[01:06:03] *** bleathem has quit IRC
[01:07:26] *** bleathem has joined #jbosstesting
[01:07:37] *** aslak has quit IRC
[01:24:43] *** johnament has joined #jbosstesting
[02:11:09] *** aamonten has joined #jbosstesting
[02:14:19] *** jdlee has quit IRC
[02:35:05] *** Elisha has quit IRC
[02:40:01] *** aamonten has quit IRC
[02:44:26] *** lightguard_jp has quit IRC
[02:56:07] *** ldimaggi has joined #jbosstesting
[03:54:40] *** johnament has quit IRC
[04:12:26] *** bleathem has quit IRC
[04:12:44] *** bleathem has joined #jbosstesting
[04:20:16] *** bleathem has quit IRC
[04:25:45] *** bleathem has joined #jbosstesting
[05:32:06] *** ldimaggi has quit IRC
[06:16:34] *** lightguard_jp has joined #jbosstesting
[06:18:20] *** lightguard_jp has quit IRC
[06:35:01] *** rruss has quit IRC
[06:35:07] *** jdlee has joined #jbosstesting
[06:35:08] *** jdlee has joined #jbosstesting
[06:53:23] *** maeste has joined #jbosstesting
[06:56:03] *** rruss has joined #jbosstesting
[07:01:14] *** michaelschuetz has joined #jbosstesting
[07:05:38] *** michaelschuetz has quit IRC
[07:22:28] *** oskutka has joined #jbosstesting
[07:32:24] *** wolfc has joined #jbosstesting
[07:41:20] *** timte has joined #jbosstesting
[08:14:44] *** alesj has joined #jbosstesting
[08:16:53] *** maeste has quit IRC
[08:17:38] *** kpiwko has joined #jbosstesting
[08:23:08] *** Elisha_ has joined #jbosstesting
[08:31:08] *** Elisha_ is now known as Elisha
[08:35:51] *** aslak has joined #jbosstesting
[08:35:51] *** aslak has quit IRC
[08:35:51] *** aslak has joined #jbosstesting
[08:45:09] *** mgoldmann has joined #jbosstesting
[08:57:09] *** bleathem has quit IRC
[09:03:57] *** maschmid has joined #jbosstesting
[09:15:26] *** rruss has quit IRC
[09:31:12] *** Jaikiran has joined #jbosstesting
[09:48:19] *** jeand has joined #jbosstesting
[10:00:16] *** michaelschuetz has joined #jbosstesting
[10:39:37] *** michaelschuetz1 has joined #jbosstesting
[10:41:39] *** michaelschuetz has quit IRC
[10:54:49] *** Jaikiran has quit IRC
[11:11:08] *** maeste has joined #jbosstesting
[11:28:00] *** michaelschuetz has joined #jbosstesting
[11:30:04] *** vtunka has joined #jbosstesting
[11:30:35] *** michaelschuetz1 has quit IRC
[11:58:32] *** alesj has quit IRC
[12:07:33] *** aslak has quit IRC
[12:17:24] *** alesj has joined #jbosstesting
[12:18:55] *** aslak has joined #jbosstesting
[12:18:55] *** aslak has quit IRC
[12:18:55] *** aslak has joined #jbosstesting
[12:34:50] *** maeste is now known as maeste_lunch
[13:02:04] *** ldimaggi has joined #jbosstesting
[13:16:57] *** maeste_lunch is now known as maeste
[13:55:03] *** pmuir has joined #jbosstesting
[14:17:24] *** ldimaggi has quit IRC
[14:18:41] *** ldimaggi has joined #jbosstesting
[14:48:03] *** kpiwko1 has joined #jbosstesting
[14:48:50] *** kpiwko has quit IRC
[14:49:22] *** kpiwko1 is now known as kpiwko
[15:25:52] *** michaelschuetz has quit IRC
[15:26:33] *** michaelschuetz has joined #jbosstesting
[15:27:47] *** pmuir has quit IRC
[15:30:51] *** michaelschuetz has quit IRC
[15:31:22] *** michaelschuetz has joined #jbosstesting
[15:36:46] *** lightguard_jp has joined #jbosstesting
[15:40:44] *** michaelschuetz has quit IRC
[15:41:30] *** michaelschuetz has joined #jbosstesting
[15:46:14] *** bambitroll has joined #jbosstesting
[15:46:25] *** bambitroll has left #jbosstesting
[15:54:44] *** Jaikiran has joined #jbosstesting
[16:00:38] *** maeste has quit IRC
[16:03:07] *** rruss has joined #jbosstesting
[16:07:32] *** alesj has quit IRC
[16:22:47] *** jdlee has quit IRC
[16:24:01] *** bleathem has joined #jbosstesting
[16:27:57] *** bgeorges has joined #jbosstesting
[16:28:45] *** oskutka has quit IRC
[16:30:56] *** lfryc has quit IRC
[16:40:03] *** ALR has joined #jbosstesting
[16:46:37] *** pmuir has joined #jbosstesting
[16:46:37] *** pmuir has quit IRC
[16:46:37] *** pmuir has joined #jbosstesting
[17:04:08] *** jdlee has joined #jbosstesting
[17:04:08] *** jdlee has joined #jbosstesting
[17:04:37] *** maschmid_ has joined #jbosstesting
[17:06:41] *** maschmid has quit IRC
[17:12:54] *** mgoldmann has quit IRC
[17:15:20] <michaelschuetz> pmuir: hi
[17:15:39] <michaelschuetz> pmuir, do you have time for quick seam2 question?
[17:19:04] *** maschmid_ is now known as maschmid
[17:21:56] *** jdlee has quit IRC
[17:47:42] *** jdlee has joined #jbosstesting
[17:52:59] *** jasondlee has joined #jbosstesting
[17:53:00] *** jdlee has quit IRC
[17:57:22] *** jasondlee has quit IRC
[17:57:22] *** jasondlee has joined #jbosstesting
[18:08:48] *** jbossbot has quit IRC
[18:09:15] *** jbossbot has joined #jbosstesting
[18:22:28] *** ALR has quit IRC
[18:24:15] *** michaelschuetz has quit IRC
[18:52:39] *** vtunka has quit IRC
[18:55:46] *** kpiwko has quit IRC
[19:01:31] *** maschmid has quit IRC
[19:21:40] *** jasondlee has quit IRC
[19:28:43] *** bleathem has quit IRC
[19:43:02] *** ALR has joined #jbosstesting
[20:04:32] *** Jaikiran has quit IRC
[20:47:42] *** bleathem has joined #jbosstesting
[20:51:13] *** jeand has quit IRC
[21:11:40] *** jdlee has joined #jbosstesting
[21:18:37] *** bleathem has quit IRC
[21:39:46] *** michaelschuetz has joined #jbosstesting
[21:41:33] *** jdlee has quit IRC
[21:47:30] *** jdlee has joined #jbosstesting
[21:58:30] *** Tashtego has joined #jbosstesting
[22:02:33] *** aslak has quit IRC
[22:10:00] *** aslak has joined #jbosstesting
[22:10:01] *** aslak has quit IRC
[22:10:01] *** aslak has joined #jbosstesting
[22:10:07] *** lightguard_jp has quit IRC
[22:26:10] *** bleathem has joined #jbosstesting
[22:31:28] *** bleathem has quit IRC
[22:31:40] *** jbossbot has quit IRC
[22:31:55] *** jbossbot has joined #jbosstesting
[22:36:49] <nickarls> michaelshuetz: throw the question up in the air and see if anyone grabs.
[23:11:48] *** ALR has quit IRC
[23:12:51] *** ALR has joined #jbosstesting
[23:15:43] <michaelschuetz> nickarls, cheers.
[23:15:49] <michaelschuetz> seam2 question:  why is asynchronous interceptor "client-typed" ?
[23:20:13] <nickarls> hmm, someone thought it must run before any ejb interceptors. don't know the reason...
[23:24:02] *** reza has joined #jbosstesting
[23:24:21] <reza> aslak, you there?
[23:24:48] <aslak> reza, heya
[23:25:28] <reza> I have a couple of things to discuss..
[23:25:42] <reza> Have you looked at JavaLobby today?
[23:26:05] <reza> Take a look at the article on CDISource
[23:26:07] <aslak> nope
[23:26:59] <aslak> aa, yea.. i saw Andy has created 'something'.. didn't read it to carefully
[23:26:59] <reza> One of our guys, Rick is involved with it.
[23:27:30] <aslak> tried to fine some source, but it seemed to be a numberguess something..
[23:27:49] <aslak> what is it? new Seam? or just a common space for knowledge+
[23:27:50] <aslak> ?
[23:27:51] <reza> The part that concerns me is that Andy is apparently not too happy with Arquillian and hence is creating his own solution (sort of)
[23:28:36] <aslak> he lost faith in Arq as well..
[23:28:43] <reza> The two concerns that I gathered is lack of conversation support, stability issues and complexity?
[23:29:04] <reza> Have you talked to him about it?
[23:29:12] <reza> What's the story there?
[23:29:21] *** maschmid has joined #jbosstesting
[23:29:43] *** jose_freitas has joined #jbosstesting
[23:29:54] *** jose_freitas has left #jbosstesting
[23:30:06] <aslak> reza, i have not spoke to him about it.. he contributed to arq and has shown interest in it. that's all i know
[23:30:40] <aslak> reza, it took some time to get the patch in since we were in the middle of a rewrite
[23:31:01] <reza> The part that really bothers me is that his testing framwork is only supposed to support CDI and nothing else
[23:31:22] <aslak> yea, he joins the lines of javaee-junit or what ever it's called
[23:31:34] <reza> I guess that means Weld SE and OpenWebBeans withoout OpenEJB...
[23:32:29] *** wolfc has quit IRC
[23:32:31] <aslak> yea
[23:33:38] <aslak> we don't leck conversation scope tho.. but besides the point
[23:33:54] <aslak> (and where does conversation scope fit into a pure SE CDI anyway)
[23:33:57] <reza> You don't? Is it documented?
[23:34:06] <michaelschuetz> nickarls, thanks for your reply. yeah, the reason would be interesting
[23:34:29] <aslak> reza, should be, the Weld EE mock container supports conversationScope if you activate it.
[23:34:30] <reza> I agree about conversations not fitting in SE..
[23:35:03] <aslak> but it will most likely only work in Weld EE mock and not in a real container..
[23:35:13] <aslak> http://docs.jboss.org/arquillian/reference/latest/en-US/html_single/#container.weld-ee-embedded-11.configuration
[23:35:24] <aslak> anyway.. that's besides the point
[23:35:29] <aslak> what are their goals ?
[23:35:49] *** ldimaggi has quit IRC
[23:36:03] <reza> OK, let me take a look so I can talk to him more.
[23:36:28] <reza> Is there a technical challenge to enabling it in a real container?
[23:36:39] <reza> Or just a low-priority item?
[23:36:59] <reza> I'm not too clear on his goals - tried hard to elicit it.
[23:37:33] <aslak> well, the scopes in the real container ionly works by luck atm, you can't control them. we have a guy on the case of making a controllable unified Context api between the different containers, weld, openwebbeans etc.. but it's not done yet
[23:37:42] <reza> The best I could gather is the concerns around conversations, stability and complexity (could not get more details on what the specifics are).
[23:38:07] <reza> That's what I thought..
[23:38:36] <aslak> the complexity i assume he is talking about dependencies, which comes from the containers not arq really. and ShrinkWrap to define the deploymetns
[23:38:57] <aslak> both are to valid 'ish' points that we are looking into.
[23:39:01] <aslak> to/two
[23:39:26] *** bleathem has joined #jbosstesting
[23:39:43] <reza> What about stability, anything to it?
[23:40:22] <reza> It supripses me because I haven't sen any issues.
[23:40:44] <reza> I don't even see the issue with depepencies.
[23:41:10] <reza> I do think there has to be another option besides ShrinkWrap...
[23:42:13] *** mojavelinux has joined #jbosstesting
[23:43:01] <reza> We scan the classpath/file system and deploy what we find for the Resin-specific test framework.
[23:43:35] <aslak> reza, yea i think that is what most ppl do.. i find that a bit insane tho but that's probably just my test
[23:43:38] <aslak> teat
[23:44:04] <aslak> taste damn it
[23:44:05] <aslak> :)
[23:44:32] <aslak> you also lose a lot of possibilities
[23:44:52] <reza> There are potential issues with dumping the entire class-path into the deployment.
[23:45:06] <aslak> reza, yea, tons..
[23:45:12] <aslak> that's why we don't
[23:45:13] <reza> We did it just for "ease-of-use"
[23:45:44] <mojavelinux> my personally feeling is that andy needs to be *a bit* more patient
[23:46:02] <aslak> we are looking at trying to scan for dependent classes etc, but there is really no 100% bullet proff way of doing it
[23:46:05] <mojavelinux> he seems to want to jump onto the next thing before we really can get the discussion started, or even part done
[23:46:21] <reza> We do allow the specification of alternate XML deplyment descriptors (CDI, EJB, JPA and Resin config).
[23:46:39] <mojavelinux> any problem he wants to solve, we can solve in arquillian, there is no need to go start another project
[23:47:29] <mojavelinux> in fact, the question about stability is largely due to us (or aslak really) reworking the arquillian core to not make so many assumptions
[23:47:41] <reza> I agree on the guessing the deployemtn  part...
[23:47:41] <mojavelinux> I mean, conversation support is really not that big of a thing to add and I realize we need to do it
[23:47:44] <aslak> it's super fast to get a 80% solution running on Weld SE, but then comes the 100% v that makes it useful.. that's a completely different story
[23:47:55] <nickarls> besides, how can one call for stability in an alpha?
[23:48:01] <reza> The trick I think is just to make the user aware of the potential hazards
[23:48:08] <mojavelinux> seriously
[23:48:22] <mojavelinux> we are holding back so that we get the requirements correct
[23:48:37] <mojavelinux> the biggest "disagreement" we've had so far within the community
[23:48:42] <mojavelinux> is how to include things on the classpath
[23:48:53] <mojavelinux> and what we've tried to communicate, at least as a starting point
[23:49:03] <mojavelinux> is that the current way of testing is totally screwed up
[23:49:17] <mojavelinux> so while the shrinkwrap archive may seem extreme, that's just coming in from the other end of the spectrum
[23:49:43] <mojavelinux> and it's also about the goal of liberating the test from the build, which you simply cannot ignore how powerful that is for integration testing
[23:49:57] <mojavelinux> as I watch people abandon ides when they start doing any sort of complex testing
[23:50:13] <nickarls> shrinkwrap shorthand functions can easily be written
[23:50:16] <mojavelinux> I think we can certainly add a "collector" in arquillian
[23:50:30] <mojavelinux> that would say "add all project compiled resources and classes"
[23:50:49] <mojavelinux> it's really not that big of a deal...what we have been trying to do is get people away from their habit of starting there
[23:51:12] <nickarls> once JBT supports right-clikcing a file and it scans it for imports and collects stuff from classpath (including maven repo) and makes the @Deployment method, we're done ;-)
[23:51:21] <mojavelinux> after all, the resources are sitting right there in target/classes and target/test-classes
[23:51:45] <mojavelinux> it's easy to put them into the shrinkwrap archive
[23:51:52] <mojavelinux> right, what nik said
[23:52:00] <mojavelinux> we want intelligence and hopefully it will come from the IDE
[23:52:12] <nickarls> containers with AS7-type startup speeds help, too
[23:52:21] <mojavelinux> calculating dependencies at runtime is much, much harder than people realize
[23:52:26] <mojavelinux> meaning, add this class and it's dependencies
[23:52:28] <nickarls> since in-container testing on an AS 6 is a bit... time consuming currently
[23:52:33] <mojavelinux> because the answer is, it depends on what you want
[23:52:53] <mojavelinux> yes, people will forget about AS6 in about 2 - 3 months ;)
[23:53:00] <mojavelinux> not that I don't appreciate it for what it is providing now, trust me
[23:53:10] <mojavelinux> I'm just saying, it's going to be way better to have that quick startup
[23:53:34] <mojavelinux> and yes, also to nik's point
[23:53:36] <reza> That is something Andy compianed about as well...
[23:54:03] <reza> I can't speak to it since I haven't tried JBoss embedded...
[23:54:04] <mojavelinux> it's not that hard to have a giftwrap api over shrinkwrap...in fact, I see a Groovy Arquillian test as being the ideal for shorthand archive setup
[23:54:12] <nickarls> auto-generation of @Deployment method is doable but it required scanning of maven deps since the repo api doesn't have any "what provides" functionality
[23:54:21] <mojavelinux> but we are also solving another *huge* problem with embedded containers
[23:54:33] <mojavelinux> when an embedded container starts, it's totally broken in the current model
[23:54:44] <mojavelinux> of how poeple are integration testing
[23:54:51] <mojavelinux> and arquillian <= 1.0.0.Alpha4
[23:55:06] <mojavelinux> because you absolutely do not realize how screwed up it is to start a container under the current project's classpath
[23:55:13] <mojavelinux> there are no guarnatees
[23:55:16] <nickarls> I usually use managed, since I manage to get CP errors with embedded usually
[23:55:21] <reza> How much longer do you think you guys are going to be in Alpha?
[23:55:28] <reza> How long has it been now?
[23:55:30] <mojavelinux> so what we've done in alpha5
[23:55:32] *** maschmid has quit IRC
[23:55:39] <mojavelinux> is to have a classpath isolation mechanism
[23:55:46] <mojavelinux> where the embedded container is starting clean(er)
[23:55:49] <mojavelinux> we can even do better
[23:56:07] <mojavelinux> reza alpha5 was going to be beta1
[23:56:17] <mojavelinux> but there was too big of an internal change, so we decided on alpha5
[23:56:17] <reza> OK...
[23:56:25] <mojavelinux> and now we are going to release more steadily
[23:56:35] <mojavelinux> hopefully getting to final by May
[23:56:36] <nickarls> so...2-3 weeks? (fishing for a date)
[23:56:37] <reza> I'm justr trying to understand what's going on...
[23:56:39] <mojavelinux> or at least into the RCs by then
[23:56:50] <aslak> ALR loves Embedded, but to me as wel.. it's insane
[23:57:04] <mojavelinux> the big delay between alpha4 and alpha5 was for a rewrite
[23:57:05] <mojavelinux> of the core
[23:57:16] <mojavelinux> which had to happen in order to support the requirements the community presented to us
[23:57:20] <reza> It's a lot easier for us since we just need to worry about Resin and have almost complete control.
[23:57:30] <aslak> and OSGi doesn't help one bit. add slf4j 1.6 to your project and see glassfish embedded blow up
[23:57:31] <mojavelinux> well, that's another important fact
[23:57:44] <mojavelinux> alpha5 gives us much, much deeper control over the individual servers
[23:58:14] <aslak> initialize JUL before you start AS7 embedded and see that blow up
[23:59:10] <reza> Is it worth it to reach oit to Andy?
[23:59:50] <reza> I don't see how he won't be running into the same issues (unless being "CDI only" helps matters)...





top