[00:00:08] <dmlloyd> bstansberry: https://github.com/bstansberry/jboss-as/commit/bae503847c018b01c262479c7d9fe92c496f16da#L3R67 looks wrong
[00:00:08] <jbossbot> git [jboss-as] bae5038.. bstansberry at jboss dot com Don't attempt composite rollback if no steps provide an rollback op...
[00:00:58] <jbossbot> git [jboss-as] push master 9e75d09.. Rémy Maucherat Add AS 6 workaround to allow context-root path which don't start with '/'.
[00:00:58] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/3a1c3bd...9e75d09
[00:01:16] <bstansberry> the OFE?
[00:01:29] <dmlloyd> yeah the formats
[00:01:43] <rmaucher> thanks !
[00:01:44] <dmlloyd> numbers should be %d, and streamIndex is in there twice instead of the max whatever
[00:01:55] <dmlloyd> and there probably could be a %s for INPUT_STREAM_INDEX too
[00:02:07] * dmlloyd will paint your bikeshed
[00:02:17] <bstansberry> that's a constant, that's why I didn't bother
[00:02:25] <dmlloyd> yeah wasn't 100% about that
[00:02:42] <dmlloyd> the max value at least looks wrong though
[00:02:51] <bstansberry> doh, you're right!
[00:02:53] <bstansberry> thanks
[00:07:06] <jbossbot> git [jboss-as] push master a627097.. bstansberry at jboss dot com Don't attempt composite rollback if no steps provide an rollback op...
[00:07:06] <jbossbot> git [jboss-as] push master URL: http://github.com/jbossas/jboss-as/compare/9e75d09...a627097
[00:07:10] *** bstansberry is now known as bstans_afk
[00:49:19] *** emuckenhuber has quit IRC
[00:53:06] *** miclorb has joined #jboss-as7
[00:59:04] *** alesj has quit IRC
[01:02:23] *** emuckenhuber has joined #jboss-as7
[01:02:23] *** ChanServ sets mode: +v emuckenhuber
[01:09:26] *** rmaucher has quit IRC
[01:34:45] *** jamezp has left #jboss-as7
[01:40:50] * smarlow changing to inject the JPA DataSources...
[01:46:40] *** aslak has quit IRC
[01:47:53] <smarlow> baileyje:  the DataSource injection works perfectly!
[01:54:00] <smarlow> stuartdouglas:  I'm not sure if you saw my earlier question about whether "Attachments.COMPOSITE_ANNOTATION_INDEX" will likely be built during the PARSE phase during the next few weeks?  JPAAnnotationParseProcessor is currently using the COMPOSITE_ANNOTATION_INDEX and is part of POST_MODULE as a result.  Maybe this would be different if I followed your advice and turned the loop around (to be on components instead).
[02:45:02] *** sguilhen has left #jboss-as7
[02:56:30] *** frainone has joined #jboss-as7
[02:56:30] *** ChanServ sets mode: +v frainone
[02:58:01] <stuartdouglas> smarlow: If you turn the loop around you should not need the composite annotation index
[03:25:22] *** jpearlin has joined #jboss-as7
[03:32:46] <smarlow> stuartdouglas:  fair enough, that gets me out of both jams then :)
[03:41:20] *** jpearlin1 has joined #jboss-as7
[03:42:09] *** jpearlin has quit IRC
[03:44:04] <smarlow> stuartdouglas:  that flipped around easily enough, I like that AbstractComponentConfigProcessor.  Thanks for the improvement!
[04:14:57] *** smarlow has quit IRC
[04:24:11] *** smarlow has joined #jboss-as7
[04:24:11] *** ChanServ sets mode: +v smarlow
[04:40:46] <baileyje> smarlow: Great
[04:44:21] <smarlow> baileyje:  thanks for the help, that will save two more jndi lookups!
[04:51:03] *** jpearlin1 has left #jboss-as7
[04:52:48] *** frainone has quit IRC
[04:59:05] *** bobmcw has joined #jboss-as7
[04:59:05] *** ChanServ sets mode: +v bobmcw
[05:03:21] *** bobmcw has quit IRC
[05:04:55] *** bobmcw has joined #jboss-as7
[05:04:55] *** ChanServ sets mode: +v bobmcw
[05:14:53] *** miclorb has quit IRC
[05:54:32] *** bgeorges has joined #jboss-as7
[05:56:18] <smarlow> dmlloyd:  I'm nuking the org.slf4j.slf4j-module (duplicates of org.slf4j module) that you mentioned a few days ago or so...
[06:04:43] <smarlow> dmlloyd:  I squashed the above change and a few other small ones.  If your around this weekend, could you please pull https://github.com/scottmarlow/jboss-as/commit/90e8d1e2d913bea0149d3a028b0ae02d7e787ae3
[06:04:45] <jbossbot> git [jboss-as] 90e8d1e.. Scott Marlow switch to using DataSource injection, flipped the loop in JPAAnnotationParseProcessor, improved service names and addressed JBAS-9034 (allow JPA operations in deployments that use JPA annotations that reference external applications persistence.xml),delete duplicate module org.slf4j.slf4j-api
[06:04:46] <jbossbot> jira [JBAS-9034] Applications should be able to inject a persistence context/unit from a separate deployment by using a fully scoped (appname#puname) reference [Coding In Progress (Unresolved) Sub-task, Major, Scott Marlow] https://issues.jboss.org/browse/JBAS-9034
[06:07:44] *** bgeorges has quit IRC
[06:21:52] *** smarlow has quit IRC
[07:20:35] *** bobmcw has quit IRC
[08:22:20] *** bgeorges has joined #jboss-as7
[11:12:04] *** alesj has joined #jboss-as7
[11:22:30] *** bgeorges has quit IRC
[12:44:36] *** alesj has quit IRC
[12:48:39] *** aslak has joined #jboss-as7
[12:48:39] *** aslak has joined #jboss-as7
[12:48:39] *** ChanServ sets mode: +v aslak
[14:05:31] *** smarlow has joined #jboss-as7
[14:05:31] *** ChanServ sets mode: +v smarlow
[14:16:29] *** alesj has joined #jboss-as7
[14:51:30] <smarlow> dmlloyd:  One detail of plugging persistence providers is that they have their own dependencies as well.  Will have to think about the top level deployment case some more and how we resolve the dependencies.  I definitely need some per provider metadata (in config or code or both).
[14:52:14] <smarlow> I'm not sure if the metadata helps in this case, just wanted to point out that I need it for other reasons.
[15:20:11] *** mbg has joined #jboss-as7
[15:20:11] *** ChanServ sets mode: +v mbg
[15:28:29] <dmlloyd> smarlow: dependencies in this case are not transitive, *and* everything goes through the API so we technically don't have to add any classes to the JPA client classpath
[15:28:40] *** pferraro has quit IRC
[15:28:41] <dmlloyd> or I should say, transitive dependencies are not relevant
[15:39:28] <smarlow> dmlloyd:  when we come across the hibernate-entitymanager-3.6.1.Final.jar in the deploy folder, where are the jars that hibernate needs found?
[15:40:08] <smarlow> or same question for openjpa jar
[15:40:40] <dmlloyd> they would have to have proper class-path or dependencies set up like any deployment
[15:40:53] <dmlloyd> but odds are good that we're shipping most, if not all, of the support libraries it would need
[15:41:13] <dmlloyd> for hibernate we definitely would of course :)
[15:41:31] <smarlow> if I download either hibernate or openjpa, they both include their support libraries, so its easy for the user to know what they are
[15:42:23] <dmlloyd> the challenge is generally the problem where users add stuff we already ship, which can potentially result in CCEs
[15:42:24] <smarlow> dmlloyd:  I'm just not sure what the user does with the support libraries.  Do
[15:43:06] <smarlow> they copy them in standalone into the deploy folder also, next to their favorite persistence provider jar?
[15:43:17] <dmlloyd> they can, in many cases
[15:43:28] <dmlloyd> but in some cases that could cause some trouble
[15:43:45] <dmlloyd> they might be better off shipping a wrapper .jar with a jboss-structure.xml (to be renamed)
[15:44:07] <smarlow> they or we?  :)
[15:44:16] <dmlloyd> both
[15:44:34] <dmlloyd> if they deploy a JAR that happens to provide some service or contain some descriptor it could cause unexpected side-effects
[15:45:41] <dmlloyd> JBAS-8983
[15:45:42] <jbossbot> jira [JBAS-8983] Detect, filter, and act upon system APIs found in deployments [Open (Unresolved) Task, Major, Unassigned] https://issues.jboss.org/browse/JBAS-8983
[15:45:45] <dmlloyd> that is relevant too
[15:48:42] <smarlow> that is an interesting but tricky game (filtering org/slf4j and using our org.slf4j module :-)
[15:48:58] <dmlloyd> yeah.  we really need to consider modules on an individual basis
[15:49:34] <dmlloyd> it's just so common for users to do
[15:50:40] <smarlow> By the way, I cleaned up that org.slf4j.slf4j-module reference that you asked me.  Not sure if you saw my irc note from last night.
[15:51:08] <smarlow> my todo list is way under 100, but I still like to reduce it when possible.  ;)
[15:53:52] <dmlloyd> :)
[15:54:08] <dmlloyd> I put most of my list into JIRA now.
[15:55:33] <smarlow> dmlloyd:  I downloaded openjpa to have a look at it, their lib folder seems to contain some option elements also.  http://fpaste.org/1t5u/ is the contents.  Also, http://fpaste.org/cHEb/ is their MANIFEST.MF
[15:55:49] <smarlow> in case you are curious
[15:56:12] <dmlloyd> we have some of those but not others.  Also it's not really clear which modules we ship will be public...
[15:56:27] <dmlloyd> ah osgi stuff
[15:56:45] <dmlloyd> yeah I think we probably ought to recommend that people create container JARs for their JPA providers
[15:56:59] <dmlloyd> it's not something an average user will want to do anyway, one would hope
[15:57:25] <dmlloyd> the other alternative is to ship multiple providers but somehow I don't think Jason would go for that :)
[15:58:09] <smarlow> I don't want to either, its too painful over time as the providers change and people want new provider jars with their five year old AS 7...
[15:58:55] <smarlow> other reasons too
[15:59:22] <smarlow> I don't know much about the proprietary providers but people will want to use those too
[16:01:36] <smarlow> dmlloyd:  for the container jar that has a jboss-structure.xml, does that work as the mc doc says?  http://docs.jboss.org/jbossmc/docs/2.0.x/userGuide/ch21s04.html
[16:03:17] <dmlloyd> no it doesn't, it's actually a different schema altogether
[16:03:23] <dmlloyd> so we were talking about renaming that file
[16:03:28] <dmlloyd> to avoid confusion etc.
[16:03:40] <dmlloyd> in fact...
[16:04:35] <dmlloyd> JBAS-9066
[16:04:36] <jbossbot> jira [JBAS-9066] Rename "jboss-structure.xml" to "jboss-deployment-structure.xml" [Open (Unresolved) Task, Major, Unassigned] https://issues.jboss.org/browse/JBAS-9066
[16:04:38] <dmlloyd> :-)
[16:07:59] <smarlow> is there a jira for lifecycle callback interceptors also that I can follow?
[16:08:26] <dmlloyd> hmmm should be, let me look
[16:08:49] <dmlloyd> guess not
[16:08:52] * dmlloyd makes one
[16:10:08] * smarlow thanks!
[16:10:15] <dmlloyd> JBAS-9067
[16:10:16] <jbossbot> jira [JBAS-9067] Lifecycle methods on interceptor objects not functioning [Open (Unresolved) Bug, Critical, Unassigned] https://issues.jboss.org/browse/JBAS-9067
[16:10:33] <dmlloyd> tbh I have another notepad around here somewhere with more stuff
[16:10:44] <dmlloyd> next week will be a busy week...
[16:14:45] * smarlow cool, I added dependencies on the above jiras from the JPA jiras :)
[16:16:03] * smarlow walking away to get ready for the day
[16:23:23] *** jpearlin has joined #jboss-as7
[16:26:52] *** jpearlin1 has joined #jboss-as7
[16:28:04] *** jpearlin has quit IRC
[16:36:03] *** mbg has quit IRC
[16:51:09] *** alesj has quit IRC
[17:16:07] *** mbg has joined #jboss-as7
[17:16:08] *** ChanServ sets mode: +v mbg
[17:24:43] *** pferraro has joined #jboss-as7
[17:24:43] *** ChanServ sets mode: +v pferraro
[17:24:56] *** pferraro has quit IRC
[17:45:53] *** ccrouch has quit IRC
[17:50:52] *** bobmcw has joined #jboss-as7
[17:50:52] *** ChanServ sets mode: +v bobmcw
[17:53:17] *** ccrouch has joined #jboss-as7
[17:53:17] *** ChanServ sets mode: +v ccrouch
[17:53:31] *** bobmcw has quit IRC
[18:21:50] *** mbg is now known as mbg|away
[18:35:25] *** ccrouch has quit IRC
[19:12:09] *** ccrouch has joined #jboss-as7
[19:12:09] *** ChanServ sets mode: +v ccrouch
[19:22:52] *** bobmcw has joined #jboss-as7
[19:22:52] *** ChanServ sets mode: +v bobmcw
[19:25:32] *** bobmcw has quit IRC
[19:48:25] *** ccrouch has quit IRC
[20:02:24] *** ccrouch has joined #jboss-as7
[20:02:24] *** ChanServ sets mode: +v ccrouch
[21:15:23] *** alesj has joined #jboss-as7
[22:20:36] *** bstans_afk has quit IRC
[22:20:55] *** bstansberry has joined #jboss-as7
[22:20:55] *** ChanServ sets mode: +v bstansberry
[22:47:11] *** jpearlin1 has quit IRC
[22:57:13] *** alesj has quit IRC
[23:04:45] *** miclorb_ has joined #jboss-as7
[23:10:16] *** miclorb_ has quit IRC
[23:15:18] *** mbg|away is now known as mbg
[23:56:25] *** baileyje has quit IRC

   March 19, 2011  
< | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 | 11 | 12 | 13 | 14 | 15 | 16 | 17 | 18 | 19 | 20 | 21 | 22 | 23 | 24 | 25 | 26 | 27 | 28 | 29 | 30 | 31 | >