[00:02:51] *** ldimaggi_ has quit IRC
[00:03:13] *** ldimaggi_ has joined #jbosstesting
[00:08:21] *** jeand has quit IRC
[00:21:47] *** aslak has quit IRC
[00:43:33] *** bleathem has quit IRC
[00:43:58] *** maeste has quit IRC
[00:54:37] *** jharting has quit IRC
[00:58:37] *** jesmon has left #jbosstesting
[01:18:00] *** lightguard_jp has quit IRC
[02:04:39] *** michaelschuetz1 has quit IRC
[02:47:21] *** ALR has quit IRC
[04:12:27] *** dmlloyd_ has joined #jbosstesting
[04:12:50] *** jbossbot has quit IRC
[04:13:10] *** jbossbot has joined #jbosstesting
[04:14:08] *** dmlloyd has quit IRC
[04:22:19] *** dmlloyd_ is now known as dmlloyd
[04:31:53] *** jdlee has quit IRC
[04:45:56] *** ldimaggi_ has quit IRC
[05:07:14] *** jdlee has joined #jbosstesting
[05:26:35] *** jdlee has quit IRC
[05:49:48] *** rruss has joined #jbosstesting
[06:14:50] *** mgoldmann has joined #jbosstesting
[06:20:10] *** mgoldmann has quit IRC
[06:47:30] *** rruss has quit IRC
[07:44:31] *** rruss has joined #jbosstesting
[08:09:07] *** oskutka has joined #jbosstesting
[08:30:19] *** aslak has joined #jbosstesting
[08:30:19] *** aslak has quit IRC
[08:30:19] *** aslak has joined #jbosstesting
[08:31:29] *** ge0ffrey has joined #jbosstesting
[08:43:43] *** lfryc has joined #jbosstesting
[08:46:03] *** GTobi has joined #jbosstesting
[08:53:11] *** lfryc has quit IRC
[08:57:06] *** lfryc has joined #jbosstesting
[08:59:23] *** aslak has quit IRC
[08:59:58] *** aslak has joined #jbosstesting
[09:19:23] *** Jaikiran has joined #jbosstesting
[09:44:00] *** michaelschuetz has joined #jbosstesting
[10:04:35] *** jharting has joined #jbosstesting
[10:20:13] *** michaelschuetz has quit IRC
[10:21:28] *** wolfc has joined #jbosstesting
[10:36:40] *** kpiwko has joined #jbosstesting
[11:11:17] *** maeste has joined #jbosstesting
[11:28:50] *** maeste has quit IRC
[11:42:05] *** maeste has joined #jbosstesting
[12:52:41] *** jdlee has joined #jbosstesting
[13:03:44] *** oskutka has quit IRC
[13:18:36] *** ldimaggi_ has joined #jbosstesting
[13:24:32] *** Jaikiran is now known as Jaikiran|AFK
[14:53:46] *** mgoldmann has joined #jbosstesting
[14:57:56] *** kpiwko has quit IRC
[15:02:50] *** al80 has joined #jbosstesting
[15:03:15] <al80> is glassfish-embedded 3.1.x supported in arquillian alpha 4?
[15:08:30] *** oskutka has joined #jbosstesting
[15:14:43] *** rruss has quit IRC
[15:19:56] <jdlee> al80: i don't think so
[16:03:36] *** mgoldmann has quit IRC
[16:22:11] <aslak> al80, not yet, coming in next v.
[16:53:00] *** rruss has joined #jbosstesting
[16:53:21] *** rruss has quit IRC
[16:55:19] *** rruss has joined #jbosstesting
[17:02:20] *** lfryc has quit IRC
[17:02:20] *** aslak has quit IRC
[17:03:22] *** aslak has joined #jbosstesting
[17:05:58] *** lfryc has joined #jbosstesting
[17:10:58] *** lfryc has quit IRC
[17:11:58] *** lfryc has joined #jbosstesting
[17:12:49] *** jharting has quit IRC
[17:17:36] *** maeste has quit IRC
[17:30:56] *** maeste has joined #jbosstesting
[17:42:38] *** mgoldmann has joined #jbosstesting
[17:52:42] *** maeste has quit IRC
[18:03:29] *** lfryc has quit IRC
[18:06:20] <bobmcw> ALR taking the day off?
[18:18:49] *** aslak has quit IRC
[18:19:37] *** aslak has joined #jbosstesting
[18:24:17] *** lfryc has joined #jbosstesting
[18:29:09] *** jdlee has quit IRC
[18:34:07] *** al80 has quit IRC
[18:35:32] *** GTobi has quit IRC
[18:36:24] *** jdlee has joined #jbosstesting
[18:36:24] *** jdlee has joined #jbosstesting
[18:48:38] *** Jaikiran|AFK has quit IRC
[19:03:52] *** oskutka has quit IRC
[19:24:06] *** ge0ffrey has quit IRC
[20:04:33] *** lfryc has quit IRC
[20:25:15] *** bobmcw has quit IRC
[20:25:36] *** bobmcw has joined #jbosstesting
[20:26:01] *** jeand has joined #jbosstesting
[20:33:23] *** ALR has joined #jbosstesting
[20:41:36] <aslak> ALR, good evening..
[20:41:39] <aslak> ALR, comment ? SHRINKWRAP-254
[20:41:40] <jbossbot> jira [SHRINKWRAP-254] Support creation of URLs from a Archive [Open, Major, Aslak Knutsen] https://issues.jboss.org/browse/SHRINKWRAP-254
[20:41:52] <ALR> Lookyloo
[20:42:04] <ALR> WTF is THIS?!
[20:42:06] <ALR> Wikiness?
[20:42:27] <aslak> wikiness?
[20:42:28] *** rruss has quit IRC
[20:42:42] <ALR> Yeah, Wiki formatting in the JIRA.
[20:42:47] <ALR> Preetttty.
[20:42:59] <ALR> Anyway.
[20:43:06] <ALR> No StreamExporter.
[20:43:08] <ALR> *Not
[20:43:10] <aslak> aa, yea.. confluence wiki render support activated.. finally
[20:43:15] <ALR> That exports things as streams.
[20:43:23] <ALR> URLs are locators.
[20:43:49] <aslak> could make it a separate view
[20:44:06] <aslak> archive.as(URLSomething.class).toURL()
[20:44:22] <aslak> archive.as(URLSomething.class).toURL(Class<? extends StreamExporter>)
[20:44:49] <ALR> What do you imagine the URL would be?
[20:44:50] <aslak> i figured on the StreamExporter since it is used
[20:45:37] <aslak> ALR, basically the same as we do in SWCL
[20:46:11] <aslak> http://shrinkwrap.pastebin.com/Z7xCcfgd
[20:46:56] <ALR> So an archive:// URL with a StreamHandler
[20:47:01] <aslak> yea
[20:47:18] <ALR> How will the thing consuming the URL know which Handler to register?
[20:47:20] <ALR> Or use?
[20:47:45] <aslak> could of course make a proper URL protocol impl trying to fetch form some static repo where we store Archives that have been URL'ified, but..
[20:48:20] <aslak> ALR, handler? the creator of the URL makes the handler
[20:48:50] <aslak> URL.openStream -> URLStreamHandler -> archive.export
[20:49:07] <ALR> Umm...
[20:49:11] <ALR> Thinking here.
[20:49:27] <ALR> We used to run into this w/ vfs:// protocols.
[20:49:33] <aslak> or new URL(archive.toURL(), "some/path/inside.txt")
[20:49:49] <ALR> There was some issue where the handlers had to be installed before these were ever used..
[20:50:05] <aslak> ALR, not if you specify the URLStreamHandler
[20:50:30] <ALR> Aha
[20:50:31] <ALR> "The URL is backed by a custom {@link URLStreamHandler} and can not be converted to String and recreated."
[20:50:34] <ALR> That's your gotcha.
[20:50:44] <aslak> it will try to look for a URLStreamHandler for the protocol if you randmly create a url.. e.g. new URL("tada://") -> exception, handler for tada not found
[20:51:00] <ALR> Making me not so much a fan of this.
[20:51:06] <ALR> As it's now a nonportable URL.
[20:51:19] <aslak> portable where?
[20:51:36] <ALR> By String serialization.
[20:51:40] <aslak> a archive is a temporary memory rep of resources, the URL is the same
[20:51:41] <ALR> Thinking more on it.
[20:52:27] <ALR> For instance, is it a reasonable or contracted expectation that a URL be roundtripped via String format?
[20:52:33] <ALR> URL > String > URL ?
[20:53:16] <aslak> ppl keep doing it, but not sure its contract
[21:02:41] <ALR> aslak: I'm still looking :)
[21:03:23] <ALR> I think this is a user assumption
[21:03:35] <ALR> Based on that most URLs have baked-in handlers
[21:03:40] <ALR> file:// or http://
[21:04:15] <ALR> There's nothing that dictates a URL must be constructed from a String
[21:04:26] <ALR> That's the issue we used to have w/ VFS and some 3rdparty libs
[21:04:28] <ALR> They assumed
[21:04:32] <ALR> And we yelled at them
[21:04:33] <ALR> :)
[21:04:42] <ALR> Because they did stuff like new File(url)
[21:06:26] <aslak> yea i know
[21:07:32] <ALR> aslak: So here's a question:
[21:07:57] <ALR> If external APIs are relying on URLs to xfer across the wire, how does this help?
[21:08:12] <ALR> On the other end all comes across?
[21:15:44] <aslak> the zip stream of the content.. same as StreamExporter.exportTo...
[21:23:02] <ALR> aslak: But the handler is on the client side...
[21:42:29] <aslak> ALR, aa, the URL object is never sent to the server.. the client side opens the URL stream dn streams it to the server side
[21:43:12] <ALR> You're confusing "client" and "server" notions.
[21:43:30] <ALR> Streaming isn't a push.  It's a pull.
[21:43:31] <ALR> Always.
[21:44:10] <ALR> So take a simple example.
[21:44:22] <ALR> JUnit VM has the archive.
[21:44:35] <ALR> Remote AS instance does not.
[21:44:43] <ALR> JUnit connects to AS.
[21:44:54] <ALR> Passes it a URL to the archive via this mechanism.
[21:45:01] <ALR> URL now is on the AS instance.
[21:45:05] <aslak> no
[21:45:06] <ALR> What do you expect to happen?
[21:45:30] <aslak> the URL is on the client side of AS. the client then opens up communication with the server to stream the content
[21:46:16] <aslak> the copy content part in distributionManager tells the client to stream it, not send the url
[21:46:56] <ALR> "The client side of AS"?
[21:47:03] <aslak> profileservice client
[21:47:13] <ALR> So not the AS VM?
[21:47:20] <aslak> right, the junit vm
[21:47:43] <ALR> I see.
[21:47:55] <ALR> And distribution manager is on the client side.
[21:48:02] <aslak> yea
[21:48:52] *** kpiwko has joined #jbosstesting
[21:48:57] <ALR> aslak: What other cases need this aside from PS?
[21:49:11] <aslak> SW CL
[21:49:25] <ALR> aslak: That's internals.
[21:49:31] <ALR> We can hide behind the SW CL.
[21:50:41] <ALR> What I'm getting at is trying to determine if adding the complexity of the feature to SW is worth it.
[21:50:54] <ALR> Especially since the String form is useless without the Handler.
[21:51:00] <ALR> Because we're defining a custom protocol
[21:51:25] <ALR> (So we'd have to expose the protocol handler in the API too, likely)
[22:00:30] *** dmlloyd has quit IRC
[22:01:37] *** jbossbot has quit IRC
[22:05:43] *** dmlloyd has joined #jbosstesting
[22:05:45] *** jbossbot has joined #jbosstesting
[22:13:20] *** jbossbot has quit IRC
[22:14:29] *** jbossbot has joined #jbosstesting
[22:40:31] *** wolfc has quit IRC
[22:45:42] *** jeand has quit IRC
[23:08:51] <aslak> ALR, oh, gf embedded as well.
[23:09:16] <aslak> ALR, not that we can't give them a file url in that case, but.. then gf can take care of where to put it
[23:09:31] *** nthngrtr has joined #jbosstesting
[23:10:58] <nthngrtr> does anyone know the status of when we will be able to use @EJB attributes such as mappedName for injection in Arquillian?
[23:12:06] <aslak> nthngrtr, i'll try to get it into the next release, out in a couple of weeks
[23:15:27] <jdlee> doesn't the container handle the injection?
[23:15:41] <nthngrtr> awesome
[23:16:30] <aslak> jdlee, depends.. for pure ejb with no cdi, arq tries to find the ejbs to inject. but there is no way to get Interface -> JNDIName in a standard way
[23:16:48] <aslak> jdlee, the TestCase it self is not managed
[23:16:56] <jdlee> oh! right
[23:17:22] <jdlee> so Arq handles the EntityManager injection, etc. into the test case classes?
[23:17:49] <aslak> jdlee, no, we don't support EntityManager alone. but via CDI we do
[23:18:18] <aslak> if a BeanManager is found, we do a non contetual injection on the whole TestClass, which means anything CD ican inject will be
[23:18:30] <aslak> contextual
[23:18:34] <jdlee> so @PersistenceContext doesn't work in the test class?
[23:18:49] <jdlee> or not directly? it's supported but CDI is doing the lifting?
[23:18:59] <aslak> right, if cdi does the lifting
[23:19:20] <jdlee> ok.  that makes perfect sense :)
[23:20:41] <aslak> jdlee, there is nothing stopping us from creating a PersistenceContext enricher as well, just havn't done it yet. for EE6 servers, adding beans.xml / using CDI is probably what you will be doing anyway so
[23:21:07] <aslak> it would be for ee5/ee4 containers.. not been prioritized
[23:22:20] * jdlee nods
[23:48:52] *** kpiwko has quit IRC
[23:59:28] *** lightguard_jp has joined #jbosstesting





top