Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-05-18

| Channels | #bioclipse index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
06:34 egonw joined #bioclipse
07:00 edrin joined #bioclipse
07:00 edrin hi again
07:01 edrin cu later
07:01 olass joined #bioclipse
07:01 edrin oh
07:01 edrin hi ola
07:01 egonw moin
07:01 zarah privet egonw
07:02 edrin olass, egonw: got my mail?
07:02 Gpox joined #bioclipse
07:02 olass edrin: yes
07:03 edrin i dont know what is wrong
07:04 edrin interestingly this error is only in the build, and it is working well from eclipse
07:06 olass edrin: what jar seems to be missing?
07:06 olass on the classpath?
07:06 masak joined #bioclipse
07:08 edrin olass: i think it's not a jar that is missing
07:08 olass edrin: so what do you think?
07:08 olass you are completely bypassing all managers with your code
07:09 olass right?
07:09 edrin currently i think that this is some strange bug that is associated with either eclipse or spring but only in build environment
07:09 olass only in exported environment you mean?
07:09 edrin yes
07:09 edrin the error is not associated with the manager code itself
07:10 olass why do you think spring is involved?
07:10 masak oh, are you discussing the mysterious export/build failure?
07:10 edrin the problem seems to be when the newly created object is returned to the called from the manager
07:10 egonw masak: related, likely
07:10 olass masak: not really
07:10 edrin masak: no, build works for me
07:10 egonw but indeed not the same thing
07:10 olass edrin: not for me
07:10 egonw edrin has been working from a pre NWO branch in SVN
07:11 edrin masak: i will forward you the error mail, what is your mail address?
07:11 masak jonalv++ recounted how things broke unexpectedly on export on Friday.
07:11 edrin egonw: no
07:11 olass edrin: please explain this turther > [09:13] < edrin> the problem seems to be when the newly created object is returned to the called from the manager
07:11 egonw edrin: no?
07:11 masak edrin: carl.masak@farmbio.uu.se
07:11 egonw edrin: back with master, are you?
07:11 edrin egonw: i got the current trunk to export here. then i still have this error
07:11 egonw ic
07:12 olass edrin: it really does not help that trunk exports just with you, it must work on all platforms
07:12 olass hence we cannot test exporting
07:12 olass and cannot reproduce your error
07:12 olass and cannot release
07:12 olass so we need to solve that problem first
07:13 olass after that, we should solve the classpath problem
07:13 olass but it seems very complicated
07:13 edrin olass: yes, sure
07:13 olass edrin: what eclispe version are you using?
07:13 olass 3.4.2?
07:13 edrin the newest on
07:13 edrin yes
07:13 edrin i downloaed it from web 3 days ago
07:13 olass edrin: 3.5.0.M7 is the latest on eclipse
07:14 olass 3.4.2 is the latest stable
07:14 edrin and installed alle svn, emf... plugins again
07:14 olass so please be specific
07:14 edrin Version: 3.4.2
07:14 edrin Build id: M20090211-1700
07:14 edrin this one
07:14 olass good
07:15 olass but when you exported you got another message than we in bug 1070
07:15 edrin yes
07:15 olass edrin: could you enter the error you got in teh bug?
07:15 edrin some circular dependencies
07:15 olass and how you resolved it?
07:16 edrin @pz 1070
07:16 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1070
07:16 olass edrin: circular dependencies should be caught on compile time...
07:17 olass strange that you do not get problem "Package uses conflict"
07:18 rojasm joined #bioclipse
07:21 edrin olass, masak: is jonalv modifying or trying to intercept the objects returned from a manager function?
07:21 olass edrin: are you in any way using a manager for executing the Jseditor content?
07:21 masak edrin: I think so. at least in some cases.
07:22 edrin masak: can you mail me a description where and what and why this is done?
07:23 edrin olass: no, the jseditor content is executed in the new bc.jsexecution plugin
07:23 masak edrin: I think you'd better talk to jonalv. I'll tell him you asked. this is all new to me too.
07:23 olass masak: I agree
07:23 olass edrin: ask jonalv
07:24 olass @tell jonalv have you seen this? http://olegz.wordpress.com/doc​uments-and-articles/osgi-uses/
07:24 zarah olass's link is also http://tinyurl.com/o6wypg
07:24 zarah Consider it noted.
07:24 edrin masak: interestingly this error does not happen in the manager code itself, but only if the manager passes code back to the javascript scope
07:24 edrin ok, must go now
07:24 edrin cu
07:24 rojasm moin all
07:24 zarah saluton rojasm
07:25 olass hi rojasm
07:25 masak hi rojasm
07:33 rojasm I made an update with Bioclipse2 and I get this error (for both my computers):
07:33 rojasm java.lang.ClassNotFoundException: org.eclipse.core.runtime.adaptor.EclipseStarter
07:33 rojasm am I missing some new plugin?
07:33 egonw rojasm: did you try reset target platform?
07:33 rojasm yes
07:33 rojasm and I removed the workspace
07:33 rojasm but I doesn't work
07:33 rojasm mmm
07:35 rojasm I try again
07:43 rojasm Oh good. After cleaning, refreshing 10000 times it works. :(
07:46 masak that's a lot of times.
07:55 edrin joined #bioclipse
07:56 jonalv joined #bioclipse
07:57 jonalv oh hai
07:57 zarah hi jonalv
07:57 zarah jonalv: You have new messages. Write '/msg zarah @messages' to read them.
07:58 jonalv zarah: just tell me...
07:58 zarah jonalv: 33 m 49 s ago, olass said have you seen this? http://olegz.wordpress.com/doc​uments-and-articles/osgi-uses/
07:58 olass jonalv: is it really uses conflict we have problems with?
07:58 jonalv olass: I hvae problems with everything...
07:58 * olass has installed Eclipse 3.5.0.M7 and tries it
07:59 olass edrin claims he managed to export
07:59 olass @pz 1070
07:59 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1070
07:59 jonalv olass: so problem solved then?
07:59 olass oh no
07:59 olass It does not work for me
08:00 jonalv olass: no?
08:00 olass please try it
08:00 jonalv hm...
08:02 olass jonalv: edrin never got the "package uses conflict"
08:02 jonalv olass: did he commit that?
08:02 olass don't think so
08:03 * jonalv is downloading latest changes...
08:04 masak karma bioclipse
08:04 zarah bioclipse has a karma of 3
08:04 masak bioclipse++
08:06 rojasm egonw: what test do I need to run in medea.business.test?
08:06 rojasm AbstractMedeaManagerPluginTest?
08:06 egonw AllMedeaManagerTest
08:06 egonw and
08:07 egonw AllMedeaManagerPluginTest
08:07 egonw the first as normal unit test
08:07 egonw the latter as Plugin test
08:07 egonw see this wiki page:
08:07 rojasm ok
08:07 egonw http://wiki.bioclipse.net/index.php?title=How_to_m​ake_a_manager_2#Setting_up_the_testing_frame_work
08:07 zarah egonw's link is also http://tinyurl.com/ouun2r
08:10 rojasm I have errors in java.util.zip.ZipException: error in opening zip file
08:10 rojasm mmm
08:16 shk3 joined #bioclipse
08:20 CIA-40 bioclipse: jonalv * r10502 /bioclipse2/trunk/plugins/net.bio​clipse.core/META-INF/MANIFEST.MF: removed Spring requirements from core. They should have been removed earlier. However this did not fix export problems.
08:20 jonalv edrin: around?
08:21 masak jonalv++ # :/
08:31 jonalv OSX-- //adds .DS_Store files everywhere
08:34 jonalv egonw: can you help me with git sometime?
08:34 egonw sure
08:35 jonalv I have a branch for the new Spring version
08:35 jonalv I am trying to rebase it with the latest changes in master
08:35 jonalv there is and evil .DS_Store(OSX--) file which it complains about
08:36 jonalv it seems I have added that file in the branch
08:36 jonalv I removed it now
08:36 jonalv so it should not be tracked
08:36 jonalv but when I do rebase I get:
08:37 jonalv uhm hm now I get something else
08:37 jonalv egonw: thanlyou
08:37 jonalv |thank you
08:39 egonw share desktop?
08:39 egonw and skype for talk?
08:39 jonalv egonw: the problem magicly solved itself
08:39 egonw git++
08:40 jonalv well it finally did what I expected it to do all the time
08:40 jonalv no idea...
08:48 edrin jonalv: I am here now
08:49 jonalv edrin: could you test exporting what is in trunk now? It doesn ot work for me...
08:49 edrin jonalv: i dont remember for sure if the error i had was "package uses conflict"
08:49 jonalv edrin: bu you managed to export without errors once?
08:49 edrin jonalv: yes, it work for me when i removed some plugins from the core manifest
08:49 edrin jonalv: yes, several times, at least for windows
08:49 jonalv edrin: did you commit those changes?
08:49 edrin no, did not want to modify stuff you did
08:50 edrin @pz 1070
08:50 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1070
08:50 edrin jonalv: see last post here http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1070
08:50 jonalv edrin: I have removed a lot of deps in trunk now. Could you see if you can export after my changes?
08:50 zarah edrin's link is also http://tinyurl.com/okb2fx
08:50 jonalv edrin: last post is mine...
08:51 jonalv :)
08:51 edrin jonalv: I am not at home now, i can't do bioclipse stuff from here
08:51 jonalv oh :(
08:51 jonalv edrin: That is very irritating...
08:51 jonalv :)
08:52 edrin jonalv: i have to mention, before i did the export stuff i had to update my eclipse, since some plugins claimed they need 3.4.2
08:52 * jonalv is running 3.4.2
08:52 jonalv hm could there be a later 3.4.2 fix??
08:53 jonalv olass: I don't suppose I could ask you to try exportign from a new clean Eclipse? :)
08:53 edrin since the update within bc did not work i reinstalled whole eclipse and checked out all EMF, svn, plugins for my eclipse again, i then switched to my old workspace where all bc plugins are, deleted the run file for bc and started with setting target platform...
08:53 * jonalv is tryiing to get his branch with the latest Spring version to start...
08:53 olass jonalv: does it work for you?
08:54 jonalv olass: I have not managed to export no
08:55 edrin jonalv: did you get my mail, can you point me to where the manager calls are intercepted for example? do you modify the classpaths of the manager calls?? or any thread from with the bc core...?
08:55 olass jonalv: I had no success with building Bioclispe with 2.5.0RC1
08:55 olass 3.5.0.RC1
08:55 jonalv olass: I see
08:55 olass did not compile
08:55 jonalv edrin: did you get my mail where I answer your mail? :)
08:55 olass the problems were these plugins: managers and recording ;)
08:56 olass (they are the new ones, right?)
08:56 edrin jonalv: oh, no, i have wron mail account here.. i'll check
08:56 jonalv olass: yes there is something fishy with those two...
08:56 olass could that be the problem here?
08:56 jonalv olass: I have no idea
08:56 jonalv bu yea...
08:57 olass let me know if you compe up with something for me to test
08:57 olass come
08:57 olass I guess updating spring is good too...
08:57 jonalv olass: I have a branch for the Spring update project
08:57 olass ok
08:57 jonalv olass: I could share it on pele if you wanna terst it...
08:57 olass I am not on git
08:58 jonalv oh :(
08:58 olass (as you very well know)
08:58 edrin ah, the recording thing ?
08:58 olass no, git is for version control
08:58 masak olass: use git!
08:58 masak seriously.
08:59 edrin manage declaration in the Spring config (xml) file <- what is this? isn't it the interface of my manager i must modify ??
09:00 jonalv edrin: huh?
09:00 jonalv edrin: oh removing the @Recording annotation might suffice
09:00 edrin jonalv: i did not use IBioObject anywhere
09:00 jonalv but you can remove the entire recording code
09:00 edrin jonalv: ok, and the xml file? where can i find and remove this?
09:01 jonalv edrin: actually I think you should try removing that advice
09:01 jonalv edrin: http://wiki.bioclipse.net/index.​php?title=How_to_make_a_manager
09:01 zarah jonalv's link is also http://tinyurl.com/dypvah
09:01 jonalv edrin: http://wiki.bioclipse.net/index.php?title=H​ow_to_make_a_manager#The_recording_advisor
09:01 zarah jonalv's link is also http://tinyurl.com/dypvah
09:02 CIA-40 bioclipse: ospjuth * r10503 /bioclipse2/trunk/plugins/net.bioclipse.qsar/ (3 files in 3 dirs): Added an extension point for response values and populated it with 4 sample response units in plugin.xml; IC50, EC50, LD50, and Kd.
09:05 egonw science outreach program: http://xkcd.com/585/
09:05 jonalv gaaaah
09:06 * jonalv hates this start order and dependency shit
09:08 edrin hehe
09:09 edrin (at egon)
09:12 * olass wants coffee
09:12 * olass gets coffee
09:15 * olass is back wit ha freshly brewed la bomba
09:17 edrin jonalv: removing recording did not help
09:17 jonalv edrin: okey
09:17 edrin jonalv: where do you intercept these manager calls and manipulate them?
09:17 jonalv edrin: which world order? :)
09:17 edrin jonalv: can yuo point me to the code where this is done`?
09:18 edrin jonalv: the current trunk
09:18 edrin jonalv: could you mail me this info? i must leave now, already
09:18 edrin latze here
09:18 jonalv edrin: current trunk contains three different generations. Which plguin?
09:19 jonalv olass: do you know which plguin edrin is referring to?
09:19 olass no
09:19 olass I really have no clue what you two are discussing
09:19 olass I don't know java execution
09:19 jonalv olass: okey, fair enough
09:19 olass I don't know spring
09:20 olass I do know some eclipse
09:20 * jonalv is moving around huge code blocks again...
09:20 jonalv it's scary...
09:29 jonalv gaaaaaaa
09:29 * jonalv is stuck in dependecy swamp
09:29 jonalv gaaaa, yet another circular dep
09:30 egonw drawings...
09:30 egonw making drawings
09:30 egonw of how it should be
09:30 jonalv drawings?
09:30 egonw yes, of the deps
09:30 egonw what it should look like
09:30 jonalv oh
09:30 jonalv it's too complex...
09:30 egonw what you feel it should look like
09:31 egonw making a drawing makes it easier to handle a complex structure
09:31 * jonalv is moving BioList to net.bioclipse.recording now...
09:32 jonalv that actually makes sense
09:34 egonw does it?
09:34 egonw doesn't that make recording non-optional?
09:35 egonw is that intentional?
09:35 egonw (fine with that, though, ...)
09:35 masak recording has always been non-optional in some sense...
09:35 jonalv egonw: BioList is only needed when you are doing recording
09:35 jonalv it actually makes it more optional I think
09:36 olass jonalv: why is org.springframework.bundle.osgi.extender not listed as dep in net.bioclipse.ui, when it is referenced in the activator?
09:36 masak jonalv: I disagree. BioList is everywhere, in a lot of managers.
09:36 jonalv olass: I am working on that in my branch
09:36 egonw masak: I'm not worried about impl details... so, 'in a sense' is not relevent...
09:36 jonalv masak: that\s because Recoring is everywhere...
09:37 jonalv masak: |Recording
09:37 masak egonw: then I don't really understand your question.
09:37 egonw masak: there is a difference between what one wants...
09:37 egonw and how it is implemented
09:37 masak jonalv: so, in effect, every manager that has 'BioList' in it will import from recording now?
09:38 jonalv masak: yes
09:38 masak egonw: yes
09:38 egonw which makes it non-optional
09:38 masak yes.
09:38 jonalv egonw: no
09:38 masak but it was already, IIUC.
09:38 jonalv egonw: why?
09:38 masak jonalv: you just said it was.
09:38 masak jonalv: the whole code base is steeped in recording.
09:38 egonw jonalv: question: why does it mean that using BioList means I want recording?
09:39 jonalv egonw: becasue that's what BioList is buil;t for
09:39 jonalv egonw: it's like asking why does using a lawnmover means I wanna move lawn..
09:40 jonalv BioList is a class for getting a recordable List. Nothing more nothing less
09:40 masak the mere fact that we're referring to BioList everywhere makes us dependent on recording. whether or not BioList is in fact in the recording package.
09:40 jonalv it's an ugly hack
09:40 masak and it's everywhere. :/
09:41 jonalv masak: but that's just becasue we want recording everywhere...
09:41 masak jonalv: Java sucks.
09:41 masak either that, or our understanding of Java sucks.
09:41 masak I wouldn't rule out the latter.
09:41 * jonalv is stuck in yet another circualr dep
09:42 jonalv bah fuck recording
09:42 jonalv I am sick of this
09:42 masak jonalv++
09:43 masak let's clean out BioList while we're at it.
09:43 masak we can always re-implement recording for Bioclipse 3.0
09:43 CIA-40 bioclipse: ospjuth * r10504 /bioclipse2/trunk/plugins/net.bioclipse.qsar/ (9 files in 5 dirs):
09:43 CIA-40 bioclipse: Renamed EP to responeUnit, implemented a ResponseUnit class, added ResponseUnit
09:43 CIA-40 bioclipse: to DescriptorModel, refactored EP reading from QSARManager to QSARHelper, added
09:43 CIA-40 bioclipse: reading of reponseUnits from EP and made accessor methods from QsarManager.
09:44 CIA-40 bioclipse: ospjuth * r10505 /bioclipse2/trunk/plugins/net.bioclipse.qsar.test/ (2 files in 2 dirs): Added test for getting ResponseUnits from QsarManager.
09:44 egonw jonalv: ah... ic... was misguides by the name BioList
09:44 egonw the lack of JavaDoc does not help either
09:44 jonalv egonw: k
09:45 egonw maybe rename to RecordableList ?
09:45 jonalv egonw: do you want your knife back or can I keep it?
09:45 masak we're a bit lousy at documentation.
09:45 egonw jonalv: no, you can keep it...
09:45 jonalv egonw: yes REcordableList. That's perfect. Can you file me a bug rapport on the topic?
09:45 egonw or give it masak
09:45 masak I have enough knives already.
09:45 egonw for when I make a amusing english spelling error :)
09:45 masak egonw: I'm off duty right now... :)
09:46 egonw jonalv: will do
09:46 egonw yes, moving BioList to recording now makes perfect sense :)
09:46 masak so, we officially have a knife now? in #perl6 we have a pumpkin...
09:47 CIA-40 bioclipse: ospjuth * r10506 /bioclipse2/trunk/plugins/net.biocl​ipse.qsar.ui/src/net/bioclipse/qsar​/ui/editors/InformationPage.java: Commented out references since it does not work.
09:47 egonw jonalv: @pz 1075
09:48 jonalv egonw: tht doesn't work...
09:48 jonalv egonw: you should talk to masak about that syntax...
09:48 egonw was sor t
09:48 egonw was sort of to make it easy to copy/paste
09:49 jonalv @pz 1075
09:49 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1075
09:49 masak egonw: jonalv doesn't implement the @pz command yet.
09:49 masak patches welcome.
09:49 egonw ack, I'll use the XMPP blast service for that
09:50 masak jonalv: you'd better watch out, that sounds dangerous. :)
09:50 * jonalv hides his source code
09:50 egonw please do
09:51 * egonw does not like naked man running in the corridor
09:51 masak hate that guy.
09:52 masak he could at least wear socks.
09:52 * jonalv is trying to work here... :)
09:52 egonw masak: one would be enough
09:52 masak egonw: even a glove would do.
09:53 egonw if a yellow one, and not a transparent latex one
09:53 masak nod.
09:59 rednael joined #bioclipse
09:59 egonw rednael: welcome back!
10:09 jonalv mad laughter!
10:10 masak quiet incomprehension!
10:10 egonw mild puzzledment!
10:11 jonalv egonw: where did our NewWorldOrderPluginsTests class go?
10:11 * jonalv can't find it...
10:11 egonw merged with AllPluginTests
10:11 egonw since everything is NWO now
10:11 egonw or about
10:12 egonw was that a mistake?
10:12 * egonw did not think so at the time...
10:12 jonalv egonw: okey where is that then?
10:12 * jonalv needs something to run so he can see how bad things are in his branch...
10:12 egonw some package
10:12 egonw AllStableManagerPluginTests
10:13 jonalv egonw: oki
10:13 jonalv egonw: thanks
10:13 egonw which reminds me...
10:13 egonw I'll set up the AllUnstableManagerTests versions too :)
10:14 jonalv egonw++
10:16 jonalv egonw: is AllBioclipsePluginsTests up to date?
10:16 CIA-40 bioclipse: egonw * r10507 /bioclipse2/trunk/plugins/net.biocl​ipse.tests/src/net/bioclipse/tests/ (3 files): Added suites for testing of unstable managers
10:17 egonw at least for the work I did yesterday
10:20 jonalv egonw: good so it's probably that one I should run?
10:21 egonw oh, wait
10:21 egonw no, those are not up to date
10:21 egonw sorry...
10:21 egonw AllStable*Tests are
10:23 egonw jonalv: ping
10:23 jonalv egonw: yea?
10:23 egonw jonalv: if there are more than one contextFOO.xml files
10:23 egonw they are both read and used?
10:23 jonalv egonw: yes
10:23 egonw good
10:23 egonw then I am proposing this naming convention:
10:23 egonw context-MANAGERNAMESAPCE.xml
10:24 egonw e.g.
10:24 egonw context-cdk.xml
10:24 egonw and for jchempaint,which comes with two managers:
10:24 egonw context-jcp.xml
10:24 egonw context-jcpglobal.xml
10:24 jonalv egonw: okey, I have slighlty larger problems than naming schemes for xml files...
10:25 egonw :)
10:25 egonw sure
10:25 jonalv egonw: maybe you want to help me?
10:25 egonw yes
10:29 jonalv egonw: I will share my branch on pele after lunch then...
10:29 edrin jonalv: i was talking about the manager of the xws4j plugin
10:29 jonalv edrin: look in the spring config file, the classes are referenced there
10:30 edrin jonalv: where is the spring config file?
10:30 jonalv edrin: you are kidding right?
10:31 egonw edrin: META-INF/spring
10:31 edrin ah
10:31 edrin jonalv: no, not kidding...
10:31 jonalv edrin: could you please read: http://wiki.bioclipse.net/index.php?title=H​ow_to_make_a_manager#The_recording_advisor
10:31 zarah jonalv's link is also http://tinyurl.com/dypvah
10:31 jonalv the whole page...
10:31 edrin jonalv: no clue how these files came in my plugin :)
10:32 jonalv edrin: I would explain but I am a bit busy...
10:32 edrin ok
10:33 edrin i will now try to check if this problem is caused by the manager thing... will try to directly call xwsapi bypassing managers
10:37 jonalv olass: okey I can get Bioclispe to start with the new Spring version if I add required plugins. How do I get it to start from feature?
10:38 maclean joined #bioclipse
10:41 edrin jonalv: that sounds good!
10:42 egonw jonalv++
10:42 egonw for having seen the end of the tunnel
10:42 CIA-40 bioclipse: egonw * r10508 /bioclipse2/trunk/plugins/ne​t.bioclipse.cdk.jchempaint/ (11 files in 5 dirs):
10:42 CIA-40 bioclipse: Converted the two JChemPaint managers to the NWO:
10:42 CIA-40 bioclipse: * added IJava/IJavaScript flavored interfaces
10:42 CIA-40 bioclipse: * updated the activator to allow getting of both
10:42 CIA-40 bioclipse: * rewritten Spring context configuration and split up in one per manager
10:42 jonalv okey now it starts from features as well
10:42 jonalv attempting to export...
10:43 jonalv that did not work.
10:43 edrin interesting now i get the error even from the eclipse run
10:44 jonalv Problems during export
10:44 jonalv Bundle net.bioclipse.ui_0.0.0 failed to resolve.:
10:44 jonalv Package uses conflict: Require-Bundle: org.springframework.osgi.extender; bundle-version="1.2.0"
10:44 * jonalv just want's to give up now
10:47 olass jonalv: go for lunch and let's talk after that
10:47 olass on skype
10:48 * olass is halfways through his lunch
10:59 * olass is back from lunch
11:01 egonw maclean: ping
11:01 egonw am I missing something or are you examples not refering to the picture?
11:01 egonw should the partitionings not have length 5, for 5 carbons?
11:06 maclean egonw: yeah, that's true.
11:07 egonw ok, no worries
11:07 maclean I was giving the partitions for C4H6, I suppose.
11:07 egonw I asked about them to better understand how they are linked
11:24 * jonalv is back
11:25 jonalv olass: did you want to talk?
11:26 olass jonalv: well...
11:26 olass don't think I can contribute much
11:26 olass have you progressed anything?
11:26 olass sounded like it, for a while
11:27 jonalv I have managed to get the same problem using the latest version of Spring
11:27 jonalv + a bunch of other problems...
11:27 jonalv olass: I am gonna read that blog post you linked to earlier now
11:27 olass ok. don't know if it helps
11:28 * egonw makes a mental note about not having tested exporting in the branch before merge...
11:29 olass jonalv: If I do                     List<ICDKMolecule> mollist = cdk.loadMolecules(file);
11:30 olass from anothr thread than the UI thread
11:30 olass is that OK?
11:30 egonw olass: depends on how you constructed 'cdk'
11:30 jonalv olass: yea I guess so
11:31 * olass always uses Activator.getd.getjavacdkmanager()
11:31 egonw olass: it is a Java flavor of the maanger?
11:31 egonw olass++
11:32 egonw OK, done with converting all test suites for the manager he is responsible for...
11:33 egonw expect the Jmol one, which is in intermediate state, with things Jonathan worked on last
11:34 olass egonw, jonalv: How do I call cdk.loadMolecules(file) with my own supplied monitor?
11:34 olass cdk.loadMolecules(file, monitor) does not seem to exist
11:34 jonalv olass: create it then
11:34 olass ?
11:34 jonalv if you need it you cdan create it
11:35 egonw there is one
11:35 egonw file == IFile
11:35 olass should that not be handled with AOP?
11:35 jonalv I have not tested that but it should work
11:35 egonw monitor = IProgressMonitor
11:35 egonw in CDKManager
11:35 olass egonw: it is not in interface AFAIK
11:35 olass not in ICDKManager
11:35 CIA-40 bioclipse: egonw * r10509 /bioclipse2/trunk/plugins/ (14 files in 6 dirs):
11:35 CIA-40 bioclipse: Converted test suite for the two JChemPaint managers to the NWO:
11:35 CIA-40 bioclipse: * created two subfolders
11:35 CIA-40 bioclipse: * added missing suites
11:35 CIA-40 bioclipse: * added API and coverage testing for JChemPaintGlobalPropertiesManager
11:35 CIA-40 bioclipse: * updated test class annotation in managers' interfaces
11:35 egonw olass: now figuring out what the interface equivalent would be
11:35 * jonalv aborts everything else he is doing to look at how cdk.loadMolecuels looks
11:36 olass jonalv: NOOO
11:36 olass export is so much more important
11:36 jonalv olass: then why do you keep asking me these questions?
11:36 egonw jonalv: I'm on it
11:36 olass jonalv: I asked egonw
11:37 egonw olass: and jonalv
11:37 egonw accidents happen
11:37 olass accidents?
11:37 egonw jonalv: we need you on the exporting
11:37 olass yes..
11:37 egonw please do disregard anything else
11:37 egonw as if it never happened
11:38 * egonw wishes he was at BMC to help jonalv
11:38 jonalv egonw: perhaps you could join me when you find some time...
11:38 olass egonw: I do                     List<ICDKMolecule> mollist = cdk.loadMolecules(file);
11:38 olass and it gives me Exception: 1
11:38 olass ArrayIndexOutOfBounds
11:38 olass it is a file with only one molecule
11:38 olass where file is IFile
11:38 olass this used to work before NWO
11:39 egonw ack
11:39 egonw write a unit test please
11:39 olass how do I do that for an IFile?
11:39 egonw I'll look into doing monitoring first
11:39 olass ok
11:39 olass will use MockFile
11:39 egonw olass: see existing tests
11:43 egonw olass: checked
11:43 egonw outcome: I am not sure...
11:43 olass checked?
11:43 egonw I have no clue how to pass a monitor
11:43 olass ok
11:43 egonw the proxy does that
11:43 egonw but
11:43 olass I am writing a unit test
11:43 egonw if you have a monitor already
11:44 egonw from a parent job
11:44 egonw no clue on how to reuse that
11:45 CIA-40 bioclipse: ospjuth * r10510 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.business.test/src/net/bioclipse/cdk/busine​ss/test/AbstractCDKManagerPluginTest.java: Updated testLoadATP() to also test loadMolecules(path) and loadMolecules(MockFile).
11:45 olass egonw: please run my new CDK unit test and tell me the outcome with you
11:45 egonw ack
11:45 olass I think you might need to review the test code too..
11:45 olass was not sure how to properly test it
11:46 egonw ack, hang on
11:46 * olass is hanging on
11:49 egonw grmph
11:49 egonw bug in the jcp config.xml's
11:49 egonw fixing
11:49 egonw sorry about that
11:54 olass egonw: how did the test go?
11:54 egonw mom
11:55 * olass is mom:ing
11:56 masak how did the test go, mom?
11:57 olass masak: your mom does not work here, you have to clean up after yourself!
11:57 * olass hates that sign
11:57 * masak tries desperately not to say something with 'your mom' back to olass
11:59 masak olass: http://en.wikipedia.org/wiki/Your_mom
11:59 edrin jonalv: these manager interception things... what do they do? is another thread copying the objects transfered there?
11:59 zarah masak's link is also http://tinyurl.com/482an
12:00 olass edrin: there is a wiki page about this
12:00 jonalv edrin: I did not understand the second part of that question
12:02 edrin jonalv: to me it seems that the thread or the classpath of the thread my javascript runs in is somehow altered by the manager interception code... could this be?
12:02 CIA-40 bioclipse: egonw * r10511 /bioclipse2/trunk/plugins/ne​t.bioclipse.cdk.jchempaint/ (3 files in 2 dirs): Fixed JChemPaint configuration: bad names in the context xml files, and not proper interface classes in Activator
12:02 masak wow, Shakespeare used 'your mom' jokes in his plays.
12:02 egonw edrin: http://wiki.bioclipse.net/index.p​hp?title=How_to_make_a_manager_2
12:02 zarah egonw's link is also http://tinyurl.com/ouun2r
12:02 jonalv edrin: yes there is an old ugly hack which I thought was gone which might be involved
12:02 jonalv edrin: no that was only used in structuredb
12:03 edrin yes/no?
12:03 egonw masak: such as?
12:03 jonalv edrin: unless you looked at structredb when you did you code you shouldn\t have that ugly hack in your code
12:03 masak egonw: Chiron: "Thou hast undone our mother." Aaron: "Villain, I have done thy mother."
12:04 egonw old english, modern english?
12:04 egonw or would that be the same
12:04 egonw actually...
12:04 egonw no clue what undone refers too
12:04 egonw now I remember why I hated literature
12:04 egonw no logic in it
12:04 egonw just punch lines
12:05 masak egonw: 'undone' here probably means something like 'devastated'.
12:05 egonw so, Aaron confirms
12:05 egonw whatever...
12:06 masak well he puns by use of the double entendre of 'done'.
12:06 egonw hahaha...
12:06 masak 'to do sth/sb'
12:06 egonw what?!?!
12:06 egonw sorry, you're out of my league
12:07 * egonw being in the linguistic irrelevant league
12:07 masak still, it's not much more involved than naked guy with the latex glove... :)
12:07 egonw :)
12:07 masak egonw: just imagine what 'I have done your mother' would have to mean for it to be insulting.
12:08 egonw yes, which is why I asked if old english == new english
12:08 egonw I know what it now means
12:08 masak no, they are different.
12:08 masak don't know if they differ in this instance, though. :P
12:08 masak seems they don't.
12:09 egonw which is nice in line with "no, they are different"
12:10 * egonw is meanwhile still trying to get the CDKmanager plugin tests running
12:10 maclean "undone" means "unravelled" or picked-apart. Like shoelaces are undone.
12:10 masak oh, right.
12:10 * olass is eagerly awaiting egonw's testing
12:11 masak I never hear of things being "ravelled", though.
12:11 maclean there are many words in english that are missing their logical partners
12:11 egonw struggling with the right order of refresh, set target platform, etc
12:11 maclean it's a non-orthoganal language :)
12:11 egonw masak: well, his mother is one
12:12 masak egonw: not anymore. :)
12:14 egonw :(
12:15 egonw the cdk manager tests do not work
12:15 olass what is wrong?
12:15 olass and when did that happen?
12:15 egonw today
12:15 egonw will try set target platform again
12:15 egonw just did that
12:15 egonw but seemed to have helped for Miguel too :)
12:16 egonw !MESSAGE An error occurred while automatically activating bundle net.bioclipse.ui (23).
12:16 egonw !STACK 0
12:16 egonw org.osgi.framework.BundleException: Exception in net.bioclipse.ui.Activator.start() of bundle net.bioclipse.ui.
12:18 olass egonw: remove all deps, add require deps in run config?
12:18 egonw right
12:18 egonw set right product
12:18 egonw BTW
12:18 egonw added screenshots for this in the wiki
12:18 egonw http://wiki.bioclipse.net/index.p​hp?title=How_to_make_a_manager_2
12:18 zarah egonw's link is also http://tinyurl.com/ouun2r
12:20 olass egonw: I can run javacdkmanagertests
12:20 egonw sure
12:20 egonw it is something local
12:21 olass oh, my test passes
12:21 olass :)
12:22 egonw ok, not here
12:22 egonw did PDE class path and project clean
12:22 egonw new set target
12:22 egonw stupid thing is
12:22 egonw bioclipse runs fine
12:23 olass egonw: remove old JavaCDKManagerpluginTest
12:23 olass create a new
12:23 olass that should do it
12:23 egonw yea, can try
12:28 egonw olass: no, that did not help at all
12:30 olass egonw: I did this process (which I always do)
12:30 olass new junit plugin test
12:30 olass stop before completion
12:30 olass open run config
12:30 olass change to bioclipse2 product
12:30 olass add -consolelog -noupdate
12:30 olass plugins tab > run selected plugins
12:30 olass deselect all
12:31 olass click net.bc.cdk.business.tests
12:31 olass select required
12:31 olass run
12:31 olass of course, set target platform before this process
12:33 egonw bioclipse2 product?
12:33 olass ?
12:33 egonw tha is not ui.bioclipse ?
12:33 egonw those arguments...
12:33 egonw are that Program or VM args?
12:33 egonw (that was a second question...=
12:33 olass in run config, Main tab, "Program to run" > net.bioclipse.ui.product in combobox
12:33 egonw (that was a second question...)
12:33 egonw ok, that's what I do too
12:34 egonw same for plugins
12:34 egonw I've done this 1000 times this weekend
12:34 egonw it just stopped working just now
12:34 egonw it worked even this morning
12:34 olass I get really really weird behaviour now
12:34 olass REALLY REALLY weird
12:34 egonw never used those args though...
12:34 olass I can do cdk.loadMolecules(file) sometimes
12:34 olass and sometimes not
12:35 olass very very strange
12:35 egonw ah, sounds like Edrin's problem
12:35 jonalv Gpox: around?
12:35 egonw it's invectious
12:35 olass what is happening!?
12:35 Gpox jonalv: yes
12:35 olass this is really starting to scare me
12:35 olass it means I cannot work
12:36 jonalv Gpox: you where trying to build Bioclispe headless back in the days? How far away from succeding where you? Do you think you could give it a try again and see if you can get a more reasonable error mesage than what I get?
12:36 olass egonw: in my debug I see:
12:36 olass INFO - Bioclipse manager: cdk added to Javascript environment.
12:36 olass INFO - Bioclipse manager: cdk added to Javascript environment.
12:36 olass twice
12:36 olass that does not sound good
12:37 olass why twice?
12:37 egonw better than not once? :)
12:37 egonw have not seen that behaviour yet
12:37 olass setting a breakpoint in CDKManager, I get two calls!
12:37 masak anyone else have a compilation error in net.bioclipse.structuredb.business?
12:38 egonw masak: yes
12:38 masak should I just remove from the list of required plugins it an run without it?
12:38 masak s/an/and/
12:38 egonw yes
12:38 egonw it's for 2.1
12:38 egonw so should not be in list of required plugins
12:39 masak ok.
12:39 olass it indeed calls this methid twice
12:39 olass egonw: please confirm that you do not add the CDK manager twice in AOP
12:39 olass it looks like it here
12:40 olass hmm, I could be wrong...
12:40 Gpox jonalv: sure i will work on it with two different aproches
12:40 egonw olass: where do you thikn I do that
12:41 egonw I have no clue how to add it more than once
12:41 egonw olass: but if you want me to try just that...
12:41 olass no
12:41 jonalv olass: your problem sounds much more fun than mine...
12:41 olass let me look further first
12:41 olass I propably messed up somewhere
12:42 olass will ensure this is not the case first
12:42 jonalv olass, egonw: I am totally stuck. Don't know where to look next...
12:42 egonw jonalv, olass: I guess we have a really long session tomorrow?
12:42 egonw jonalv: can't do much from here...
12:43 jonalv egonw: me netiehr...
12:43 olass damn damn damn
12:43 jonalv it's usually olass wwho get thigns to export...
12:43 jonalv I am usually happy if it starts from within Eclipse...
12:43 olass jonalv: I never have problems with spring
12:43 jonalv and it does...
12:43 jonalv so I can't debug it
12:43 olass this was clearly induced in NWO
12:43 jonalv I can't run osgi commands
12:44 olass I have no attack plan
12:44 * maclean has a bad feeling about this...
12:44 jonalv because everythign just works when I do it from Eclipse
12:44 olass jonalv is the spring master
12:44 jonalv olass is the Eclipse master
12:44 olass if he cannot solve it...
12:44 jonalv if he cannot solve it...
12:45 olass I see no other option but to revert to old world order
12:45 olass that would be REALLY sad
12:45 olass really really sad
12:45 olass but hey
12:45 edrin jonalv: how can i turn off that a manager call is put in a job/thread
12:45 edrin ?
12:46 olass jonalv: can we move classes between plugins?
12:46 jonalv edrin: don't have a progressmonitor on your manager methods
12:46 olass and remove the managers plugin?
12:46 olass since that is causing the error
12:46 jonalv olass: no if we remove the managers plugin we end up with circular deps
12:46 olass jonalv: could its classes be put somewhere else?
12:46 olass jonalv: why so?
12:47 edrin jonalv: "progressmonitor on your manager methods" is this one of the @ things?
12:47 olass edrin: is this not in the wiki page?
12:47 olass then it should be added there
12:47 jonalv edrin: sry I think I am talking about newest world order
12:47 jonalv edrin: but it should be the same for you
12:48 jonalv edrin: if you amanger method takes an Iprgroessmontir then you will get a Job. If not then you won't
12:48 edrin ok
12:48 olass jonalv, egonw: skype on the export error?
12:49 egonw yes
12:49 egonw in 5 mins?
12:49 egonw need to get something to drink
12:49 olass jonalv: ok?
12:49 jonalv sure
12:49 olass ok, 5 mins
12:49 * olass makes more coffee
12:49 edrin jonalv: i noticed that when i run my script sometimes I get two jobs in the progress view. one that reports an error, and one that says OK... how can this be? the code in jsexecution creates only one job!
12:50 jonalv edrin: that sounds strange but it also sounds like what olass reported...
12:50 edrin jonalv: what did olas report?
12:50 olass jonalv: no, it was my fault
12:50 jonalv multiple calls to cdk.loadMoleules
12:50 olass but the debug of two cdk published to JS was a correct observation
12:51 edrin hm...
12:52 edrin however here i sometimes have two jobs, although u strated only one...
12:53 edrin they finished both at the same time, and one is actually the js script that failed (correctly)... and the other just says OK!
12:53 edrin but this is not always the case, only sometimes
12:54 edrin jonalv: do manager calls result in new threads! (somehow intercepted and put in a new thread) ?
12:54 jonalv edrin: sometimes
12:54 jonalv a job is a new thread...
12:55 edrin jonalv: what means sometimes? how can i predict what the manager intercpetor code does ? it is not obvious for me, where is the code located that hooks and filters all calls?
12:55 jonalv edrin: in the old days there are multiple classes that does this
12:55 jonalv in the new there is one
12:56 edrin jonalv: the old classes are removed in trunk?
12:56 jonalv edrin: what pluign are you working on?
12:56 edrin jonalv: the xws4j plugin
12:56 jonalv edrin: your manager comes in one flavuour
12:57 jonalv same fo java and javascript
12:57 edrin huh?
12:57 jonalv that manager has the following advisoers
12:57 jonalv recordingAdvisor
12:57 jonalv classname: RecordingAdvcie
12:58 jonalv edrin: I am reading your context-osgi.xml in your META-INF/spring folder
12:58 edrin ok
12:58 olass egonw: call me and jonalv when you have booted skype
12:58 jonalv it also has wrapInProxyAdvice classname: WrapInProxy|Advice
12:59 edrin what is a warpInProxyAdvice ??
12:59 olass warp speed for managers
12:59 jonalv edrin: you can read the code in the class
12:59 jonalv basically it wraps all IBioObjects returned by the manager in a proxy so that we can record calsl to them
12:59 jonalv edrin: just remove it if you don't want it
12:59 edrin jonalv: hm...
12:59 jonalv edrin: it also has creatJobAdvice
13:00 jonalv class: CreateJobAdvice
13:00 edrin jonalv: but i thought the use of these tricks entirely depend on the @blah things in the managers interface!?
13:00 jonalv those three classes are called on all method calls to you manager
13:00 jonalv edrin: no
13:00 edrin huh!?
13:01 jonalv edrin: no that is not the case
13:01 edrin why not?
13:01 edrin jonalv: what is the difference of these @ stuff in the interface?
13:01 jonalv edrin: because we made that choice...
13:01 edrin so the @ stuff is old stuff?
13:02 jonalv edrin: what @ stuff are we reffering to?
13:02 edrin like @Recorded
13:02 edrin the wiki page talks about @Recorded, @GuiAction, @TestMethods ...
13:02 masak edrin: some attributes have been deprecated, others haven't.
13:03 masak edrin: this all happened very recently, and not all the docs have been updated to reflect that.
13:03 edrin what is this then? isn't this stuff supposed to be the thing that define how a manager call is perforemd?
13:03 edrin uhm....
13:03 edrin maybe then this is the reason!?
13:04 edrin masak: do i need to have the interface at all!?
13:04 edrin or is the whole interface depracted?
13:04 masak edrin: I'm sorry, I didn't follow this conversation from the start. what is your concern?
13:05 edrin masak: i have very strange "random" errors regarding having the wrong classpath/threads... not predicatble, makes no sense.
13:05 masak ok.
13:05 edrin thus i had the feeling something behind the scene was intercepting my calls and put the in new threads
13:07 masak it doesn't sound entirely improbable.
13:07 edrin what happens if i delete the sping folder in the meta inf  of my plugin? will then all the spring hooks leave my manager code alone, and dont touch/intercept stuff???
13:07 masak edrin: I don't know. why don't you try it and tell me how it went?
13:07 edrin hm
13:08 edrin who knows this?
13:08 edrin masak: can you tell me what was changed with this manager stuff?
13:08 masak as far as I know, no-one has ever thought of removing the Spring folder.
13:08 masak edrin: I'm also slowly getting a working knowledge on this. I only have a superficial overview of what changed.
13:09 edrin masak, what is now doing what? the @ stuff in the interface vs. the spring xml config file?
13:09 edrin http://wiki.bioclipse.net/index.p​hp?title=How_to_make_a_manager_2 <- is this page outdated?
13:09 zarah edrin's link is also http://tinyurl.com/ouun2r
13:09 masak edrin: I think so.
13:10 masak edrin: jonalv, the person who made the changes, is sitting in a conference call behind me.
13:10 masak edrin: I suggest you wait until he's done with the call, and ask him.
13:11 edrin ok
13:11 masak edrin: there's a bit of a situation here, because exporting broke with the merging of the new branch.
13:11 edrin i noticed this
13:11 edrin masak: interestingly for me export is working
13:12 masak edrin: that _is_ interesting.
13:13 edrin masak: the others know that it is working for me
13:13 edrin but the changes I did did not help them
13:13 edrin ok, modifying the spring xml file now killed my plugin...
13:14 edrin masak: is there a tutorial for this xml file?
13:14 masak edrin: I'm sure there is. Spring is fairly well documented.
13:21 egonw join #bioclipse-spring
13:21 egonw oh crap
13:25 jonalv left #bioclipse
13:31 edrin egonw: what is #bioclipse-spring ?
13:31 egonw a non-existing channel
13:32 * egonw is away for urgent stuff
13:32 egonw bbl
13:37 edrin ok, since i modified this manager spring stuff i always get this error even when running stuff from bc
13:37 edrin erro from eclipse
13:38 edrin obviously this error is tightly associated with these spring things
13:39 edrin olass: this class loading issue is associated with the spring/manager stuff
13:40 edrin i dont know how and why but it is
13:40 olass edrin: we are working on this, I'll let you know how things turn out
13:41 edrin olass: you mean you found the reason?
13:41 olass edrin: no
13:41 edrin what means working on this then? do you think there is an error in the manager code?
13:43 olass edrin: no
13:43 edrin i think this intercepting code has a bug
14:08 CIA-40 bioclipse: egonw * r10512 /bioclipse2/trunk/plugins/ (3 files in 3 dirs):
14:08 CIA-40 bioclipse: Revert "Not net.bioclipse.core but net.bioclipse.ui is booted first (BC2 boots again)"
14:08 CIA-40 bioclipse: This reverts commit 6eb65c1cdc985df3a2ff4aece6ad8914bfd24c6c (aka 10384).
14:10 CIA-40 bioclipse: ospjuth * r10513 /bioclipse2/trunk/plugins/net.bi​oclipse.ui/META-INF/MANIFEST.MF: Added org.springframework.bundle.osgi.extender as dep to net.bc.core.ui since it is being read in its Activator.
14:12 * Gpox awaits jonalv's commit
14:19 CIA-40 bioclipse: egonw * r10514 /bioclipse2/trunk/plugins/ (3 files in 3 dirs): Removed backup files
14:20 CIA-40 bioclipse: shk3 * r10515 /bioclipse2/trunk/plugins/n​et.bioclipse.spectrum/src/ (3 files in 2 dirs): transferred spectrum similarity calculation to spectrummanager
14:20 edrin Gpox: what is he committing ?
14:23 Gpox i don't know, maby it should have been 'Gpox waints for jonalv's commit' as i did't realy know if it would be one
14:31 CIA-40 bioclipse: shk3 * r10516 /bioclipse2/trunk/plugins/net​.bioclipse.spectrum.test/src/ (2 files in 2 dirs): tests for the wcc similarity
14:32 edrin egonw: are you there?
14:34 masak edrin: I think people are all a bit busy right now, tracking down the problem.
14:36 egonw @pz 1070
14:36 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1070
14:41 egonw all, please check bug report #1080 and vote for it
14:41 egonw /end adv
14:41 masak not 1070, then?
14:41 egonw working on that
14:41 egonw you want to help?
14:41 * masak checks #1080
14:42 masak egonw: sure. it seems top priority. where do I start?
14:42 maclean re : 1080 - I'm not sure that is what 'published' refers to in this context :)
14:42 egonw context: DOI
14:43 maclean doi!
14:43 maclean :) (I know, I was only joking...)
14:44 masak maclean: :)
14:44 maclean (also, try saying "doi" out loud - like 'doyyy!" - it means 'what a fool I've been' )
14:46 masak maclean: related to the "annoyed grunt" of Simpsons fame?
14:46 maclean I think that is "d'oh".
14:46 masak maclean: it is.
14:47 maclean hey, masak : http://gist.github.com/113422
14:47 masak egonw: do you have a particular example of when a doi field would be useful? I'm reluctant to even ask this question, because I sense the whole ticket is based on a pun...
14:47 egonw hahaha
14:48 maclean if only it was in perl6, which I now learn has a postfix<!> operator like 5! = 5*4*3*2*1...
14:48 egonw well, it was triggered by this method
14:48 egonw spectrum.calculateSimilarityWCC()
14:48 egonw consider running:
14:48 egonw man spectrum.calculateSimilarityWCC
14:48 masak maclean: not 'has' as much as 'can create easily'.
14:48 masak maclean: anyway, looks cool.
14:48 egonw and getting pointed to a paper explaining what a WCC is
14:48 masak egonw: couldn't the method description do that?
14:49 egonw (surely then man can also do that)
14:49 maclean 'create easily' =?= 'has'
14:49 egonw masak: but then we cannot make them clickable URLs just as easily
14:49 masak maclean: it's not part of core Perl 6.
14:49 maclean where '=?=' is the 'as near as dammit' operator.
14:49 masak egonw: that's a problem regardless. :)
14:49 masak maclean: in that case, allright, damnit.
14:50 masak :)
14:51 edrin @pz 1080
14:51 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1080
14:51 maclean I look forward to the day that programs can be written with full mathematical operators, like summation (capital sigmas) and product (capital pi).
14:51 maclean although there probably is such a language...
14:51 edrin egonw: I mailed it...
14:51 masak maclean: you should take a look at Fortress.
14:52 edrin egonw: can you reproduce the error?
14:53 shk3 has somebody a few minutes for nwo advise?
14:53 shk3 looks really good, btw
14:53 maclean masak wow. http://projectfortress.sun.com/Project​s/Community/wiki/MathSyntaxInFortress
14:53 shk3 great you did this
14:53 zarah maclean's link is also http://tinyurl.com/rdmrql
14:53 egonw maclean: that exists
14:53 egonw ah, right fortress
14:54 masak maclean: I've been wanting to write something in Fortress for a while now.
14:54 maclean seems more like post-processing your code to get the maths, rather than v.v.
14:54 masak anyone who uses TeX to format their source code is onto something, if you ask me.
14:55 masak maclean: IIU[the page]C, you get immediacy in Emacs.
14:56 masak as usual, Emacs is decades in front of everyone else...
14:56 maclean hmmmmmmm
15:04 CIA-40 bioclipse: shk3 * r10518 /bioclipse2/trunk/plugins/net.bioclipse.seneca/ (24 files in 10 dirs): Seneca jobs can now be run from command line. Also net.bc.compute is not longer used. Still lots to do, but getting there.
15:04 shk3 joined #bioclipse
15:06 CIA-40 bioclipse: ospjuth * r10517 /bioclipse2/trunk/plugins/net.bioclipse.cdk​.business.test/src/net/bioclipse/cdk/busine​ss/test/AbstractCDKManagerPluginTest.java: Added a test to read molecules in a WorkspaceJob. Passes.
15:07 masak egonw: I'm not opposed in principle to a doi field. It'll be a while before we have links in the scripting consoles, though...
15:09 edrin where is jonalv?
15:09 CIA-40 bioclipse: carl_masak * r10519 /bioclipse2/trunk/plugins/net.bioclipse.ui/src/net​/bioclipse/ui/dialogs/UpdatesAvailableDialog.java: [UpdatesAvailableDialog] tidied up dialog, fixes #198
15:09 zarah bug #198 | http://tinyurl.com/po3cgo
15:09 egonw edrin: jonalv is really busy, and has not time for anything else
15:10 masak edrin: he is currently hard at work trying to fix the problems we're having with the build.
15:10 egonw masak: sure, no worries about that
15:11 rojasm joined #bioclipse
15:13 CIA-40 bioclipse: miguelrojasch * r10520 /bioclipse2/trunk/_playground_/org.op​enscience.cdk.qsarweka/src/org/opensc​ience/chemojava/libio/weka/Weka.java: Removed unnecessary System print
15:15 masak in the long run, we should be thinking of moving away from _playground_ and towards branches.
15:16 egonw long run =:=??=== directly after 2.0.0
15:16 masak yes, please.
15:16 egonw no clue about the right operator anymore, after that Perl 6 disaster
15:16 * egonw ducks
15:16 maclean :)
15:16 masak no-one has any clue about the right operator.
15:16 egonw who is he?
15:17 masak hm, Windows' mode of operation seems to be to annoy the user into submission...
15:18 maclean I think that ⊕ is often used for 'generic operation'
15:19 edrin egonw: could you reproduce the bug i have ?
15:20 egonw edrin: no, cannot reproduce anything right npw
15:20 CIA-40 bioclipse: ospjuth * r10521 /bioclipse2/trunk/plugins/net.bioclip​se.ds.dblookup/src/net/bioclipse/ds/d​blookup/impl/DBExactMatchTest.java: Commented out unimplemented stuff to get things moving.
15:21 masak shk3: I'm working on your #1076 right now. just about to confirm that Paste by Ctrl+V actually works on Windows. if it does, then I have to resort to OS detection somehow.
15:22 shk3 Ctrl+V works on Linux
15:22 shk3 Ctrl+C doesn't
15:22 shk3 at least for me
15:22 masak oh, ok.
15:22 masak hm, did I misread the ticket?
15:22 * masak checks
15:23 masak ah, yes.
15:23 edrin masak: can you reproduce the error?
15:23 masak edrin: "the error"?
15:23 edrin masak: i just sent a mail to the dev mailing list
15:24 edrin its really easy to reproduce
15:24 edrin one xml file edited to remove an element, one gist downloaded, gist run, error
15:24 masak edrin: good. I'll see what I can do.
15:25 CIA-40 bioclipse: ospjuth * r10522 /bioclipse2/trunk/plugins/net.bioclips​e.ds.dblookup/src/net/bioclipse/ds/dbl​ookup/impl/DBNearestNeighborTest.java: Commented out unimplemented stuff to get things moving.
15:25 edrin thanks :)
15:25 masak shk3: aha, so you pressed Ctrl+C and it sent you into the 'Paste forwarding' clause?
15:26 shk3 yes
15:26 masak that would explain why it doesn't work, yes. :/
15:26 shk3 sorry if that wasn't clear
15:26 shk3 I am double checking
15:26 masak it was, but I'm apparently one mug of coffee short...
15:27 shk3 yes
15:28 shk3 ctrl+c ends in line 192
15:28 shk3 or it doesn't end there, but it passes there
15:28 masak right.
15:36 CIA-40 bioclipse: gilleain * r10523 /bioclipse2/trunk/plugins/net.bioc​lipse.seneca/META-INF/MANIFEST.MF: Compute plugin removed from dependencies.
15:44 edrin masak: are the reproduce instructions clear?
15:44 masak edrin: sorry, still compiling in Windows, trying to track down shk3's bug.
15:44 edrin ok
15:45 shk3 edrin: shall I have a go at reproducing this?
15:45 edrin shk3: yes sure :)
15:45 edrin would be great
15:53 shk3 how do I get that gist?
15:53 shk3 sorry, did not manage to get into gists till now.
15:54 shk3 can you give me a url?
15:54 * masak 's Eclipse on Windows is 3.4.1, too old for net.bioclipse.managers which wants 3.4.2
15:58 olass @pz 1083
15:58 zarah http://pele.farmbio.uu.se/cgi-bi​n/bugzilla/show_bug.cgi?id=1083
16:03 egonw who is still running 3.4.1?
16:03 egonw (that's not mocking, but a serious question)
16:05 * maclean is still running 3.4.0 on windows
16:07 masak running 3.4.1 on Windows, but 3.4.2 here in the real world.
16:10 masak shk3: could you please try r10522?
16:10 masak shk3: see if that fixes it.
16:10 CIA-40 bioclipse: carl_masak * r10524 /bioclipse2/trunk/plugins/net.bioclipse​.scripting.ui/src/net/bioclipse/scripti​ng/ui/views/ScriptingConsoleView.java: [ScriptingConsoleView] an attempt at fixing #1076
16:10 masak oops, r10524 :)
16:11 * masak 's abilities to do basic arithmetics are failing.
16:12 shk3 masak: you mean the ctrl+c?
16:12 egonw well... ++ -- more operator fuziness
16:12 masak shk3: yes.
16:13 * masak reboots and installs Mac OS 10.5.7
16:19 edrin shk3: sorry: http://gist.github.com/109331
16:19 edrin shk3: ola also could reproduce this error...
16:19 shk3 mom
16:19 shk3 testint masak's stuff
16:20 edrin shk3: going home, will be online again in 40 minutes
16:20 edrin left #bioclipse
16:51 shk3 @tell masak ctrl+c does not work sorry
16:51 zarah Consider it noted.
17:04 CIA-40 bioclipse: goglepox * r10525 /bioclipse2/trunk/plugins/ (3 files in 3 dirs):
17:04 CIA-40 bioclipse: Removed src.includes to fix errors when running Buckminster
17:04 CIA-40 bioclipse: From:
17:04 CIA-40 bioclipse: net.bioclipse.data, netbioclipse.logger and net.bioclipse.usermanager
17:07 egonw Gpox: ping
17:11 Gpox egonw: pong
17:12 egonw you're trying building from the command line?
17:13 Gpox yes, but knowing that several plugins dont compile i forsee some problems
17:13 egonw you cannot exclude those?
17:13 egonw and only build for the core feature plugins?
17:13 Gpox i am working on that in paralell
17:22 shk3 egonw: short question on nwo?
17:22 egonw yes
17:23 shk3 thanks
17:23 shk3 I changed the createSenecaJob method
17:23 CIA-40 bioclipse: ospjuth * r10526 /bioclipse2/trunk/plugins/ (4 files in 2 dirs): Added functionality in manager to add/remove response units.
17:23 shk3 it was createSenecaJob(String)
17:24 shk3 and is now
17:24 shk3 createSenecaJob(IContainer parent, String jobTitle)
17:24 shk3 now I get wrong number of arguments
17:24 shk3 any explanation?
17:24 shk3 is the string converted to file or so?
17:24 egonw is that in the ISenecaManager or SenecaManager?
17:25 edrin joined #bioclipse
17:25 shk3 both have the same signature
17:27 egonw no, don't know
17:27 egonw looks good
17:27 egonw a single method, not run as job
17:28 shk3 well, I will try to solve it
17:28 shk3 thanks anyway
17:37 edrin egonw: i changed the config.xml code according to the instructions
17:37 edrin but the error is still there
17:38 egonw edrin: did you add the new interfaces?
17:38 edrin yes
17:38 edrin but the behavior is still the same!
17:38 edrin java.lang.NoClassDefFoundError: org/apache/xmlbeans/XmlFloat
17:38 edrin at java.lang.Class.getDeclaredMethods0(Native Method)
17:38 edrin at java.lang.Class.privateGetDeclaredMethods(Unknown Source)
17:38 edrin at java.lang.Class.privateGetPublicMethods(Unknown Source)
17:38 egonw OK, please commit your NWO update
17:38 edrin in 50% of calles
17:38 egonw bbl
17:39 CIA-40 bioclipse: edrin_t * r10527 /bioclipse2/trunk/plugins/net.b​ioclipse.xws4j/META-INF/spring/ (context-osgi.xml context.xml): updated to nwo
17:41 maclean joined #bioclipse
17:47 CIA-40 bioclipse: shk3 * r10528 /bioclipse2/trunk/plugins/net.bioclipse.seneca/ (5 files in 4 dirs): worked on new seneca job wizard to comply with standards. unfortunately, the create method in manager is not found. will work on that.
17:51 CIA-40 bioclipse: ospjuth * r10529 /bioclipse2/trunk/plugins/net.biocli​pse.qsar.test/src/net/bioclipse/qsar​/test/business/TestQsarManager.java: Added tests for add/remove response values.
17:53 CIA-40 bioclipse: ospjuth * r10530 /bioclipse2/trunk/plugins/net.biocl​ipse.qsar.ui/src/net/bioclipse/qsar​/ui/prefs/CreateNewUnitDialog.java: Added a prefs page to create new response units.
17:53 CIA-40 bioclipse: ospjuth * r10531 /bioclipse2/trunk/plugins/net.bioclipse.q​sar.ui/src/net/bioclipse/qsar/ui/editors/ (InformationPage.java MoleculesPage.java QsarEditor.java): Added GUI support/dialog to add/edit response units from qsar model.
19:07 * egonw is testing if exporting the last commit before the NWO merge actually worked
19:08 egonw OK, that marks a 'git bisect good'
19:08 egonw and the bad is clear
19:08 egonw here goes
19:10 CIA-40 bioclipse: egonw * r10532 /bioclipse2/trunk/features/ne​t.bioclipse.minimal_feature/ (. .project build.properties feature.xml): Added minimal feature of Bioclipse2 (for testing purposes; fails exporting too)
19:18 egonw Bisecting: 77 revisions left to test after this
19:22 egonw Bisecting: 38 revisions left to test after this
19:23 egonw Bisecting: 19 revisions left to test after this
19:24 edrin olass: interesting, you remember the changes in the config.xml that causes this error i recently showed you=
19:24 edrin ?
19:25 olass yes?
19:25 edrin olass this causes only error when run from the jseditor but not from the jsconsole
19:25 olass edrin: jseditor is your doing so you know best its secrets
19:26 edrin yes, but in fact this all makes no sense to me
19:26 edrin they must be doing it differently in jsconsole
19:26 edrin or do they have any special jsconsole hooks there?
19:27 egonw Bisecting: 9 revisions left to test after this
19:31 egonw Bisecting: 4 revisions left to test after this
19:33 CIA-40 bioclipse: ospjuth * r10533 /bioclipse2/trunk/plugins/net.bioclipse.qsar.ui/ (3 files in 2 dirs): Now catches faulty qsar.xml and will not open editor.
19:39 egonw Bisecting: 2 revisions left to test after this
19:39 egonw exporting is increasingly difficult
19:39 egonw as to having to repeatedly fixing errors...
19:39 edrin interesting, the code xws4j binding does not work even when calling it directly via dynamicimport
19:40 * egonw makes mental note to suggest jonalv to always make patches that leave SVN in a compilable state...
19:40 egonw *even in a branch* !!
19:40 egonw edrin: why not bisect?
19:40 egonw edrin: and determine which commit broke things
19:40 edrin egonw: what do you mean?
19:40 egonw (assuming it worked at some point)
19:40 edrin egonw: i cannot
19:40 egonw google for 'git bisect'
19:40 zarah egonw: http://www.kernel.org/pub/softwa​re/scm/git/docs/git-bisect.html
19:41 egonw ha, zarah, thanx!
19:41 edrin egonw: it just breaks now after i did this modification,
19:41 egonw edrin: did it once work?
19:41 edrin yes, it always worked in eclipse, and it often works in export.
19:41 edrin but often work in export is not good
19:42 edrin so i must find out why it is only often and not always
19:42 egonw so, in which most recent commit did it work in the export?
19:42 edrin egonw: i dont know, could be that we missed it since it is not always but quite rare
19:43 egonw then you first find out if it used to work
19:43 egonw if you have no clue something recent broke it...
19:43 egonw then you may be searching for the complete wrong thing
19:43 egonw knowing which commit broke things is *very* useful
19:43 edrin egonw: i would have no idea where to start...
19:44 egonw and old bioclipse version of which you suspect it worked?
19:44 edrin i have no suspect where it worked...
19:44 egonw then start with the first commit where you introduced a state were xmpp worked at all
19:51 egonw Bisecting: 0 revisions left to test after this
19:58 egonw done
19:58 egonw bad commit found
19:58 edrin egonw: good
19:59 edrin egonw: the problem with the xws4j thing will be more complicated i fear
19:59 egonw why?
19:59 edrin egonw: because i think it might never have worked
19:59 egonw ah, ic
19:59 egonw then you can forget an easy fix
20:00 egonw and we should adapt the paper to do without that feature
20:00 edrin not possible
20:00 edrin we must get this to work
20:00 edrin (and actually it should work, even bc1 was able to do this)
20:01 edrin egonw: i would really like you to have a look at this
20:01 edrin i found some traces but could not figure out why it is behaving like this
20:01 egonw edrin: really...
20:02 egonw I have no clue how this spring stuff works
20:02 egonw I really don't think I can be of any help if it never worked
20:02 egonw not within a reasonable time frame
20:02 egonw let's await tomorrows decision by olass on NWO
20:02 egonw and then we talk
20:03 edrin i am not sure if the NWO is the reason
20:03 edrin egonw: may i demo the problem to you?
20:03 egonw sorry, in the middle of this NWO problem
20:03 egonw can we do this tomorrow?
20:03 edrin ok
20:04 egonw sorry about that
20:04 edrin egonw: just to explain: the code i must get to work is a hack that modifies the classpath to load the dynamically created IoFactories at runtime...
20:05 edrin the problem is this implementation is not stable, since (only) in the export the IoFactory failes because of some obscure errors
20:06 egonw and why can't we remove generating bindings inside bioclipse from the paper?
20:07 edrin because i think this is a cool feature that should be in...
20:07 edrin how should we demo the Bio Schemata service this way?
20:08 egonw it's certainly a cool feature
20:08 edrin yes
20:08 egonw from the command line?
20:09 egonw with some Ant script
20:09 egonw a groovy script
20:09 egonw enough options
20:09 egonw it's not critical to the paper to create the binding inside bc2
20:09 edrin egonw: the funny thing is, the hack seems to work without problems from the js console...
20:09 egonw and apparently never worked
20:09 egonw oh...
20:09 edrin i think it worked
20:09 egonw can you confirm in the export?
20:10 edrin second...
20:10 egonw yes, please do make sure
20:10 edrin have to export again
20:10 egonw which commit revision worked, in JS console, also in editor, also in export?
20:10 egonw please write those things down
20:10 egonw particularly what worked in which revision is important
20:10 edrin i dont know
20:11 egonw test it
20:11 edrin the thing was never tested in an export
20:11 egonw then do that now
20:11 egonw it's a lot of work
20:11 edrin i cannot cheout 1000 workspaces now
20:11 egonw sure
20:11 egonw that's where bisecting comes in
20:11 egonw 1000 is only 2^10
20:12 egonw so, maybe some 12 commits to pinpoint the commit that broke something
20:12 egonw ...
20:12 egonw *assuming* it once worked
20:12 edrin egonw: we cannot assume this
20:12 egonw no, true, but you can test that too
20:12 edrin this is a hack of the classpath
20:12 egonw computer software != patients
20:13 egonw with computer software you *can* go back in time
20:13 edrin sure, but even if i checkout old workspaces, i dont know how to test if it worked
20:14 egonw testing if it exports does not change...
20:14 edrin because testing this requires modification of code or exporting and doing very complicated thing to test this random error
20:14 egonw yes
20:14 egonw that's why I say, we should drop it
20:14 egonw and do a nasty workaround for the paper
20:14 egonw well... nasty... not even that...
20:15 edrin i am sure that jonalv would be able to solve this issue
20:15 egonw but we clearly do not understand the Spring/Eclipse stuff...
20:15 edrin since he knows what spring does
20:15 egonw I'm sure he does not
20:15 edrin what is the difference between the jsconsole and the jsexecution/jseditor!?
20:15 egonw he is getting is brain already stressed up with the current export issue
20:16 egonw anyway... you can invite him for the paper...
20:16 egonw getting this working would clearly be a non-trivial contribution to the work
20:16 edrin indeed
20:16 egonw but no clue if he has time for it... if he can make time for it
20:16 edrin hm
20:16 egonw if he actually would know how to fix it...
20:17 egonw but that is, surely, for him to answer
20:17 edrin I'll go to the spring chang and ask them
20:26 edrin i think i have an idea
20:27 egonw I'm listening
20:28 edrin have to elaborate this first
20:28 edrin it will be a workaround but maybe solve this issue for the future
20:29 egonw oh...
20:29 egonw I made progress!
20:29 egonw I found a problem!
20:29 egonw don't understand it...
20:29 egonw but I made progress!
20:30 edrin well
20:30 edrin i will try to create the classloader for the iofactory by hand instead of passing the classloader of the plugin to it...
20:31 edrin egonw: i assume your progress was meaning my progress ...
20:31 egonw let's hope
20:31 egonw but let my progress not stop you
20:31 egonw custom class laoders are always a good bet
20:31 edrin yes
20:32 edrin currently i was doing Thread.currentThread().getContextClassLoader();
20:32 edrin and i think the currentThread is not predictable when having spring
20:32 edrin and that is the source of the problem �maybe
20:33 egonw OK, that will explain some trouble indeed
20:33 egonw because you have no idea what plugin the current thread comes from
20:33 egonw blah...
20:33 egonw did not one review your code?
20:33 edrin egon, it used to work :)
20:33 egonw I could have said that
20:33 edrin for months
20:33 egonw sure... with the proper single threading
20:33 egonw but that got 'fixed' at some moment
20:34 egonw because that did not work for many other situations
20:34 egonw currentThread was always the JS threa
20:34 egonw giving trouble with the GUI thread
20:34 egonw and blocking
20:34 egonw and etc...
20:34 edrin egonw: i think it is even more complicated now
20:34 egonw than what?
20:35 edrin i think that spring "plays with any thread" in the background
20:35 edrin ok, i will now try to modify my code a bit
20:37 edrin oh, maybe this is even more easy than i thought
20:46 egonw damn, and now back to square one :(
21:55 edrin joined #bioclipse
22:24 egonw OK, another step forward
22:29 egonw and stuck again
22:29 CIA-40 bioclipse: lc69 * r10534 /bioclipse2/trunk/plugins/net.bioclipse.​ds.signsig.bursi/exec/signatures-linux: Fixed a bug in the signatures code. On output the signatures were sorted which meant that we lost the order of the atoms corresponding to the signatures.
22:31 CIA-40 bioclipse: edrin_t * r10535 /bioclipse2/trunk/plugins/net.bioclipse.xws​4j/src/net/bioclipse/xws4j/Activator.java: added some internal xws for java static code setting (just internal) will do some tricks
22:36 CIA-40 bioclipse: edrin_t * r10536 /bioclipse2/trunk/plugins/net.bioclipse.​xws4j/src/net/bioclipse/xws4j/business/ (3 files): udapted to NWO
22:37 CIA-40 bioclipse: edrin_t * r10537 /bioclipse2/trunk/plugins/net.bioclipse.xws4j/ (7 files in 3 dirs): updated to current xws4j svn
22:40 egonw edrin: please check the license of IJavaXwsManager and IJavaScriptXwsManager
22:40 egonw please also use the header format I sent to the -devel ML
22:45 edrin egonw: header format?
22:45 edrin egonw you mean eclipse public license?
22:45 egonw right
22:46 egonw no, the template I sent to the list
22:46 * edrin checks
22:47 egonw edrin: is the license GPL?
22:47 egonw of xws4j?
22:47 edrin no
22:47 edrin xws4j is lgpl
22:47 egonw OK, then your license text is actualy also wrong
22:48 egonw and not just the template
22:48 * egonw gives up on spring
22:48 egonw tried
22:48 edrin [Bioclipse-devel] Template copyright header <- this one?
22:48 egonw tried really hard
22:48 egonw yes, that one
22:48 edrin egonw
22:48 edrin problem solved btw
22:48 egonw even fiddled with the spring jars...
22:48 edrin my problem
22:48 egonw edrin: but we won't know it if works in the export
22:49 edrin i do
22:49 edrin because export works for me :)
22:49 egonw oh, right, you can still export...
22:49 egonw ic
22:49 egonw what eclipse version, platform, etc are you using?
22:50 edrin i downloaded a fresh eclipse 4 days ago, then reinstalled all plugins like svn, EMF...
22:50 egonw so, eclipse 3.4.2 ?
22:50 edrin then switched to my old workspace, deleted bc run config
22:50 edrin yep
22:50 egonw what OS?
22:50 edrin windows
22:50 egonw you're on windows, right?
22:50 egonw ok
22:51 edrin egonw: the problem is now really solved in xws :) i can neither reproduce the random problem
22:51 egonw ok, good
22:51 edrin and the manager works now even when removing this interceptor stuff...
22:51 edrin from eclipse
22:52 egonw edrin: do you have the delta pack installed?
22:52 edrin yes
22:53 edrin egonw: did anyone try to export with a fresh eclipse?
22:53 edrin @pz
22:53 egonw I think they tried 3.5RC1
22:53 zarah edrin: Pelezilla is at http://bugs.bioclipse.net
22:54 edrin egonw: i can create builds for the xmpp wiki
22:54 egonw well, I guess the good news is...
22:54 egonw that the current SVN version is not broken in itself
22:55 egonw but that it has something to do with our installations...
22:55 egonw are you running/exporting with features?
22:55 edrin egonw: btw. olas hivpred is now working too
22:55 egonw ha, good :)
22:55 edrin yes, i export with features
22:55 egonw yeah... feeling stupid, for not having paid more attention to your code...
22:56 egonw would have raised questions about getting the thread classloader...
22:56 edrin hm :/
22:57 egonw but the flux of code is so high
22:57 edrin egonw: again a good reason to keep the xmlbeans and stuff in the jar folder of my plugin...
22:57 egonw well, maybe not...
22:57 egonw but it sounds so logical
22:57 egonw at least after the reason why NWO was started
22:58 edrin what was the reason?
22:58 egonw threading
22:58 egonw things not ending up in the correct thread
22:58 edrin like having multiple js threads?
22:58 edrin :)
22:58 egonw so, the AOP stuff now ensure things are in the proper threa
22:58 edrin ok
22:58 egonw but that makes currentThread rather undefined
22:58 edrin ic
22:58 egonw yes, like multiple js threads, I think
22:59 edrin the NWO stuff was really new for me
22:59 edrin it came from one day to the other
22:59 egonw it came last Thursday
22:59 edrin yes
22:59 edrin and i did not notice it
23:00 edrin anyway
23:00 edrin thing is working
23:00 edrin i did not cheout current trunk
23:00 edrin i could create builds for all platforms
23:00 edrin i mean i better do not checkout stuff now
23:01 edrin since it seems to work now...
23:01 egonw edrin: please do...
23:01 egonw linux 32/64 bit
23:01 egonw OS/X
23:01 egonw well, whatever is listed for bioclipse-xmpp now
23:01 edrin i could also switch back to Ola's old screenshot
23:02 egonw yes
23:02 egonw sounds good
23:02 edrin but will leave cdk calculate mass in the tutorial
23:04 egonw OK, installing delta pack did not help
23:05 egonw going to try a fresh 3.4.2 install now...
23:05 egonw last thing really
23:05 egonw for today
23:05 edrin egonw: maybe try a new eclipse?
23:05 egonw [01:05:06] <egonw> going to try a fresh 3.4.2 install now...
23:05 edrin ok
23:05 edrin :)
23:06 edrin the
23:06 edrin i will stay and create some builds...
23:07 egonw thanx
23:07 egonw much appreciated
23:07 egonw this basically means that we do not need to revert the patches
23:07 egonw but only figure out who export works for you, but not for us
23:09 edrin ok, building
23:09 edrin will take some time...
23:15 * egonw is rebuilding workspace in fresh 3.4.2
23:15 egonw edrin: did you make a fresh workspace too?
23:15 edrin egonw no
23:16 edrin egonw: i just deleted the bioclipse run configuration
23:16 edrin and then updated to current trunk
23:17 egonw mmm
23:17 egonw reboot needed
23:17 egonw another missing feature
23:17 edrin and then i had to remove some stuff of the core manifest (as written in the bugtraq)
23:17 egonw ack
23:17 egonw that seems to be in SVN now
23:18 edrin egonw: and then i had to checkout 1 billion cdk plugins to build from cheminformatics feature
23:18 egonw :)
23:18 egonw sure
23:18 edrin one after the other...
23:18 egonw you're welcome :)
23:22 egonw OK, exact same error
23:22 egonw new Eclipse 3.4.2 did not help here yet
23:24 egonw ok, good luck with the exports!
23:24 egonw cu tomorrow
23:25 edrin cu
23:26 edrin exports donw
23:26 edrin how can i upload them in the wiki?
23:30 edrin hm disabled

| Channels | #bioclipse index | Today | | Search | Google Search | Plain-Text | summary