40] <ALR> aslak: So that you can't in the future compile against it
[09:43:58] <aslak> good catch
[09:44:04] <ALR> maschmid: Easily recognized?  Dude we called him "Aslan" today.
[09:44:57] <ALR> aslak: Yah, I catch mad dependency issues.
[09:45:24] <aslak> ALR, to Rayme's defense.. he's never gotten it right
[09:45:35] <ALR> aslak: That's a really poor defense.
[09:45:39] <aslak> :)
[09:45:44] <ALR> Especially for a guy named "Rayme".
[09:45:46] <ALR> Haha
[09:48:04] <aslak> ALR, there we go, https://github.com/arquillian/arquillian-core/commit/2e596ede562fdabdea4b8acbc9bf6fa7e0f2796e
[09:48:04] <jbossbot> git [arquillian-core] 2e596ed.. Aslak Knutsen ARQ-49 Move ShrinkWrap Impl dependency to test scope, only needed runtime.
[09:48:05] <jbossbot> jira [ARQ-49] Arquillian depends upon ShrinkWrap internals [Resolved (Done) Task, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-49
[09:48:13] <ALR> :D
[10:01:43] *** rruss has joined #jbosstesting
[10:03:56] *** rachmatowicz has joined #jbosstesting
[10:10:54] *** tombee has joined #jbosstesting
[10:10:55] *** tombee has joined #jbosstesting
[10:10:55] *** tombee has joined #jbosstesting
[10:13:38] *** lightguard_jp has quit IRC
[10:20:15] *** jeand__ has joined #jbosstesting
[10:23:11] *** ALR has quit IRC
[10:27:55] *** rruss has quit IRC
[10:31:55] *** stuartdouglas has quit IRC
[10:33:55] *** alesj has joined #jbosstesting
[10:35:56] *** stuartdouglas has joined #jbosstesting
[10:36:18] *** jeand_ has joined #jbosstesting
[10:37:42] *** jharting has quit IRC
[10:38:25] *** jharting has joined #jbosstesting
[10:39:36] *** jeand__ has quit IRC
[10:45:20] *** jharting has quit IRC
[10:46:00] *** jharting has joined #jbosstesting
[11:06:38] <kpiwko> ge0ffrey: ping
[11:06:46] <ge0ffrey> kpiwko: good morning
[11:06:46] <kpiwko> ge0ffrey: hi
[11:07:15] <ge0ffrey> kpiwko: tnx for working on the maven @deployment thing :)
[11:07:33] <kpiwko> ge0ffrey: Hi, I'm just wondering how come nobody noticed sooner...by default war plugin is executed in package phase, so output is not available during tests
[11:07:59] <ge0ffrey> damn, you're right
[11:08:35] <ge0ffrey> although all the generated files are there then, it's just not like that
[11:08:52] <ge0ffrey> kpiwko: there is an extra phase though, called "test-integration" or something
[11:08:58] <ge0ffrey> which is called after the package phase
[11:09:04] <kpiwko> right, integration test will do
[11:09:21] <kpiwko> but it requires modification of the samples
[11:09:32] <ge0ffrey> I didn't notice it myself, because I am running the tests from intellij on my fork atm
[11:09:36] <ge0ffrey> don't have a hudson set to my fork yet
[11:10:01] <kpiwko> I was pretty sure that in package phase result is only collected
[11:10:12] <ge0ffrey> kpiwko: if they are integretions test, is it just a matter of changing the src dir from src/test/main or something?
[11:10:38] <kpiwko> ge0ffrey: well, it is sufficient to change the default phase
[11:11:02] *** Rostifant has joined #jbosstesting
[11:11:12] <Rostifant> hi @all
[11:11:24] <ge0ffrey> kpiwko: yea, still feels like a hack though, as other maven phases might rely on the default order
[11:11:34] <kpiwko> but I'd like to keep the phase test anyway
[11:12:08] <ge0ffrey> kpiwko: let's first investigate how the maven integrestion-test phase works
[11:12:38] <ge0ffrey> it's run after package, before verify: http://maven.apache.org/guides/introduction/introduction-to-the-lifecycle.html
[11:12:45] <kpiwko> ge0ffrey: it's easy...it's actually the same as test, however there's a failsafe plugin
[11:12:59] <kpiwko> so there's a extra cleanup phase available after tests
[11:13:12] <ge0ffrey> kpiwko: where does the sources of those tests go?
[11:13:44] <Rostifant> Maybe you can help me. I am trying to use arquillian on a jboss7 evironment. I know that there are some problems with the manifest.mf, where the jboss sets its dependencies. I removed all external dependencies to my deployment and now I am trying to run a test but i run into several NCF Exceptions.
[11:14:00] <kpiwko> ge0ffrey: by default src/test/java
[11:14:01] <Rostifant> In fact Jboss is missing some corba classes?!?
[11:14:13] <kpiwko> ge0ffrey: so you have to skip surefire execution
[11:14:24] <Rostifant> Caused by: java.lang.ClassNotFoundException: org.omg.CORBA.TypeCode from [Module "deployment.test.war:main" from Service Module Loader]
[11:14:25] <kpiwko> or create an extra directory
[11:15:17] <aslak> Rostifant, what do you mean removed all external deps ?
[11:15:18] <kpiwko> Rostifant: so you removed all dep jars from deployent expecting AS7 will provide them?
[11:16:29] <kpiwko> Rostifant: here's the list of libraries which AS7 provides on classpath including triggering conditions https://docs.jboss.org/author/display/AS7/Implicit+module+dependencies+for+deployments
[11:16:30] <Rostifant> No, sorry for the misunderstanding. My app dependend on an external module (I used slf4j logging). So I do not include any external modules from the jboss.
[11:16:37] <ge0ffrey> kpiwko: another theory in my head: put the integration tests in a seperate module and depend on the war?
[11:16:51] <ge0ffrey> we have normal unit tests too
[11:17:25] <kpiwko> ge0ffrey: right, this will work but is sooo inconvenient
[11:17:53] <Rostifant> Because external modules have to be included via manifest.mf which doesn't work with arq right now. (arq defines it's own manifest.mf)
[11:18:14] <ge0ffrey> kpiwko: yea
[11:18:16] <aslak> Rostifant, which AS v?
[11:18:24] <Rostifant> 7.01
[11:18:31] <aslak> Rostifant, which NCE are you getting ?
[11:18:35] <ge0ffrey> kpiwko: with includes and excludes on the surfire and failsafe plugin issuppose
[11:19:02] <ge0ffrey> surefire includes everything not ending in IT, failsafe includes everythig ending in IT
[11:19:10] <Rostifant> Caused by: java.lang.ClassNotFoundException: org.omg.CORBA.TypeCode from [Module "deployment.test.war:main" from Service Module Loader]
[11:20:42] <aslak> corba?
[11:20:48] <Rostifant> This occurs at a cdi Injection point, cause the stacktrace tells me that weld is trying to deploy a bean
[11:20:52] <aslak> Rostifant, what are you deploying ?
[11:21:35] <Rostifant> A domain layer with several ejbs, some cdi Injects and a persistence context.
[11:21:36] <kpiwko> ge0ffrey: we have basically two possibilities here...either simulate the plugin or provide user a template how to properly configure integration-test with arquillian
[11:21:36] <aslak> or let me rephrase.. do you depend on org.om.CORBA any where ?
[11:21:41] <Rostifant> It's just a domain layer
[11:21:42] <Rostifant> no
[11:22:12] <Rostifant> if i would use corba the I would know that it's my dependency
[11:22:19] <ge0ffrey> "simulate the plugin"? you mean the war plugin?
[11:22:48] <kpiwko> ge0ffrey: war & ear
[11:22:48] <ge0ffrey> problem is the war plugin could do stuff like generate the web.xml etc (not doing that myself)
[11:22:56] <kpiwko> ge0ffrey: right
[11:23:03] <aslak> Rostifant, at which injection point is weld complaining ?
[11:23:06] <ge0ffrey> and that's probably just the start of the rabbit hole
[11:23:17] <kpiwko> so we definitely have to execute in integration-test
[11:23:37] <ge0ffrey> kpiwko: afraid so, that means good documentation in the manual on how to set that up properly :(
[11:23:53] <aslak> Rostifant, you running with the Serlvet protocol now btw?
[11:23:54] <kpiwko> ge0ffrey: or we should get Arquillian-Maven users more educated in Maven :)
[11:24:04] <ge0ffrey> and try to run some guidelines down their throth, like integration tests should end with IT as the maven guys propose :)
[11:24:34] <Rostifant> @aslak yes servlet 3.0
[11:24:34] <ge0ffrey> kpiwko: yes, same thing: good documentation on how to set it up
[11:24:49] <Rostifant> and I don't know which injection point is the problem
[11:24:59] <aslak> Rostifant, can you pastebin.com the full stacktrace your getting
[11:25:02] <ge0ffrey> kpiwko: or, there's a third option: maven-arquillian-plugin
[11:25:06] <Rostifant> of course
[11:25:17] <kpiwko> ge0ffrey: it is already there
[11:25:31] <ge0ffrey> It is? What does it do?
[11:25:50] <aslak> ge0ffrey, not what you want it to.. ;)
[11:25:57] <kpiwko> ge0ffrey: ask Aslak for details, but basically deploy/undeploy, start/stop container
[11:26:17] <aslak> https://github.com/arquillian/arquillian-maven
[11:26:37] <aslak> ge0ffrey, can deploy/undeploy your modules generated war, but it does not run any tests
[11:26:51] <aslak> ge0ffrey, think maven-jboss plugin / jetty-run plugin
[11:27:29] <ge0ffrey> kpiwko: let's call our proposal (nr 3), maven-arquillian-test-plugin (wheter we go for it or not (probably not))
[11:27:47] <ge0ffrey> or maven-arquillian-it-plugin, idunno
[11:28:02] <ge0ffrey> aslak: tnx, makes sense, kinda like cargo?
[11:28:21] <aslak> ge0ffrey, yea, basically
[11:28:43] <aslak> ge0ffrey, except when running Managed etc we don't keep it alive between runs etc
[11:28:59] <aslak> so deploy to a managed makes little sense atm
[11:29:13] <aslak> it will start, deploy, and shut it down
[11:29:42] <ge0ffrey> kpiwko: I think 2) "maven configuration documentation" should be the first way to go about it and once that works, 3) a "specific maven plugin" that just works and does that for you, should be investigated?
[11:29:51] <Rostifant> @aslak one moment. I have to cut the stacktrace it's too long for pastebin
[11:31:06] <Rostifant> http://pastebin.com/1LTRp5B7
[11:31:32] <Rostifant> here is the start of my deployment. and the first ncf exceptions
[11:32:11] <Rostifant> as you can  see. My Beans where deployed as I intended it but then some strange things happen
[11:34:08] <Rostifant> Maybe it has to do with the persistence context
[11:34:25] <aslak> HibernateAnnotationScanner seems to be the one that triggers it
[11:34:34] <kpiwko> ge0ffrey: another gotcha...where should test-classes go in war archive?
[11:35:04] <ge0ffrey> kpiwko: not in a maven build, but for the stuff that's deployed to the arquillian container, yes
[11:36:02] <ge0ffrey> kpiwko: that's another point: currently maven writes all my stuff to the exploded directory, then arquillian grabs that stuff, creates a war file (= zip) gives that to tomcat, tomcat unzips that and deploys
[11:36:03] <kpiwko> ge0ffrey: right, but where
[11:36:10] <Rostifant> ok, but how to resolve this issue, I mean I cannot deploy without annotations
[11:36:15] <kpiwko> ge0ffrey: int WEB-INF/classes?
[11:36:17] <kpiwko> *into
[11:36:25] <aslak> Rostifant, hehe i agree
[11:36:26] <ge0ffrey> kpiwko: yep, I think so. with a fail-fast for now if it's already there
[11:36:58] <ge0ffrey> (as it's easier to later change the fail-fast behaviour into overwrite behaviour then vica versa)
[11:37:29] <aslak> Rostifant, Failed to link org/jboss/com/sun/corba/se/impl/encoding/WrapperInputStream CNF org.omg.CORBA_2_3.portable.InputStream
[11:39:21] <Rostifant> ??
[11:39:45] <Rostifant> maybe I should update my hibernate
[11:40:26] <aslak> Rostifant, which data base do you use ?
[11:40:56] <Rostifant> postgres 9
[11:42:30] <maschmid> aslak: for the arquillian-jacoco, should I wait for arquillian-bom release with SW 1.0.0-beta-6, or is it safe to override the version?
[11:43:09] <Rostifant> I will update to Hibernate 4 rc3. It seems that they are patching it to newer jboss libraries
[11:43:20] <aslak> Rostifant, you 100% sure you haven't by accident imported org.omg.CORBA.portable.InputStream instead of java.io.InputStream somewhere ? org.omg is part of the jdk as well, so you don't need any extra deps to import it
[11:44:10] <Rostifant> I am 100% sure. I have searched for *corba* in my codebase.
[11:44:26] <aslak> you search for *CORBA* as well ? :)
[11:44:40] <aslak> searched
[11:44:52] <Rostifant> hehe
[11:44:54] <Rostifant> yes
[11:44:59] <Rostifant> case insensitive
[11:45:01] <Rostifant> ;)
[11:46:22] <aslak> Rostifant, based on the stack trace it seems like hibernate is reading the annotation index that is provided for the deployment trying to load the classes that is 'known' to create a unique package name list, but fails to load one due to it being compiled against the CORBA lib
[11:49:23] <aslak> it's suppose to handle CFNE, but in this case it gets a LinkageError
[11:56:27] <Rostifant> I updated my hibernate, maybe it will help
[11:58:35] *** jhuska has joined #jbosstesting
[12:07:10] <aslak> maschmid, you can update to 1.0.0.Final-SNAPSHOT v. for now
[12:14:39] <Rostifant> ok nothing helps
[12:15:37] <Rostifant> @aslak one remarkable information I have forgotten to provide. If I deploy my app as a War file manually or via eclipse, everything runs perfectly
[12:18:38] <aslak> Rostifant, seeing if Hardy is alive, maybe he has some insight
[12:19:46] <aslak> Rostifant, can you pastebin pom / arquillian.xml and @Deployment method ?
[12:29:35] <Rostifant> sorry, smoker's break.
[12:30:24] *** ldimaggi has joined #jbosstesting
[12:31:51] <Rostifant> @aslak http://pastebin.com/7d2aUaAL
[12:32:42] <aslak> lol
[12:32:57] <Rostifant> ??
[12:33:01] <aslak> your are literally deploying half the world there
[12:33:08] <aslak> .addPackages(true, "org")
[12:33:27] <Rostifant> I thought this will deploy just my app
[12:33:32] <aslak> that means add org.* recursivly from what ever is on your classpath
[12:33:38] <Rostifant> aaaaaaaaaaaaarggggg
[12:33:52] <Rostifant> i thought it's recursly in my app
[12:34:13] <aslak> no, Class / Package adding depends on ClassPath
[12:34:29] <aslak> ShrinkWrap has really no understand of 'your project'
[12:34:40] <Rostifant> ok, dummy fault
[12:34:49] <Rostifant> fuck
[12:35:10] <aslak> :)
[12:35:27] <Rostifant> sorry sorry sorry
[12:35:33] <Rostifant> it seems to work
[12:35:40] <aslak> excellent!
[12:35:59] <Rostifant> So, now I will kick myself a few times
[12:39:21] <aslak> you could do something like .addAsResources(new File("target/classes/org")), that would limit it to your module
[12:40:33] *** maeste has quit IRC
[12:42:38] <Rostifant> i just expanded it to org.atmosfair.
[12:42:39] <Rostifant> ^^
[12:42:49] <aslak> that should work as well
[12:46:01] <Rostifant> thx 4 all
[13:07:45] *** maeste has joined #jbosstesting
[13:15:30] *** timte has quit IRC
[13:18:30] <kpiwko> ge0ffrey: hi
[13:18:50] <kpiwko> ge0ffrey: once again I need to consult something :)
[13:20:01] *** kevinpollet_ has joined #jbosstesting
[13:21:53] <ge0ffrey> kpiwko: hi :) ask away
[13:21:57] <ge0ffrey> 2 heads are better than 1
[13:22:16] <kpiwko> ge0ffrey: so now I'm wondering how do you want to include testClasses in EARs
[13:22:32] <kpiwko> because you can package classes in EAR
[13:22:44] <kpiwko> *you can't
[13:22:51] <kpiwko> unless packaged in a jar file
[13:23:24] <ge0ffrey> kpiwko: I 'd presume to add an extra jar file to package them, yes
[13:23:48] <ge0ffrey> I don't use ears myself in droolsjbpm
[13:27:27] <kpiwko> ge0ffrey: I think I'll disable this in EARs
[13:28:01] <kpiwko> ge0ffrey: behavior is not consistent between JARs and WARs anyway
[13:28:04] <ge0ffrey> kpiwko: sounds good for the first version. If people want it, they can always open an issue
[13:31:37] *** sofasofa has joined #jbosstesting
[13:40:52] <sofasofa> can anybody help me? im getting a NoClassDefFoundError for org.apache.xerces.xni.parsers.XMLDocumentFilter. i'm using arq-1.00-CR1 jsfunit-2.0-beta2 and glassfish-remote-1.0.0-CR1
[13:52:18] *** jamezp_afk has quit IRC
[13:52:46] *** timte has joined #jbosstesting
[13:54:34] <sofasofa> the package org.apache.xerces.xni.parser gets correctly added to the deployment by the jsfunitArchiveAppender. maybe the glassfish is the hound
[13:54:43] *** jeand_ has quit IRC
[13:58:57] *** mbg has quit IRC
[14:05:48] <sofasofa> aslak: have you an idea, where i can look at?
[14:08:54] *** jamezp_afk has joined #jbosstesting
[14:14:03] <Rostifant> maybe different xerces versions? Can you explore your Xerces Impl and Api jar for the containment of xmldocumentfilter
[14:14:28] <Rostifant> and make sure that api and implementation of xerces were added
[14:17:34] <ge0ffrey> Who's working on jboss-7-embedded?
[14:18:29] <ge0ffrey> I am using tomcat-managed, but it's unreliable (If kill my test during debug, the background process stays alive and the port is still bound)
[14:21:20] <aslak> sofasofa, not sure.. jose_freitas ?
[14:21:31] <aslak> ge0ffrey, ALR
[14:23:07] <ge0ffrey> aslak: tnx
[14:23:16] <ge0ffrey> If he needs guinea pigs, count me in :)
[14:24:04] <aslak> ge0ffrey, a bit odd that tomcat hangs tho, the container is registering a shutdown hook that kills the process it's started in
[14:25:05] <ge0ffrey> aslak: I am running the test in intellij
[14:25:15] <ge0ffrey> so I set a break point in an RunASClient test
[14:26:00] <ge0ffrey> apprently there's some timeout or something happening, because resume doesn't work and it alwasy leaves me in state where the test process is killed but the tomcat managed process is not
[14:27:15] <ge0ffrey> btween that (think hudson, think other team members) and not being able to debug out-of-the-box (think other team members again), I don't see our future with tomcat-managed for day-to-day testing
[14:41:33] *** jose_freitas has quit IRC
[14:44:15] *** Jaikiran has joined #jbosstesting
[14:44:31] *** kevinpollet_ has quit IRC
[14:47:07] <timte> I can't find any indications on the blog or anywhere else that Arquillian has any later release than alpha 5. But in the maven repository there is betas and event a CR5
[14:54:39] <aslak> timte, correct..  a tad bit behind on release notes.. give it a week and we'll give you a big splash..
[14:54:52] <timte> :)
[14:57:53] <timte> ok, cool
[14:58:48] <aslak> timte, you don't follow jira ? ;)
[14:59:09] <timte> sorry but no
[14:59:58] <timte> I subscribed to the blog rss feed hoping to get updated  :)
[15:04:28] <aslak> :)
[15:04:55] <sofasofa> Rostifant: thx, but doesn't help so far. it seems that glassfish strays in his classloader delegation, but i don't know why. i have now found a workaround with <class-loader delegate="false"/> in the glassfish-web.xml.
[15:10:47] *** pilhuhn has joined #jbosstesting
[15:10:47] *** pilhuhn has joined #jbosstesting
[15:39:29] <aslak> Jaikiran, any input here? http://community.jboss.org/message/628015#628015
[15:46:34] <Jaikiran> aslak: replied
[15:46:44] <Jaikiran> my guess is that the ejb-jar.xml ends up at wrong location
[15:50:53] <kpiwko> aslak: ge0ffrey: maschmid: jharting: oskutka: jhuska: if you guys have time, I'd love to hear some feedback on https://github.com/kpiwko/resolver/commit/20957339e2976047bc4d4281dae21ce84996cc17
[15:50:53] <jbossbot> git [resolver] 2095733.. Karel Piwko SHRINKWRAP-325 Implementation of MavenImporter....
[15:50:54] <jbossbot> jira [SHRINKWRAP-325] ShrinkWrap.create(MavenImporter.class) that creates a WebArchive that adds all WEB-INF descriptors, classes, test classes and webapp resources too (not just dependencies and test dependencies) [Open (Unresolved) Feature Request, Critical, Karel Piwko] https://issues.jboss.org/browse/SHRINKWRAP-325
[15:51:31] <aslak> Jaikiran, unless we're seeing a bug in ShrinkWrap, addAsResource("META-INF/ejb-jar.xml") should end in the same location as addAsManifestResource("META-INF/ejb-jar.xml", "ejb-jar.xml")
[15:52:09] <aslak> in a JavaArchive
[15:52:45] <ge0ffrey> kpiwko: is it enough to clone just your fork of resolver, build that and then switch what dependencies to what version?
[15:54:19] <kpiwko> ge0ffrey: it should be enough to get SHRINKWRAP-325 branch and the use 1.1.0-alpha-1-SNAPSHOT version
[15:54:20] <jbossbot> jira [SHRINKWRAP-325] ShrinkWrap.create(MavenImporter.class) that creates a WebArchive that adds all WEB-INF descriptors, classes, test classes and webapp resources too (not just dependencies and test dependencies) [Open (Unresolved) Feature Request, Critical, Karel Piwko] https://issues.jboss.org/browse/SHRINKWRAP-325
[15:54:48] <ge0ffrey> kpiwko: tnx, I 'll try it out today or monday
[15:56:06] <kpiwko> ge0ffrey: version should be switched for shrinkwrap-resolver-* and probably shrinkwrap deps might be updated to 1.0.0-beta-6 which was released today and contains some required fixes
[15:56:25] <kpiwko> well, not really required, required for test in order to pass :)
[15:57:31] <kpiwko> ge0ffrey: thx
[15:57:46] <aslak> kpiwko, not required to run, but to work.. ;)
[15:58:19] <kpiwko> aslak: :)
[15:58:21] <ge0ffrey> k, I 'll see what I can make of that :)
[15:58:53] <Jaikiran> aslak: hmm yeah, just tested. it should be the same
[16:01:36] <aslak> Jaikiran, aa..
[16:01:43] <aslak> Jaikiran, i know what it is..
[16:01:50] <Jaikiran> what's that? :)
[16:02:20] <aslak> Jaikiran, ejb library jars are not allowed to have ejb-jar.xml files when packaged as a library in a web archive right ?
[16:02:29] <aslak> ejb-jar has to be placed in WEB-INF ?
[16:02:44] <dblevins> aslak: right
[16:02:52] <dblevins> (unfortunately)
[16:02:54] <Jaikiran> aslak: if that jar ends up in .war/lib/ejb.jar then it's valid
[16:03:01] <Jaikiran> i mean they will be picked up
[16:03:16] <Jaikiran> dblevins: EJB3.1 allows that
[16:03:55] <dblevins> the ejb-jar.xml is only honored in WEB-INF/ejb-jar.xml
[16:04:08] <aslak> hmm..  i thought war/lib/ejb.jar/META-INF/ejb-jar.xml would be ignored. and should be placed in war/WEB-INF/ejb-jar.xml
[16:04:11] <dblevins> i argued heavily for allowing to be in the jars themselves and lost
[16:04:19] <Jaikiran> A .war contains at most one ejb-jar.xml. The ejb-jar.xml can be packaged either as
[16:04:19] <Jaikiran> WEB-INF/ejb-jar.xml or as META-INF/ejb-jar.xml within one WEB-INF/lib .jar file,
[16:04:19] <Jaikiran> but not both.
[16:04:36] <Jaikiran> that's from 20.4 section of EJB3.1 spec
[16:04:52] <dblevins> oh yea, .. the "one jar" rule
[16:05:03] <aslak> right, we don't have both.. so lib/ejb.jar/META-INF/ejb-jar.xml should be valid then
[16:05:39] * dblevins grumbles about the complexity of not just allowing it everywhere like CDI does... and JPA does.. and Servlet fragments do
[16:06:23] <dblevins> Jaikiran: thanks for the reminder... I need to do some fixing now :)
[16:06:46] <Jaikiran> dblevins: actually wait :) i think that got "revised" in the final version
[16:06:48] <aslak> no ejb tck test on this ? :)
[16:07:07] <dblevins> definitely not covered in the TCK
[16:07:26] <dblevins> we have just a handful of failures left and this is definitely not one of them
[16:07:37] <aslak> yea, why test concrete rules like that in a test suite.. makes sense
[16:07:51] <dblevins> indeed
[16:08:08] <Jaikiran> dblevins: turns out it got removed in the final version of spec
[16:08:09] <dblevins> better to test things that aren't in the spec and just the RI
[16:08:14] <Jaikiran> it now is "A .war may contain an ejb-jar.xml file. If present, the ejb-jar.xml is packaged as
[16:08:14] <Jaikiran> WEB-INF/ejb-jar.xml. "
[16:08:44] <dblevins> ok, that's where I thought it ended up... but I've been wrong before
[16:09:00] <aslak> well, that's the problem then.. :)
[16:09:27] <Jaikiran> so the thing is - you can have annotation based EJBs in .war/WEB-INF/lib/.jar files but if a ejb-jar.xml is to be processed, it has to be in .war/WEB-INF/
[16:09:45] <Jaikiran> aslak: yeah, that's the side effect of the arquillian war packaging then
[16:09:51] <aslak> Jaikiran, yea
[16:10:11] <aslak> i blame the spec.. :)
[16:10:22] <aslak> i'll reply
[16:10:37] <Jaikiran> :)
[16:11:40] <timte> arquillian-jbossas-remote-6 does not exist for CR5, only CR2
[16:12:03] <timte> what's the latest official version?
[16:12:10] <Rostifant> @aslak should the injection of a stateless ejb via cdi inject work on as7? Because I am getting a NPE
[16:21:18] <aslak> timte, Arq Core and Containers X are on different release cycles..
[16:21:56] <aslak> timte, latest Arq Core is CR5, latest JBoss AS Containers is CR2
[16:22:34] <timte> aslak: ok
[16:22:41] <aslak> timte, use Arquillian Core BOM and dependencyManagement to upgrade the Containers Arq Core deps to the Arq Core v. of your liking: https://gist.github.com/1154075
[16:22:52] <aslak> Rostifant, it should..
[16:23:00] <kpiwko> lfryc: na tebe jsem zapomnel s https://github.com/shrinkwrap/resolver/pull/7 :)
[16:23:32] <timte> aslak: thanks
[16:24:02] <aslak> kpiwko, in SHRINKWRAP-325, you assume Maven has been ran previously and created the output file right ?
[16:24:03] <jbossbot> jira [SHRINKWRAP-325] ShrinkWrap.create(MavenImporter.class) that creates a WebArchive that adds all WEB-INF descriptors, classes, test classes and webapp resources too (not just dependencies and test dependencies) [Pull Request Sent (Unresolved) Feature Request, Critical, Karel Piwko] https://issues.jboss.org/browse/SHRINKWRAP-325
[16:24:11] <aslak> kpiwko, so no direct support from IDE
[16:24:34] <kpiwko> aslak: right, this is maven importer, so it expect output is there
[16:25:05] <kpiwko> aslak: actually with IDE support, in idea it would work, as Geoffrey suggested
[16:25:23] <kpiwko> for eclipse it might a a matter of eclipse plugin configuration
[16:25:29] <timte> aslak: also, are a bunch of classes moved? for instance  org.jboss.arquillian.api.Deployment ==> org.jboss.arquillian.container.test.api.Deployment
[16:25:51] <kpiwko> not sure about that though
[16:25:56] <ge0ffrey> aslak: this works in intellij unless you do really black magic in the maven pom's
[16:26:24] <aslak> to my knowledge, eclipse does not build the artifact it self
[16:26:24] <ge0ffrey> even filtered resources work, as intellij runs "process resources" phase during maven synchronization
[16:26:41] <aslak> timte, http://community.jboss.org/wiki/MigrationToArquillian100CR1
[16:26:53] <ge0ffrey> ah yea, it configures the configuration automatically, but you need to hook itup
[16:27:39] <timte> aslak: awsome
[16:27:41] <ge0ffrey> in "run config", enabled "build guvnor-webapp:war exploded" artifact
[16:31:16] <aslak> kpiwko, what does ShrinkWrap.create(MavenImporter.class).configureFrom("pom.xml").as(WebArchive.class) do ?
[16:32:41] <kpiwko> aslak: this will discard parsed pom as the pom configuration is stored in EffectivePomMavenImporter and user has to parse it again
[16:33:23] <kpiwko> aslak: the idea is you parse, you modify, you change to a different archive and do the modifications by hand if required
[16:33:47] <kpiwko> so your call will consume some time and creates an empty war archive
[16:33:59] <aslak> right, that was my assumption
[16:35:06] <kpiwko> more important is that it works other way around, that is ShrinkWrap.create(WebArchive.class)...as(MavenImporter.class)...
[16:35:07] <aslak> aa yea, configureFrom is settings.xml, not a pom
[16:35:57] <kpiwko> aslak: if the name is not clear, it should be renamed during api review...configureFrom is to align with MavenDepedencyResolver which has the method with same name
[16:36:11] <kpiwko> which actually does the same thing
[16:36:49] <kpiwko> withMavenSettings() might be a candidate name :)
[16:39:48] <aslak> hmm yea
[16:40:43] <timte> DependencyResolvers and MavenDependencyResolver are no more?
[16:42:14] <Rostifant> Ok, I can inject Beans into my Testcases but if I inject a bean into another bean then this injection will result into a NPE
[16:42:37] <Rostifant> so @Inject only works in the @Test method
[16:42:51] <kpiwko> timte: they are still there, but MavenImporter api might supersede the usage in future ;)
[16:42:52] <aslak> timte, they are, just not pulled in transitivly from arq, but they are defined in the bom, so just add a dep to org.jboss.shrinkwrap.resolvers:shrinkwrap-maven-resolver-impl or what it was called
[16:44:03] <aslak> Rostifant, hmm..  that doesn't sound right..
[16:44:03] <kpiwko> timte: ahh, different discussion
[16:46:07] <aslak> :)
[16:46:10] *** mbg has joined #jbosstesting
[16:48:48] <Rostifant> hmpf, why am I getting into trouble every fucking time.
[16:54:17] *** timte has quit IRC
[16:56:46] *** ALR has joined #jbosstesting
[16:57:38] <ALR> aslak: Begging forgiveness :)
[16:58:20] <aslak> ALR, hehe
[16:59:35] *** jhuska has quit IRC
[16:59:36] <aslak> let me grab a coffee and fun can begin
[17:00:04] <aslak> http://www.oracle.com/technetwork/articles/java/integrationtesting-487452.html
[17:00:05] <aslak> :)
[17:00:27] *** kpiwko has quit IRC
[17:01:13] <ALR> aslak: And let me figure out WTF Google Hangout is
[17:06:34] *** jhuska has joined #jbosstesting
[17:06:38] *** jhuska has quit IRC
[17:10:16] <Rostifant>  @aslak there is a failure in the oracle article
[17:10:26] <Rostifant> addAsManifestResource(new ByteArrayAsset("<beans/>".getBytes()),ArchivePaths.create("beans.xml"));
[17:11:02] <Rostifant> has to be changed to addAsWebInfResource
[17:11:23] <Rostifant> if it will be no jar
[17:11:59] <Rostifant> Do you have any idea why injection only works particular for me?
[17:12:22] <aslak> Rostifant, the article is written for Alpha5 so..
[17:13:34] *** jeand has joined #jbosstesting
[17:18:22] *** jamezp_afk is now known as jamezp
[17:22:38] *** jamezp has quit IRC
[17:24:08] *** jharting has quit IRC
[17:24:42] *** rruss has joined #jbosstesting
[17:31:40] *** jamezp has joined #jbosstesting
[17:35:54] *** sofasofa has quit IRC
[18:02:34] *** bleathem has quit IRC
[18:03:52] *** alesj has quit IRC
[18:05:59] *** rachmatowicz has quit IRC
[18:06:17] *** rachmatowicz has joined #jbosstesting
[18:14:11] *** bleathem has joined #jbosstesting
[18:17:39] *** oskutka has quit IRC
[18:23:12] *** pilhuhn has quit IRC
[18:24:25] *** kevinpollet has joined #jbosstesting
[18:31:17] *** pilhuhn has joined #jbosstesting
[18:38:52] *** lfryc has quit IRC
[18:40:08] *** maschmid has quit IRC
[18:45:58] *** Jaikiran is now known as Jaikiran|Dinner
[18:48:22] *** ge0ffrey has quit IRC
[18:49:31] *** ianbrandt has joined #jbosstesting
[19:03:12] *** ge0ffrey has joined #jbosstesting
[19:55:12] *** Jaikiran|Dinner has quit IRC
[20:05:09] *** jose_freitas has joined #jbosstesting
[20:13:50] *** vnvarsete has quit IRC
[20:29:48] *** oskutka has joined #jbosstesting
[20:29:53] *** oskutka has quit IRC
[20:31:36] *** rruss has quit IRC
[20:33:03] *** lightguard_jp has joined #jbosstesting
[20:40:45] *** rruss has joined #jbosstesting
[20:41:10] *** rruss has quit IRC
[20:50:35] *** kevinpollet has quit IRC
[21:05:55] *** Rostifant has quit IRC
[21:13:01] *** mgoldmann_ has joined #jbosstesting
[21:13:13] *** mgoldmann has quit IRC
[21:14:21] *** ge0ffrey has quit IRC
[21:24:37] *** jamezp is now known as jamezp_afk
[21:25:46] *** kevinpollet has joined #jbosstesting
[21:31:56] <tombee> damn jboss as 7 and arquillian are sweet!
[21:32:04] <tombee> with JPA!
[21:43:30] *** mhuniewicz has joined #jbosstesting
[21:43:35] <mhuniewicz> aslak, hi
[21:51:09] *** jamezp_afk is now known as jamezp
[21:53:40] *** lfryc has joined #jbosstesting
[21:56:59] *** maeste has quit IRC
[22:07:15] *** rachmatowicz has quit IRC
[22:18:37] *** maeste has joined #jbosstesting
[22:21:18] *** mgoldmann has joined #jbosstesting
[22:24:13] *** mgoldmann_ has quit IRC
[22:24:49] *** maeste has quit IRC
[22:26:41] *** mgoldmann has quit IRC
[22:27:28] *** maeste has joined #jbosstesting
[22:37:39] *** maeste has quit IRC
[22:48:09] *** maeste has joined #jbosstesting
[22:49:27] <mhuniewicz> ALR, hi
[22:53:59] *** maeste has quit IRC
[23:06:38] *** maeste has joined #jbosstesting
[23:17:23] *** maeste has quit IRC
[23:17:40] *** rruss has joined #jbosstesting
[23:37:26] *** jeand has quit IRC
[23:39:15] *** pilhuhn has quit IRC
[23:57:09] *** kevinpollet has quit IRC





top