[00:09:12] *** antollinim_lunch has quit IRC
[00:28:19] *** igarashitm has joined #switchyard
[00:31:45] *** bfitzpat has quit IRC
[00:54:23] *** kcbabo has joined #switchyard
[00:54:23] *** kcbabo has joined #switchyard
[01:49:30] *** ldimaggi has joined #switchyard
[01:58:43] *** dbevenius has joined #switchyard
[02:06:42] *** dbevenius has quit IRC
[02:22:56] <tcunning> igarashitm: are you in tomorrow?
[02:23:14] <tcunning> (going to westford)
[02:29:30] <igarashitm> tcunning: yep, I'll get there about 9am
[02:29:47] <tcunning> igarashitm: i think i'm coming in, want to grab lunch?
[02:29:59] <igarashitm> tcunning: that's good :)
[02:32:17] <tcunning> igarashitm: cool - i'll msg you on here or try to come find you - i think i remember where you are sitting.
[02:33:53] <igarashitm> tcunning: ok, if you didn't find me, please call me in this IRC
[02:48:01] <rcernich> kcbabo: sun spots
[03:21:26] *** lance|afk has quit IRC
[03:21:51] *** lance|afk has joined #switchyard
[03:26:50] *** lance|afk has quit IRC
[04:04:05] <kcbabo> rcernich:  haha ? I've gone with that excuse once or twice
[04:05:02] <rcernich> kcbabo: seriously, there was a massive expulsion the last couple days
[04:05:13] <rcernich> kcbabo: northern lights were visible in atlanta
[04:06:16] <rcernich> kcbabo:  this is very odd.  nothing has changed and it's barfing on the maven tycho plugins
[04:07:37] <kcbabo> rcernich:  yeah, I don't know what the deal is there
[04:07:46] <kcbabo> rcernich:  just wiped the workspace and triggered a new build
[04:08:01] <kcbabo> rcernich:  my local build is fine, although I skip tests on OS X
[04:08:20] <rcernich> kcbabo: ahh, so it's kinda building on osx now?
[04:08:42] <kcbabo> rcernich:  yep, but it's always built on OS X if I skip tests
[04:08:57] <rcernich> kcbabo: interesting
[04:09:09] <rcernich> kcbabo: you'll have to send me the error info at some point
[04:09:28] <rcernich> kcbabo: tom had it failing resolving dependencies
[04:09:34] <rcernich> kcbabo: so i think you got further
[04:10:05] <kcbabo> rcernich:  I think it's still a dependency resolution issue - it's just in the tests
[04:11:32] <rcernich> kcbabo: interesting.  that's more to go on.
[04:12:04] <rcernich> kcbabo: worse comes to worse, i'll try bumping the tycho version to 0.13
[04:12:35] <rcernich> kcbabo: tools guys updated to .13, so maybe that's horking tycho somehow
[04:13:42] *** lance|afk has joined #switchyard
[04:19:15] *** lance|afk is now known as lanceball
[04:19:16] *** lanceball has joined #switchyard
[04:23:47] *** igarashitm has quit IRC
[04:26:09] <rcernich> kcbabo: ci.jboss.org is down
[04:26:53] <kcbabo> rcernich:  go tot he internal hudson instance
[04:27:00] <kcbabo> err ? go to
[04:27:12] <rcernich> kcbabo: yeah, just wondering if that might be contributing?
[04:27:35] <kcbabo> rcernich:  don't think so ? the internal instance just publishes to the external instance
[04:27:50] <kcbabo> rcernich:  now, it's possible that nexus is horked at the same time
[04:27:57] <kcbabo> rcernich:  guess we'll see if the release build fails
[04:28:13] <rcernich> kcbabo: i don't know what to say
[04:30:56] <kcbabo> rcernich:  I think the only thing to do at this point is to try again tomorrow
[04:31:12] <kcbabo> rcernich:  if it's a transient issue, then it will magically resolve itself
[04:31:27] <kcbabo> rcernich:  if not, maybe we can get fennelly to spin up a build on his EC2 instance
[04:31:55] <rcernich> kcbabo: i'll post something to jbosstools-dev
[04:33:59] <kcbabo> rcernich:  sounds good
[04:49:57] *** rcernich has quit IRC
[04:56:45] *** jliu has joined #switchyard
[05:02:50] *** kcbabo has quit IRC
[05:40:14] *** dbevenius has joined #switchyard
[05:54:45] *** ldimaggi has quit IRC
[06:02:23] *** dbevenius has quit IRC
[06:21:46] *** lanceball is now known as lance|zzz
[07:13:30] *** tcunning has quit IRC
[07:52:15] *** errantepiphany is now known as ee_zzz
[08:50:46] *** rbalent has joined #switchyard
[08:50:47] *** rbalent has joined #switchyard
[09:01:48] *** dbevenius has joined #switchyard
[09:49:59] *** rbalent has quit IRC
[10:13:25] *** rbalent has joined #switchyard
[10:13:26] *** rbalent has joined #switchyard
[10:32:23] *** jliu has quit IRC
[12:07:22] *** rbalent has quit IRC
[12:23:08] *** aslak has quit IRC
[12:27:32] *** rbalent has joined #switchyard
[12:38:54] *** kcbabo has joined #switchyard
[12:38:54] *** kcbabo has joined #switchyard
[12:42:15] *** tcunning has joined #switchyard
[12:51:21] *** ldimaggi has joined #switchyard
[12:59:19] *** tcunning has quit IRC
[13:26:43] <kcbabo> dbevenius:  hey dan - yt?
[13:27:03] <dbevenius> kcbabo: hey Keith, was just writing an email to you :)
[13:27:12] <kcbabo> dbevenius:  cool, that will work
[13:27:19] <dbevenius> kcbabo: I just tried this now and you are right....
[13:27:35] <dbevenius> kcbabo: for some reason I though the actual service impl would be part of the transaction...
[13:28:00] <dbevenius> kcbabo: but it end with the SwitchYardProducer?so I had it throw a runtime exception and redelivery kicked in
[13:28:35] <kcbabo> dbevenius:  so in SwitchYard, I would expect that the transaction would propagate to the service implementation
[13:28:42] <kcbabo> dbevenius:  but that would not be part of the Camel route
[13:29:38] <kcbabo> dbevenius:  so I would expect the transaction to commit after delivery has completed to the SwitchYard endpoint
[13:30:37] *** vinicius has joined #switchyard
[13:32:03] <kcbabo> dbevenius:  and then there are two possibilities
[13:32:43] <kcbabo> dbevenius:  1) the transaction has already been set to rollback only, so the transaction will not commit
[13:33:25] <kcbabo> dbevenius:  2) a runtime exception is generated from a SwitchYard handler and mapped to a fault.  This should also trigger a rollback.
[13:33:43] *** tfennelly has joined #switchyard
[13:34:05] <kcbabo> so case (2) is what the transaction policy is used for in Camel, IINM
[13:37:28] <dbevenius> kcbabo: I think I'm with you?let me verity a few things first and get back to you on this
[13:38:38] <kcbabo> dbevenius:  sure thing
[13:38:52] <kcbabo> dbevenius:  fwiw, I'm happy to just get (1) working first and then going from there
[13:49:09] <tfennelly> kcbabo: hey Keith... so are TC's forge changes pushed or not?
[13:49:19] <tfennelly> kcbabo: https://github.com/jboss-switchyard/core/pull/310
[13:49:42] <tfennelly> kcbabo: said pushed and then said waiting to push
[13:50:15] <tfennelly> kcbabo: doesn't look like they were pushed
[13:50:23] <kcbabo> tfennelly:  this is another wonderful thing about git and github
[13:50:39] <kcbabo> tfennelly:  I did push his changes earlier in the week
[13:50:49] <tfennelly> kcbabo: :)
[13:50:58] <tfennelly> kcbabo: and then rewrote history?
[13:50:58] <kcbabo> tfennelly:  then I realized that the error actually wipes out the switchyard.xml
[13:51:11] <tfennelly> kcbabo: ok
[13:51:14] <kcbabo> tfennelly:  so I felt we didn't want that hanging around
[13:51:18] <kcbabo> tfennelly:  ha!  right
[13:51:35] <kcbabo> tfennelly:  so I used 'git revert', which takes any commit in history and creates an "undo" commit
[13:51:37] <kcbabo> awesome
[13:51:50] <tfennelly> kcbabo: awesome
[13:51:51] <kcbabo> tfennelly:  I then pushed that and went into github and re-opened the pull request
[13:51:56] <kcbabo> tfennelly:  also re-opened the JIRA
[13:52:10] <kcbabo> tfennelly:  so now we are back at the same point before they were originally pushed
[13:52:12] <tfennelly> kcbabo: the catholic church here in Ireland should do all their publications through git
[13:52:13] <kcbabo> tfennelly:  super simple
[13:52:17] <kcbabo> tfennelly:  hehe
[13:52:29] <tfennelly> kcbabo: can rewrite history then when it suits them
[13:52:40] <tfennelly> kcbabo: and remove all evidence lol
[13:52:41] <kcbabo> tfennelly:  things have been pretty tame in here this week and you're threatening to take us to new lows!  ;-)
[13:52:50] <kcbabo> tfennelly:  if only tc was around
[13:52:57] <tfennelly> kcbabo: hahaha
[13:53:06] <tfennelly> kcbabo: it's nearly Friday
[13:53:14] <tfennelly> kcbabo: my last Friday
[13:53:26] <tfennelly> kcbabo: we gotta make it special
[13:53:39] <tfennelly> kcbabo: only one way... a new low
[13:54:03] <kcbabo> tfennelly:  true dat
[13:54:06] <kcbabo> tfennelly:  leave your mark
[13:54:20] <kcbabo> tfennelly:  FBI and HR will be sniffing around and you'll be long gone
[13:54:36] <tfennelly> kcbabo: exactly
[13:54:57] <tfennelly> kcbabo: drop a bomb just before I exit the building
[13:55:27] <tfennelly> kcbabo: so... how are TC's mods depedent on the piddly little mod I made to the BPM stuff?
[13:55:47] <kcbabo> tfennelly:  it's not specifically related to the BPM stuff
[13:55:53] <tfennelly> kcbabo: are are you saying that at all?
[13:56:03] <kcbabo> tfennelly:  any time we go to save the config, that error will surface
[13:56:04] <tfennelly> kcbabo: ok... just want me to test it:?
[13:56:18] <kcbabo> tfennelly:  yeah, just adding the facet and creating a service will be enough
[13:56:43] <tfennelly> kcbabo: ok
[13:57:02] <tfennelly> kcbabo: so you just want me to take it for a spin and make sure I don't see the errors I was seeing?
[13:57:09] <kcbabo> tfennelly:  right
[13:57:13] <tfennelly> kcbabo: before we push it
[13:57:16] <tfennelly> kcbabo: grand
[13:57:19] <tfennelly> kcbabo: will do
[13:57:24] <kcbabo> tfennelly:  thanks, tom
[13:57:32] <tfennelly> kcbabo: nps man
[13:58:03] <tfennelly> kcbabo: I issued pull requests for that unit test component activation stuff... not sure if you saw it or not
[13:58:15] <tfennelly> kcbabo: have a look and check it's what you wanted
[13:59:57] <kcbabo> tfennelly:  did see it, thanks ? trying to reason out the rest of this transaction stuff
[14:00:04] <kcbabo> tfennelly:  will definitely push it today though
[14:21:03] *** lance|zzz is now known as lanceball
[14:27:43] *** bfitzpat has joined #switchyard
[14:51:26] *** igarashitm has joined #switchyard
[15:02:13] *** bfitzpat is now known as bfitzpat_afk
[15:06:13] <gaYak> There's no discussion on the developer forum about transaction support..?
[15:12:35] <tfennelly> kcbabo: Keith... what exactly do you mean by "... until Beta3 is ready.."?
[15:12:44] <tfennelly> kcbabo: forge Beta3 is out there
[15:12:54] <tfennelly> kcbabo: and Beta4 I think
[15:13:41] <tfennelly> kcbabo: you mean the SwitchYard changes to bring it in line with Beta3?
[15:14:25] <tfennelly> kcbabo: ala https://issues.jboss.org/browse/SWITCHYARD-488
[15:16:53] <kcbabo> tfennelly:  yep
[15:17:12] <tfennelly> kcbabo: ok... just check'n
[15:17:18] <kcbabo> tfennelly:  so that's why I backed it out
[15:17:36] <kcbabo> tfennelly:  and it can wait in the queue until tom's other changes are ready
[15:17:43] <kcbabo> tfennelly:  then we'll push 'em all at once
[15:18:59] *** ee_zzz is now known as errantepiphany
[15:19:19] <tfennelly> kcbabo: sure... if it's OK with you, I'm going to change those JIRAs to make this clearer i.e. set 488 as a dependency of the others (Vs it just being "relates to")
[15:19:34] <kcbabo> tfennelly:  sounds great, thanks
[15:19:36] *** tcunning has joined #switchyard
[15:21:16] *** aslak has joined #switchyard
[15:30:06] *** bfitzpat_afk is now known as bfitzpat
[15:39:28] *** igarashitm has quit IRC
[15:45:06] *** igarashitm has joined #switchyard
[15:58:54] *** GitHub26 has joined #switchyard
[15:58:54] <GitHub26> [core] kcbabo pushed 1 new commit to master: http://git.io/S2SHTA
[15:58:54] <GitHub26> [core/master] Undoing fix for SWITCHYARD-508 until we have Forge Beta 3 support available. - Keith Babo
[15:58:54] *** GitHub26 has left #switchyard
[16:19:24] *** rbalent has quit IRC
[16:23:24] *** antollinim_lunch has joined #switchyard
[16:29:50] *** tcunning1 has joined #switchyard
[16:29:50] *** tcunning has quit IRC
[17:01:36] <kcbabo> dbevenius:  hey - what registry are we using right now in Camel component?
[17:01:45] <kcbabo> dbevenius:  looks to be loaded from a service loader
[17:02:15] <dbevenius> kcbabo: the CDIRegistry which falls back to the JndiRegistry
[17:02:31] <kcbabo> dbevenius:  oh, crap ? that's where it's at then
[17:02:43] <kcbabo> dbevenius:  I was looking for the META-INF/services entry
[17:04:14] <dbevenius> kcbabo: I'll poke around some more and see what I can find. Heading off shortly though so I'l continue on this tomorrow
[17:04:22] <kcbabo> dbevenius:  I'm wondering if we should just instantiate that directly instead of using ServiceLoader
[17:04:36] <kcbabo> dbevenius:  and we could add a method to add beans directly to it
[17:04:51] <dbevenius> kcbabo: would be nice to be able to add beans in a clear way...
[17:05:27] <kcbabo> dbevenius:  well, we have that already in a way
[17:05:31] <kcbabo> dbevenius:  for application beans
[17:05:44] <kcbabo> dbevenius:  in this case, the transaction manager bean would be shared across all deployments
[17:06:22] <kcbabo> dbevenius:  instead of requiring people to provide a bean with each application
[17:06:41] <kcbabo> dbevenius:  hmmm ? need to think about it a bit
[17:08:29] <dbevenius> kcbabo: still we should be able to use CDI for this I would have though. So I'll just see why this is not working
[17:08:44] <kcbabo> dbevenius:  cool, thanks dan
[17:08:54] <dbevenius> kcbabo: Lets sync up tomorrow again
[17:10:09] *** dbevenius has left #switchyard
[17:20:38] *** kcbabo has quit IRC
[17:54:49] *** rcernich has joined #switchyard
[17:54:59] *** rcernich has quit IRC
[17:54:59] *** rcernich has joined #switchyard
[18:04:47] <tcunning1> igarashitm: hungry?
[18:10:45] <igarashitm> tcunning1: yep, let's go :)
[18:11:10] <tcunning1> igarashitm: be right down
[18:24:01] *** lanceball is now known as lance|afk
[18:53:02] <tcunning1> antollinim_lunch: ping
[18:53:37] <antollinim_lunch> tcunning: mi nick alwyas gets back to _lunch!
[18:53:40] <antollinim_lunch> tcunning; hi
[18:53:51] <tcunning1> antollinim_lunch: how's it going?
[18:54:10] <antollinim_lunch> tcunning: me good, forge so so
[18:54:22] <antollinim_lunch> tcunning: you?
[18:54:22] <tcunning1> antollinim_lunch: i think i found an easy workaround
[18:54:48] <antollinim_lunch> tcunning: good, what is it?
[18:55:09] <tcunning1> antollinim_lunch: i think for the other plugins (soap/bean/etc) that we can just copy them into the switchyard-plugin module and then add resource-root entries
[18:55:52] <antollinim_lunch> tcunning: that is the exact problem I am trying to solve
[18:56:07] <antollinim_lunch> tcunning: I am working with camel now
[18:56:25] <antollinim_lunch> tcunning: what happens is that when I do "camel-service create"
[18:57:08] <antollinim_lunch> tcunning: it fails cause any of the classloaders can find one resource located inside of switchyard-forge-camel-plugin.jar
[18:57:48] <antollinim_lunch> tcunning: I have tried many things but I am starting to thnik that the Resource is actually loaded, but Switchayrd is not finding it properly
[18:58:32] <tcunning1> antollinim_lunch: i think we just create a lot of problems for ourselves with the separate modules, it might be better now to just combine them
[18:58:34] <antollinim_lunch> tcunning: the rouserce is java/RouteInterfaceTemplate.java
[18:59:04] <antollinim_lunch> tcunning: what I am trying to say is that I am still facing this problem using the strategy you mention
[18:59:13] <tcunning1> antollinim_lunch: ah
[18:59:18] <antollinim_lunch> tcunning: I installed ONLY camel
[18:59:22] <tcunning1> antollinim_lunch: let me give it a try
[18:59:29] <antollinim_lunch> tcunning: and then I added switchard to it
[18:59:50] <antollinim_lunch> tcunning: everyhting loads OK but it fails during runtime of the execution of that command
[19:00:03] <tcunning1> gotcha
[19:00:20] <tcunning1> let me set it up and try
[19:00:26] <antollinim_lunch> tcunning: so, in a few words: we already are able to compile, install and boot plugins
[19:01:11] <antollinim_lunch> tcunning: but there is a very specific problem regarding loading a resource, i have not been able to solve it. This resource is not a class, is a non-compiled java file
[19:02:02] <antollinim_lunch> tcunning: this is the code that does not find the resource: swtrichyard-commong: Classes.java -> getResources(String path, ClassLoader loader) throws IOException
[19:02:34] <antollinim_lunch> tcunning: this class is in charge of loading finding java/RouteInterfaceTemplate.java from switchyard-forge-camel-plugin but it cannot
[19:04:24] <antollinim_lunch> tcunning: to the best of my understanding, forge modeles and dependencies are good, but the file is not loading, so I am starting to think that switchyard-common is the responsible here
[19:05:54] <antollinim_lunch> tcunning: I will actually have lunch now :)
[19:07:51] <tcunning1> antollinim_lunch: :)
[19:12:39] *** kcbabo has joined #switchyard
[19:12:39] *** kcbabo has joined #switchyard
[19:19:53] <rcernich> kcbabo: looks like that build problem resolved itself
[19:20:31] <rcernich> kcbabo: the first build that failed looked like it was having some issues wrt file locks
[19:20:40] <rcernich> kcbabo: wonder if that didn't hork the repo
[19:20:51] <rcernich> kcbabo: anyway, working now
[19:23:42] <kcbabo> rcernich:  guess it was sun spots ;-)
[19:26:04] <igarashitm> tcunning1: ok, I've noticed TransformSwitchYardScanner is only for JavaTransform and others should be created by factory. TransformerUtil#newTransformers() shows it
[19:26:26] <igarashitm> tcunning1: anyway, thanks
[19:26:36] <tcunning1> igarashitm: sorry i wasn't much more help there
[19:32:09] <igarashitm> tcunning1: np, I think you'll help me next time... I'll stuck with another prob :)
[19:33:13] <igarashitm> I'm lucky this was easy prob
[19:53:38] *** echelog-2 has joined #switchyard
[19:57:25] *** rcernich is now known as rcernich_lunch
[20:33:50] *** lance|afk is now known as lanceball
[20:51:24] *** tfennelly has quit IRC
[21:51:20] <tcunning1> antollinim_lunch: i ran into a forge bug with forge-snapshot and camel
[21:51:29] <tcunning1> antollinim_lunch: bpm and drools seem to work pretty easily though
[21:51:52] <antollinim_lunch> tcunning: good! and bad! :)
[21:51:56] <antollinim_lunch> tcunning
[21:52:11] <antollinim_lunch> tcunning: what is the bug? is the resource not found thing?
[21:52:34] <tcunning1> antollinim_lunch: for some reason forge thinks the com.sun.xml.bind:main module is missing
[21:52:39] <tcunning1> even when it's there
[21:55:29] <antollinim_lunch> tcunning: I hit it and solved it
[21:56:11] <antollinim_lunch> tcunning: the camel pom needs to look like this
[21:56:12] <antollinim_lunch> <dependencies>     <dependency>       <groupId>org.switchyard.components</groupId>       <artifactId>switchyard-component-camel</artifactId>     </dependency>     <dependency>         <groupId>org.jboss.forge</groupId>         <artifactId>forge-shell-api</artifactId>         <version>1.0.0.Beta3</version>         <scope>provided</scope>     </dependency>         <dependency>         <groupId>javax.enterprise</groupId>         <artifactId>cdi
[21:56:30] <antollinim_lunch> tcunning: and the module.xml like this
[21:56:32] <antollinim_lunch> <dependencies>       <module name="javax.api"/>       <module name="javax.enterprise.cdi-api" services="import" optional="false">         <imports>             <include path="**" />             <include path="META-INF" />          </imports>       </module>           <module name="org.jboss.forge.shell-api" services="import" optional="false">         <imports>             <include path="**" />             <include path="META-INF" />
[21:57:28] <tcunning1> antollinim_lunch: nice!
[21:59:13] <antollinim_lunch> tcunning: it should install with that
[21:59:52] <antollinim_lunch> tcunnning; then, when you execute "camel-service create"
[22:00:02] <antollinim_lunch> tcunning: it will fail
[22:05:04] *** bfitzpat1 has joined #switchyard
[22:06:10] *** bfitzpat has quit IRC
[22:11:53] *** antollinim_lunch is now known as antollinim
[22:24:27] *** rcernich_lunch is now known as rcernich
[22:36:15] *** kcbabo has quit IRC
[22:48:58] *** tcunning1 has quit IRC
[23:23:05] <antollinim> anyone: does anyone know if a ClassLoader instance is able to "see" class files only or it also is able to see other files contained in a jar, like a .java ?
[23:23:54] <errantepiphany> antollinim: A ClassLoader can see anything that's in it's path.  If you bundled up jpgs, txt files, properties files, source java files, then yes.  If you didn't, then no.
[23:24:07] <errantepiphany> antollinim: You would use getResource(...) from the ClassLoader, however...............
[23:24:39] <errantepiphany> antollinim: ...........in SwitchYard, please use the org.switchyard.common.type.Classes.getResource(...)  methods.
[23:25:07] <antollinim> errantepiphany: I am not modifying, I am just troubleshooting something that is not workin
[23:25:33] <errantepiphany> antollinim: ah
[23:25:37] <antollinim> errantepiphany: obviously, right? why would I troubleshoot sth that works ? :)
[23:25:57] <antollinim> errantepiphany: forge is loading switchyard camel plugin
[23:26:24] <errantepiphany> antollinim: Yeah.... Um.... I'm not a "forge" expert by any stretch.
[23:26:32] <antollinim> errantepiphany: when I run a sw camel plugin within forge, the command fails cause a resource is not being found. The resource is a template java file
[23:26:57] <errantepiphany> antollinim: well, I doubt the xyz-sources.jar file is on the classpath. ;)
[23:28:11] <antollinim> errantepiphany: ok then, so I have to focus on the the classpath then. I have done that but maybe I must check thoroughly again
[23:28:21] <antollinim> errantepiphany: thanks for your help
[23:28:58] <errantepiphany> antollinim: there's a maven command that shows your dependency heirarchy... I would see what that comnmand is and try that.  you can also see what m2eclipse brings in right in eclipse under maven dependencies.
[23:29:21] <errantepiphany> antollinim: sorry I don't remember the command
[23:30:10] <antollinim> errantepiphany: don't worry I use eclipse  and I have also used tattletale, good tool
[23:30:44] <antollinim> errantepiphany: CU
[23:31:37] <errantepiphany> antollinim: laters.
[23:33:03] *** igarashitm has quit IRC
[23:39:04] *** antollinim has quit IRC
[23:41:17] *** vinicius has quit IRC
[23:52:27] *** ldimaggi has quit IRC
[23:57:40] *** babo has joined #switchyard





top