13:01:12 <sochotni> #startmeeting #startmeeting Java SIG --
13:01:12 <zodbot> Meeting started Wed Mar 16 13:01:12 2011 UTC.  The chair is sochotni. Information about MeetBot at
13:01:12 <zodbot> Useful Commands: #action #agreed #halp #info #idea #link #topic.
13:01:18 <sochotni> #meetingname java-sig
13:01:18 <zodbot> The meeting name has been set to 'java-sig'
13:01:24 <sochotni> #topic roll-call
13:01:32 <mbooth> Good afternoon
13:01:36 <akurtakov> here :)
13:01:49 <nthykier> here
13:02:26 <pingou> .fas pingou
13:02:27 <zodbot> pingou: pingou 'Pierre-YvesChibon' <>
13:02:40 <sochotni> #info mbooth akurtakov nthykier pingou sochotni attending
13:02:57 <sochotni> maybe more people will turn up during the meeting
13:03:34 <sochotni> #topic guidelines changes
13:04:33 <sochotni>
13:04:42 <sochotni> that is where changes have been happening
13:05:08 <sochotni> #link is difference between current guidelines and draft
13:05:36 <sochotni> not many changes really
13:05:56 <akurtakov> so it's use jpackage_script usage and maven 3.x usage recommendation
13:05:59 <sochotni> just maven2/3 distinction, jpackage_script addition
13:06:18 <sochotni> yes basically just that
13:06:38 <sochotni> I couldn't make "clean-binary-files" to work easily
13:06:47 <akurtakov> does anyone has any objections?
13:07:00 <mbooth> Looks good to me :-)
13:07:31 <pingou> stupid question but what works on maven3 will work on maven2 right ?
13:07:46 <sochotni> I would like to see a few additions to jpackage-utils (such as simplified macro for cleaning jars/class files)
13:07:48 <akurtakov> pingou: nope
13:08:01 <akurtakov> it might work but it's not mandatory
13:08:03 <pingou> ok akurtakov so no need for a separate maven2/maven3 :)
13:08:08 <sochotni> weeell...answer is "should" :-)
13:08:12 <mbooth> pingou: I think the other way round. What works on Maven2 should work on Maven3
13:08:26 <pingou> s/no// on my last sentence
13:09:04 <akurtakov> pingou: we have quite a lot of packages depending on maven2 but we want to discourage its usage
13:09:08 <sochotni> pingou: not really, maven 3 (maven package) is able to build maven2 projects just fine (with a few exceptions that should be filed as bugs)
13:09:49 <pingou> sochotni: ok that was my question :)
13:10:01 <akurtakov> that's why we want to put guidelines on maven 3 and I would say trying to kill maven 2 for F-17
13:10:19 <akurtakov> reduce usage to the minimum for F-16
13:10:24 <akurtakov> deprecate for F-17
13:10:28 <pingou> I was seeing this from a packager pov of java apps, if I have a maven2 app will it work with these new guideline, and from what I read answer is yes :)
13:10:37 <akurtakov> yes it will
13:10:58 <sochotni> pingou: yes, I found just one app that didn't work so far and I have yet to figure out why exactly
13:11:05 <sochotni> it might have been my refactoring
13:11:36 <akurtakov> pingou: note that upstream no longer care for maven 2 so we are on our own
13:12:46 <sochotni> now there is another guideline thing that is in the making...
13:12:48 <sochotni> JNI changes
13:13:23 <sochotni> #link 64/32 bit jvm and JNI
13:14:08 <sochotni> the bug contains discussion on the topic to a big degree and I believe whatever solution comes out of that bug should be considered as a JNI guideline
13:14:29 <sochotni> it's not finished yet and I have a feeling it will take some time till the dust settles
13:17:01 <sochotni> so the question there was..
13:17:25 <sochotni> do we skip the JNI for now and just focus on changes mentioned earlier?
13:18:38 <mbooth> Unless a concrete JNI proposal is going to emerge Soon(tm)...
13:18:41 <sochotni> another change I'd like to see is inclusion of %add_maven_depmap in guidelines but that won't happen before it's in jpackage-utils
13:19:03 <akurtakov> let's skip jni for now
13:19:20 <akurtakov> I would vote +1 for the first jni guideline
13:19:26 <akurtakov> proposal
13:19:33 <akurtakov> any guideline is better than none
13:20:10 <sochotni> ok, I changed the mvn template line to:
13:20:10 <sochotni> mvn-rpmbuild install javadoc:javadoc # javadoc:aggregate for multi-module projects
13:20:19 <pingou> :)
13:20:33 <pingou> +use ?
13:20:56 <sochotni> there :_)
13:20:59 <akurtakov> sochotni: we might simplify by suggesting aggregate always
13:21:11 <akurtakov> it works even for non multimodule
13:21:19 <sochotni> akurtakov: it does?
13:21:31 <sochotni> I remember it didn't
13:21:32 <akurtakov> I'm pretty sure about this
13:21:41 <akurtakov> let's check it after the meeting
13:21:51 <pingou> (if so, then we can just mention javadoc:javadoc below as a possible alternative in case of problems)
13:22:04 <akurtakov> such change could go in with the new macro
13:22:10 <akurtakov> so are we skipping jni now?
13:22:51 <sochotni> akurtakov: ok, aggregate works for single module..
13:23:00 <pingou> by skipping you mean submit the new draft w/o including the results of the discussion on the bug mentionned earlier, right ?
13:23:44 <sochotni> pingou: I believe yes, we'd like to submit these changes ASAP and not wait for JNI resolution in that bug
13:24:04 <pingou> +1 for me
13:24:12 <mbooth> When does the FPC meet next?
13:24:28 <sochotni> it might be tomorrow/thursday
13:24:38 <sochotni> or later today even?
13:24:50 <sochotni> eh, tomorrow is thursday :-)
13:25:07 <mbooth> Oh, might as well get your maven3 related updates approved straight away then :-)
13:25:09 <pingou> every Wednesday at 16:00 UTC
13:25:13 <sochotni> FYI I changed the draft to have javadoc:aggregate
13:26:15 <akurtakov> so formally +1 for the new guidelines changes and +1 for skipping jni for now
13:26:43 <sochotni> obviously +1 from me for both things
13:26:47 <mbooth> Yes
13:28:31 <sochotni> #agreed Submit new draft for FPC approval
13:28:41 <sochotni> #topic Tracker bug status
13:29:04 <sochotni> it looks much better than last time I'd say
13:29:11 <sochotni> 10 bugs
13:29:42 <akurtakov> there are few long standing reviews still
13:29:58 <akurtakov> and idea/jps build failure
13:30:02 <sochotni> yes also a few non-review bugs that have been around for some time
13:30:44 <mbooth> I spoke to upstream about the java-gnome FTBFS, by the way
13:30:57 <akurtakov> libnotify problem?
13:31:11 <mbooth> They should should have a new release out before F15 is release :-)
13:31:27 <mbooth> Yeah, libnotify had API changes
13:31:56 <sochotni> that's good
13:32:19 <sochotni> (that they will have release before f15, not that there were API changes :-) )
13:32:34 <hicham> the fix should be trivial, I wonder why no one did it
13:32:56 <akurtakov> because noone cared enough ?
13:33:44 <hicham> it is just just an extra parameter that should be removed from some call
13:34:07 <akurtakov> same as the idea case - there is a workaround in the FTBFS bug but it is not applied and package is not rebuild
13:34:26 <mbooth> The bindings are programmatically generated... If it was that trivial, it should not broken?
13:35:09 <hicham> I believe libnotify binding are not autogenerated
13:35:44 <mbooth> I see
13:37:19 <sochotni> I floor...
13:37:24 <sochotni> #topic open floor
13:37:45 <mbooth> Helios SR2 was released, I beleive...
13:37:58 <mbooth> Should we build updates only for Rawhide?
13:38:08 <mbooth> Or do you want SR2 in F15?
13:38:21 <akurtakov> I definetely plan to push it to F-15
13:38:25 <mbooth> Cool
13:38:32 <sochotni> mbooth: not related but for now I add most of my packages to F15 still
13:38:35 <akurtakov> but there is a problem with the browser integration
13:38:50 <akurtakov> once it's fixed I'll do a F-15 build
13:39:12 <mbooth> sochotni: Me too, but eclipse is it's own little eco-system :-)
13:40:10 <akurtakov> well, we tend to push SR releases to latest version
13:40:45 <akurtakov> new releases only in development phase of a distro version
13:40:53 <akurtakov> I mean new feature releases
13:41:03 <mbooth> Ok
13:41:15 <akurtakov> mbooth: btw, I would really appreciate your help with pushing SR2 builds into rawhide
13:41:32 <akurtakov> we can backport easily after that
13:41:50 <nthykier> akurtakov: are there any complications with SR2?
13:42:07 <nthykier> other than the browser integration
13:42:08 <akurtakov> nthykier: except for the browsers no
13:42:12 <mbooth> Sure, I'll see what I can do
13:42:17 <nthykier> ok
13:43:52 <sochotni> ok, so that's it I guess?
13:44:06 <akurtakov> we skipped tomcat
13:44:16 <sochotni> I didn't plan it
13:44:23 <akurtakov> but noone do anything so yeah
13:44:30 <sochotni> yup
13:44:38 <sochotni> it's not a priority right now
13:44:50 <sochotni> too late for f15 and plenty of time till f16
13:44:58 <sochotni> #endmeeting