[00:12:40] *** alesj1 has joined #jbosstesting
[00:13:55] *** alesj has quit IRC
[00:31:55] *** alesj1 has quit IRC
[01:15:55] *** kevinpollet has quit IRC
[01:30:43] *** rruss has quit IRC
[01:47:28] *** ldimaggi has joined #jbosstesting
[01:55:16] *** mbg has quit IRC
[01:57:42] *** aslak has quit IRC
[02:01:45] *** lightguard_jp is now known as lightguard_jp_aw
[02:30:18] *** jamezp is now known as jamezp_afk
[03:11:31] *** mbg has joined #jbosstesting
[04:36:27] *** rruss has joined #jbosstesting
[06:06:57] *** ldimaggi has quit IRC
[06:08:26] *** lightguard_jp_aw is now known as lightguard_jp
[06:10:22] *** maeste has joined #jbosstesting
[06:22:50] *** bleathem has quit IRC
[06:40:15] *** mbg has quit IRC
[06:40:46] *** mbg has joined #jbosstesting
[06:46:33] *** dblevins has quit IRC
[06:49:08] *** dblevins has joined #jbosstesting
[07:07:43] *** oskutka has joined #jbosstesting
[07:12:40] *** oskutka has quit IRC
[07:13:53] *** oskutka has joined #jbosstesting
[07:20:56] *** alesj has joined #jbosstesting
[07:40:42] *** vnvarsete has joined #jbosstesting
[07:53:06] *** Jaikiran has joined #jbosstesting
[07:55:36] *** Jaikiran is now known as Jaikiran|brb
[08:28:48] *** Jaikiran|brb is now known as Jaikiran
[08:31:57] *** tdiesler has joined #jbosstesting
[08:42:35] *** OndraZizka has quit IRC
[08:42:50] *** OndraZizka has joined #jbosstesting
[08:51:07] *** maschmid has joined #jbosstesting
[09:00:14] *** jharting has joined #jbosstesting
[09:01:05] *** lfryc has joined #jbosstesting
[09:04:25] *** ge0ffrey has joined #jbosstesting
[09:28:55] *** jhuska has joined #jbosstesting
[09:35:11] *** kpiwko has joined #jbosstesting
[09:41:51] *** rruss has quit IRC
[09:49:39] *** jeand has joined #jbosstesting
[09:53:56] *** pilhuhn has joined #jbosstesting
[09:53:57] *** pilhuhn has joined #jbosstesting
[09:59:17] *** timte has joined #jbosstesting
[10:01:41] *** aslak has joined #jbosstesting
[10:32:55] *** lightguard_jp has quit IRC
[10:34:26] *** jeand_ has joined #jbosstesting
[10:38:02] *** jeand has quit IRC
[10:53:12] *** pilhuhn is now known as pil-afk
[12:28:53] *** maeste has quit IRC
[13:02:06] *** alesj has quit IRC
[13:14:23] *** alesj has joined #jbosstesting
[13:18:19] *** dblevins has quit IRC
[13:19:13] *** aLBa^ has joined #jbosstesting
[13:20:09] *** maeste has joined #jbosstesting
[13:22:29] *** timte has quit IRC
[13:22:39] *** dblevins has joined #jbosstesting
[13:23:09] *** timte has joined #jbosstesting
[13:28:53] *** vtunka has joined #jbosstesting
[13:39:33] *** jamezp_afk has quit IRC
[13:40:09] *** jamezp_afk has joined #jbosstesting
[14:00:54] *** timte has joined #jbosstesting
[14:15:26] <aLBa^> how does JavaArchive's addAsResource(String resourceName) look for its resource?
[14:15:28] *** dblevins has quit IRC
[14:17:22] <alesj> aLBa^: probably checks its classpath, from root, for the same resourceName
[14:19:39] *** dblevins has joined #jbosstesting
[14:20:11] *** pil-afk is now known as pilhuhn
[14:28:24] *** maschmid has quit IRC
[14:28:40] *** maschmid has joined #jbosstesting
[14:30:13] *** ldimaggi has joined #jbosstesting
[14:41:07] <aLBa^> Ive used a file reference instead and that seems to work
[14:44:39] <aLBa^> but the file in question is my persistence.xml and during deployment it says it cannot find my persistence unit
[14:47:59] <aslak> aLBa^, which type of archive?
[14:48:35] <aLBa^> JavaArchive
[14:48:50] <aslak> aLBa^, you probably want addAsResource(new File("persistence.xml"), "/META-INF/persistence.xml")
[14:49:32] *** rruss has joined #jbosstesting
[14:51:11] <aLBa^> all better
[14:54:50] <vnvarsete> aslak: About persistence.xml. Say I have a couple of categories of tests, some tests where I want to use "plain manual transaction management" where transaction-type=RESOURCE_LOCAL, while in other tests I want the container (ARQUILLIAN) to control the t ransaction management (transaction-type=JTA). If I specify both persistence units in the same persistence.xml I ended up with some failures (don't remember what right now, but can find it tomorro
[14:55:50] <aslak> vnvarsete, ok ?
[14:56:28] *** ldimaggi has quit IRC
[14:57:14] <vnvarsete> aslak: not, ok? think it was something about validation of the persistence unit using RESOURCE_LOCAL when deployed to JBoss by Arquillian
[14:57:42] <vnvarsete> aslak: something about how to define non-jta-data-source
[14:58:08] <aslak> aa yes
[14:58:30] <vnvarsete> aslak: I would really like to maintain only ONE persistence.xml in test
[14:58:35] *** vineetr has joined #jbosstesting
[14:58:51] <vnvarsete> but still have the ability to run tests both using manual and container managed transaction management
[15:02:25] <aslak> vnvarsete, you are then defining javax.persistence.jdbc-driver|url|user|password as properties in the RESOURCE-LOCAL persistence unit?
[15:02:44] <vnvarsete> aslak: Yes, I use hsqldb
[15:04:18] <aslak> defining the javax.persistence properties and using RESOURCE-LOCAL should work in SE, not sure if it's allowed in EE
[15:05:08] *** vnvarsete has quit IRC
[15:05:39] <vineetr> aslak, vnvarsete, RESOURCE_LOCAL transactions for JPA work in EE.
[15:05:43] *** vnvarsete has joined #jbosstesting
[15:05:58] <aLBa^> it is, but he'll need to create 2 deployments each with a different persistence unit
[15:06:03] <vineetr> Yes
[15:06:39] <aslak> vineetr, aLBa^ sure, but you'll have to define a non-jta-datasource jndi lookup right ?
[15:06:50] *** vnvarsete has joined #jbosstesting
[15:07:00] <aslak> vnvarsete, try pastebin.com next time.. ;)
[15:07:04] <vineetr> agree with aLBa^. With a single persistence.xml his tests would have to use UserTransaction, I suppose, instead of CMTs
[15:07:13] <vnvarsete> aslak: Ah..yes, I was kicked out:) sorry
[15:08:00] <vineetr> aslak, I think a non-jta-datasource is not required if the JPA provider manages the conn pool.
[15:09:23] <aslak> vnvarsete, didn't catch any of the content of the files..
[15:10:01] <vnvarsete> aslak: persistence-arquililan.xml: http://pastebin.com/FDDRThmH (using JTA), persistence.xml: http://pastebin.com/ixKt9kGM (using RESOURCE_LOCAL)
[15:10:23] <vnvarsete> vnavarsete: will have to leave now, but will be back tomorrow, I will read all comments then
[15:11:59] *** ldimaggi has joined #jbosstesting
[15:13:21] <vineetr> vnvarsete, I think aslak is right in this case, since the second file has transaction-type="RESOURCE_LOCAL" and no mention of a non-jta-data-source. Check out section 8.2.1.2 of the the JPA 2.0 spec.
[15:13:22] <aLBa^> vineetr that I am not sure of :) We previously did embedded testing on glassfish and I suppose this way you can easily switch between DBs
[15:13:58] <vineetr> Quoting from the spec: "In general, in Java EE environments, a transaction-type of RESOURCE_LOCAL assumes that a non-JTA datasource will be provided."
[15:14:05] <vnvarsete> vineetr: thanks, will do. I think I stopped on this last time since I didn't figure out how to define the non-jta-data-source:(
[15:14:41] <vnvarsete> vineetr/aslak: What should be the value of the non-jta-data-source?
[15:14:53] <aLBa^> jndi location
[15:16:50] <vnvarsete> aLBa^: examples are good;)
[15:19:39] <vnvarsete> vineetr: examples are good;)
[15:21:10] <vineetr> Just create a JNDI datasource. Say jdbc/__default for instance, that connects to your test database. It is like a jta-data-source entry, except that this datasource would/should not participate in JTA transactions.
[15:21:43] *** alesj has quit IRC
[15:22:34] <vineetr> aslak, I've a query on setting up the Eclipse project for the Arquillian-Glassfish repo (first time on git here). I've forked the arquillian-container-glassfish project, and then imported the forked repo into Eclipse.
[15:23:45] <vineetr> I get a few compiler errors like "The method register(LoadableExtension.ExtensionBuilder) of type GlassFishExtension must override a superclass method", because Eclipse sets the project's compiler compliance level to 1.5 and not 1.6
[15:26:06] <aslak> vineetr, yea.. we use @Override on interfaces which is a 1.6 options, but compile for 1.5
[15:26:24] <aslak> vineetr, eclipse doesn't support that setup, source 1 v compile with target for another
[15:26:46] <aslak> vineetr, i normally just update to 1.6 in the pom.xml, and update the import
[15:26:53] <vineetr> ok, sure thanks.
[15:27:09] <vineetr> Got to go now. Running low on battery power.
[15:27:22] *** vineetr has quit IRC
[15:36:38] *** rachmatowicz has joined #jbosstesting
[15:44:59] *** vineetr has joined #jbosstesting
[15:47:12] *** mbg has quit IRC
[15:56:43] *** bleathem has joined #jbosstesting
[15:59:14] *** pilhuhn is now known as pil-afk-biab
[16:04:13] *** aLBa^ has quit IRC
[16:12:56] *** ldimaggi has quit IRC
[16:21:42] *** jharting has quit IRC
[16:25:09] *** pil-afk-biab is now known as pilhuhn
[16:26:00] *** ldimaggi has joined #jbosstesting
[16:31:23] *** kpiwko has quit IRC
[16:42:47] *** oskutka has quit IRC
[16:53:23] *** timte has quit IRC
[16:58:34] *** tdiesler has quit IRC
[17:10:32] *** vineetr has quit IRC
[17:11:05] *** mbg has joined #jbosstesting
[17:11:13] *** mbg has joined #jbosstesting
[17:21:30] *** maeste has quit IRC
[17:27:34] *** jamezp_afk has quit IRC
[17:45:08] *** lfryc has quit IRC
[17:49:41] *** jamezp has joined #jbosstesting
[18:03:01] *** vtunka has quit IRC
[18:05:40] *** maschmid has quit IRC
[18:10:34] *** pilhuhn is now known as pil-dinner
[18:17:06] *** bobmcw has quit IRC
[18:20:12] *** bobmcw has joined #jbosstesting
[18:29:06] *** mbg has quit IRC
[18:32:24] *** ge0ffrey has quit IRC
[18:33:19] *** Jaikiran has quit IRC
[18:42:18] *** oskutka1 has joined #jbosstesting
[18:50:33] *** rruss has quit IRC
[18:54:29] *** jhuska has quit IRC
[19:03:25] *** durandal_ has joined #jbosstesting
[19:12:08] *** jose_freitas_ has joined #jbosstesting
[19:12:49] *** lightguard_jp has joined #jbosstesting
[19:14:13] *** jose_freitas has quit IRC
[19:17:12] *** kevinpollet has joined #jbosstesting
[19:17:58] *** Jaikiran has joined #jbosstesting
[19:21:20] *** Jaikiran has quit IRC
[19:22:00] *** aslak has quit IRC
[19:22:37] *** aslak has joined #jbosstesting
[19:23:00] *** rruss has joined #jbosstesting
[19:24:02] *** jose_freitas_ has quit IRC
[19:26:45] *** jose_freitas has joined #jbosstesting
[19:51:31] *** jose_freitas has quit IRC
[19:51:55] *** jose_freitas has joined #jbosstesting
[20:08:01] *** pil-dinner is now known as pilhuhn
[20:16:12] *** kevinpollet has quit IRC
[20:20:49] *** dblevins has quit IRC
[20:22:22] *** dblevins has joined #jbosstesting
[20:33:09] <ALR> aslak: Gotta sec?
[20:36:18] <aslak> ALR, sure
[20:36:35] <ALR> aslak: I'm just having some issues hooking in from container to test enricher
[20:36:41] <ALR> Container (server-side) has:
[20:36:53] <ALR> @Inject @ContainerScoped InstanceProducer<T>
[20:36:59] <ALR> And I set that in the setup method.
[20:37:10] <aslak> server-side?
[20:37:36] <ALR> No?  That won't get set on the server side?  What's the @ContainerScoped for?
[20:37:54] <ALR> Basically followed https://github.com/arquillian/arquillian-container-openshift/blob/master/openshift-express/src/main/java/org/jboss/arquillian/container/openshift/express/OpenShiftExpressContainer.java#L62
[20:38:15] *** bmajsak has joined #jbosstesting
[20:38:18] <aslak> ALR, @ContainerScoped is in the scope of this Container instance
[20:38:31] <ALR> So only on the client VM then?
[20:38:41] <aslak> Suite|Class|Test|Container|Deployment scope
[20:38:42] <aslak> yes
[20:38:46] <ALR> Basically I just wanna get what's from the container to the test enricher
[20:38:57] <ALR> So I can inject it into the server-side test
[20:39:13] <aslak> ALR, you can do what you are doing for Client side enrcihment
[20:39:18] <aslak> you probably want that as wel
[20:39:29] <ALR> Yes.
[20:39:46] <ALR> So what needs to be in the enricher class?  Another @Inject of something?
[20:39:58] <ALR> I couldn't find a similar example of an enricher getting something set from the container
[20:39:58] <aslak> for incontianer, you need a AuxilliaryArchiveAppender that creates a Archive'ed v. of your extension
[20:40:12] <aslak> this adds a RemoteLoadableExtension to the Archive
[20:40:36] <aslak> this adds/this should add
[20:40:56] <aslak> and in that RemoteLoadableExtension impl you need to register a TestEnricher
[20:41:10] <ALR> Example?
[20:41:41] <ALR> And then I can do websitey stuff :)
[20:41:42] <aslak> that TestEnricher probably needs to read a ThreadLocal/Static that is set by the msc service you also need to get hold of the contianers ModelClient
[20:42:21] <aslak> all the enrichers do this.. pick one. ejb / resource
[20:42:25] <aslak> :)
[20:42:47] <aslak> https://github.com/arquillian/arquillian-core/tree/master/testenrichers/ejb/src/main/java/org/jboss/arquillian/testenricher/ejb/container
[20:43:03] <aslak> https://github.com/arquillian/arquillian-core/blob/master/testenrichers/ejb/src/main/java/org/jboss/arquillian/testenricher/ejb/client/EJBEnricherArchiveAppender.java
[20:44:12] *** jamezp has quit IRC
[20:44:35] *** jamezp has joined #jbosstesting
[20:48:58] <ALR> aslak: Thanks. I must just be missing some reg point then
[20:51:31] <tombee> Humm why doesn't my JBoss AS 7 print exceptions to console :(
[20:52:17] <tombee> So difficult to workout why something won't deploy when all it says is it failed to deploy!
[20:52:18] <tombee> :(
[20:56:53] *** bmaxwell has joined #jbosstesting
[20:57:50] *** bmaxwell has left #jbosstesting
[21:03:13] *** jamezp is now known as jamezp_afk
[21:04:29] *** jose_freitas has quit IRC
[21:04:35] <tombee> bleh incorrectly named driver for my datasource, produced no output as to what was going wrong heh
[21:06:21] *** alesj has joined #jbosstesting
[21:08:23] *** jose_freitas has joined #jbosstesting
[21:23:07] *** jeand_ has quit IRC
[21:30:01] *** ALR has quit IRC
[21:30:15] *** ALR has joined #jbosstesting
[21:37:54] *** jeand_ has joined #jbosstesting
[22:01:07] *** alesj has quit IRC
[22:01:53] *** jamezp_afk is now known as jamezp
[22:02:30] *** lightguard_jp is now known as lightguard_jp_aw
[22:16:24] *** nilian has joined #jbosstesting
[22:16:52] *** nilian has quit IRC
[22:18:14] *** pilhuhn has quit IRC
[22:18:29] *** ldimaggi has quit IRC
[22:34:39] *** mhuniewicz has joined #jbosstesting
[22:35:49] <mhuniewicz> aslak, ping
[22:38:24] <aslak> mhuniewicz, heya
[22:38:59] <mhuniewicz> aslak, persistence.xml is, obviously, leaking in somehow.
[22:43:43] <aslak> mhuniewicz, you don't see from where ?
[22:44:10] <mhuniewicz> I know where the file is, I added it myself to test this behaviour. I don't know how to prevent it from happening, and I assume this is what you asked me to do?
[22:44:28] <aslak> mhuniewicz, correct
[22:44:53] <aslak> i meant which classloader it is being loaded from and where / why it goes beyond the sw classloader
[22:45:11] <mhuniewicz> aslak, no idea.
[22:46:15] <aslak> well, that's no good
[22:46:19] <aslak> :)
[22:46:35] <aslak> everyone will have a persistence.xml in META-INF already
[22:47:41] <mhuniewicz> I think the test folder is the problem only.
[22:50:18] <mhuniewicz> Should we activate ALR? :)
[22:50:38] <aslak> mhuniewicz, have you tried creating a new ShrinkWrap classloader with a null parent?
[22:50:39] <ALR> What'd I do?
[22:50:54] <aslak> brb
[22:51:44] <mhuniewicz> ALR, I got a simple class that creates a JavaArchive
[22:51:50] <ALR> mhuniewicz: k
[22:51:56] <mhuniewicz> and it pulls is a file called persistence-text.xml
[22:51:59] <mhuniewicz> and calls it persistence.xml.
[22:52:04] <ALR> k
[22:52:23] <mhuniewicz> It comes from the src/test/resources folder.
[22:52:39] <mhuniewicz> But if that folder already has a persistence.xml file, then both are available when I run this test in Eclipse.
[22:52:57] <ALR> Yeah
[22:53:08] <ALR> Because one is on the SW ClassLoader and the other is in the appCL
[22:53:22] <mhuniewicz> Yes.
[22:53:25] <ALR> Try isolating the SW ClassLoader from the app CL
[22:53:35] <ALR> By explicitly setting a null parent during its creation.
[22:54:34] <mhuniewicz> ALR, how?
[22:55:04] <ALR> http://docs.jboss.org/shrinkwrap/latest/org/jboss/shrinkwrap/api/classloader/ShrinkWrapClassLoader.html#ShrinkWrapClassLoader(java.lang.ClassLoader, org.jboss.shrinkwrap.api.Archive...)
[22:55:10] <ALR> Pass null as 1st argument
[22:56:41] <ALR> mhuniewicz: ^
[22:58:21] <mhuniewicz> ALR, where? http://pastebin.com/R5vFcTvS
[22:58:43] <aslak> mhuniewicz, in the TestContainre
[23:01:38] *** dblevins has quit IRC
[23:01:58] *** dblevins has joined #jbosstesting
[23:02:38] <mhuniewicz> aslak, in the TestContainer?
[23:02:48] <aslak> weld-ee
[23:03:04] <mhuniewicz> We're not even instantiating that there.
[23:03:57] <aslak> https://github.com/arquillian/arquillian-container-weld/blob/master/weld-ee-embedded-1.1/src/main/java/org/jboss/arquillian/container/weld/ee/embedded_1_1/WeldEEMockContainer.java#L89
[23:06:19] <mhuniewicz> aslak, that certainly affected everything else. NoClassDefFoundError.
[23:09:12] <aslak> on what?
[23:09:27] <mhuniewicz> LifecycleMixin.
[23:10:16] <durandal_> should testing jpa with arq1.0.0.Alpha5 with glassfish-emebedded 3.1 be doable?  stuck on getting the embedded container resources setup, I see a resolved issue regarding sunResourceXML in arquillian.xml (https://issues.jboss.org/browse/ARQ-397)
[23:10:16] <aslak> aa yea
[23:10:17] <jbossbot> jira [ARQ-397] Re enable support for sunResourceXML [Closed (Done) Feature Request, Major, Aslak Knutsen] https://issues.jboss.org/browse/ARQ-397
[23:10:44] <aslak> weld will use the classloader where the class was found to load everything.  so yea, that won't work
[23:11:32] <aslak> durandal_, upgrade to Core 1.0.0.CR5 and GlassFish 1.0.0.CR1
[23:11:58] <durandal_> aslak, thanks will do
[23:11:59] <aslak> durandal_, example setup with tomcat, https://gist.github.com/1154075
[23:12:42] <durandal_> great, appreciate it
[23:15:58] <mhuniewicz> aslak, are we doomed?
[23:17:14] <ALR> Doomed.  Interesting.
[23:17:32] <ALR> Seems we should have an option to make SW CL child-first delegation too.
[23:17:35] <ALR> aslak: WDYT?
[23:29:25] <aslak> ALR, child first won't help in this case
[23:30:05] <aslak> ALR, if you don't add a persistence.xml it will still find the one in your appcl
[23:30:26] <ALR> It's more of an aside.
[23:30:33] <ALR> Like, "would that also be useful"
[23:30:34] <ALR> ?
[23:34:52] <aslak> ALR, sure, it might
[23:46:47] <mhuniewicz> ... Well?
[23:48:06] <ALR> The real problem is that you want selectively some stuff on your appCL but not all in test/resources
[23:50:09] <mhuniewicz> ALR, should we abandon the whole idea of excluding then?
[23:50:28] <ALR> mhuniewicz: Excluding what?  Excluding appCL?
[23:50:49] <ALR> Yes, unless you construct some other CL which contains what you need only, and set that as the parent
[23:51:01] <ALR> See, this is the problem with Embedded containers.
[23:51:19] <ALR> The onus is on the *user* to have a correctly-setup environment.
[23:51:22] <ALR> And yours...is not :)
[23:51:42] <ALR> So managed containers provide for better isolation from this kind of thing.
[23:51:50] <mhuniewicz> aslak, what do you think?
[23:52:39] <aslak> i agree.. embedded sucks
[23:52:39] <aslak> :)
[23:53:13] <aslak> mhuniewicz, it's a bit to scary as it is now, finding database configurations that has not been defined in the deployment
[23:53:48] <aslak> have some thoughts about moving to JBoss Modules on the client for better CP isolation, but that's far from done/started
[23:55:38] <mhuniewicz> aslak, so you're not willing to accept it the way it is right now?





top