Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-04-14

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

All times shown according to UTC.

Time Nick Message
07:01 Gpox joined #bioclipse
07:42 rojasm joined #bioclipse
07:46 egonw joined #bioclipse
07:54 jonalv joined #bioclipse
07:55 olass joined #bioclipse
07:56 masak joined #bioclipse
08:08 shk3 joined #bioclipse
08:25 jonalv git++
08:25 jonalv time-travel++
08:26 masak git++
08:27 masak @karma travel
08:27 egonw git++++
08:27 zarah travel has a karma of 1
08:27 masak @karma time-travel
08:27 zarah time-travel has a karma of 0
08:27 masak jonalv: .
08:27 masak (time-travel)++
08:27 masak @karma time-travel
08:27 zarah time-travel has a karma of 1
08:27 jonalv zarah--
08:28 masak jonalv-- # refuses to accept that he's doing it wrong, blames zarah
08:28 jonalv masak: so if I think zarah is doing things wrong than I get less karma?
08:28 masak jonalv: yes, if you're unfair.
08:28 masak zarah: you're a smart girl.
08:28 zarah masak: Never mind the compliment! Take your hand off my knee!
08:28 jonalv masak: I don't think I was unfair
08:29 jonalv masak: I think you are though
08:29 masak jonalv: the rules are very clear and easy to follow.
08:29 jonalv masak: yes, complain and masak removes karma
08:29 masak :)
08:29 jonalv masak: now I know...
08:29 masak right.
08:30 masak jonalv-- # you did it again (translated ignorance to the fault of others)
08:42 jonalv masak: remind me not to talk to you
08:42 masak seems we got off on the wrong foot this morning. was it something I said?
08:52 CIA-73 bioclipse: shk3 * r9776 /bioclipse2/trunk/plugins/net.bioclipse.chemoinfo​rmatics/src/net/bioclipse/chemoinformatics/conten​tlabelproviders/MoleculeFileContentProvider.java: empty folders no longer added to children in the MoleculeFileContentProvider. Note it only counts children, not MoleculeChildren, which would require lots of checks for ContentType.
09:01 CIA-73 bioclipse: shk3 * r9777 /bioclipse2/trunk/plugins/net.bio​clipse.spectrum/build.properties: added files to build
09:09 jonalv masak: I hate you
09:09 jonalv masak: was it something you said? No it ws something you did. You took away karma
09:09 masak jonalv: relax. it's just digits.
09:10 jonalv masak: it's the spirit that counts. Imagine what would happend if I took away karma from olass...
09:14 masak jonalv: well, you criticize zarah for no good reason, so I removed some karma. but I don't hold grudges. if you don't do it again, I won't either.
09:15 jonalv masak: I critised zarah because she does't recognize that time-travel is a word
09:15 jonalv or a thing
09:16 masak that's the way she works.
09:16 masak the source is open.
09:16 jonalv masak: yes and I critized that way but appharently I am not allowed to do that...
09:16 masak jonalv: I know. and I removed karma for exactly that.
09:16 masak case closed.
09:17 jonalv masak: so I am not allowed to critizise zarah whatsoever?
09:17 jonalv masak: this is not china you know...
09:17 masak I'm not silencing you. just removing some karma.
09:17 masak not for your opinions as such, just for your attitude.
09:17 masak zarah isn't broken or anything.
09:17 masak she just abides by very simple rules.
09:18 jonalv masak: what's wrong with my attitude? IT's not me that thinks that simply should be defended.
09:18 masak jonalv: look.
09:19 masak I've written zarah.
09:19 masak she's "my" bot.
09:19 masak when you criticize her, you criticize my work.
09:19 jonalv masak: appharenlty you don't take critizism very good...
09:19 masak maybe not.
09:20 masak or maybe I just think that you're not taking the time to figure out how things work instead of complaining.
09:20 maclean joined #bioclipse
09:20 masak maclean: welcome. we're having a flame party.
09:20 jonalv masak: of course I don't. zarah is a black box for me. What's wrong with that?
09:20 egonw it sounds like we could all use very much a Bioclipse 2.0.0 release
09:20 maclean oh good.
09:21 egonw and have something to be happy about
09:21 * maclean backlogs
09:21 masak jonalv: it's that attitude that I'm giving you -- for.
09:21 egonw black boxes are evil
09:21 egonw they should be green
09:21 masak egonw++
09:21 jonalv maclean: don't comment on anyhting masak has written.
09:22 masak :)
09:22 maclean er...okay.
09:22 jonalv you will regret that...
09:22 masak maclean: he's not threatening you, just trying to warn you.
09:22 maclean I have never had cause to complain about masak's code.
09:22 maclean and I love to complain about code.
09:22 maclean but maybe I just haven't seen enough of it...
09:23 masak I love to complain about code too.
09:23 masak but I prefer to do it when I know I'm right.
09:23 masak not when I think I know what's going on inside the black box.
09:24 masak besides, calling zarah a black box is just silly.
09:24 jonalv masak: I know that I am right. time-travel++ should not give travel more karma
09:24 masak jonalv: that's the way it works, and it's worked like that from the beginning.
09:24 masak - is not a word character.
09:24 jonalv masak: what kind of argument is that?
09:25 masak an arguemnt for simplicity.
09:25 maclean oh right, it splits words on \s*\w+\s* - right?
09:25 masak maclean: something like that.
09:25 masak jonalv: see, maclean understands.
09:25 jonalv just because things always have worked in one way doesn't mean that way is the best. I am supriosed I have to tell this to you...
09:25 maclean I'm rubbish at regexes, though
09:25 masak jonalv: that's not what I'm saying either.
09:25 jonalv I am not saying that I don't understand I am only saying that I think it shouldn't work that way
09:25 masak jonalv: if you'll excuse me, I'd like to work a little now.
09:26 jonalv lol
09:26 masak patches welcome.
09:27 jonalv the cdk manager is a mess... :(
09:27 * maclean timidly raises the point that 'it is only a chat-bot [or (chat-bot) ...]
09:28 egonw only a chat bot?!?!? how can you be so incensitive ?!?!
09:28 masak :D
09:28 egonw it's zarah you are talking about
09:28 CIA-73 bioclipse: carl_masak * r9778 /bioclipse2/trunk/plugins/net.biocli​pse.scripting.ui/src/net/bioclipse/s​cripting/ui/views/RConsoleView.java:
09:28 CIA-73 bioclipse: [net.bioclipse.scripting.ui] removed stubbed RConsole
09:28 CIA-73 bioclipse: The real one lives in net.bioclipse.r.
09:28 masak maclean: what egonw said. I'm tempted to give you -- too now. :)
09:28 masak actually...
09:28 masak maclean-- # in all fairness
09:28 masak zarah: you're a smart girl.
09:28 zarah masak: Not so loud, dear. I don't want my boyfriend to hear. Over.
09:28 CIA-73 bioclipse: carl_masak * r9779 /bioclipse2/trunk/plugins/net.bioclipse​.scripting.ui/src/net/bioclipse/scripti​ng/ui/views/ScriptingConsoleView.java:
09:28 CIA-73 bioclipse: [ScriptingConsoleView] now catches \r and eats them
09:28 CIA-73 bioclipse: Since R sends its output in little chunks, we can't just replace "\r\n" with
09:28 CIA-73 bioclipse: NEWLINE and be done with it. Instead, we replace "\r" with "", and "\n" with
09:29 CIA-73 bioclipse: NEWLINE. Seems to work.
09:29 maclean Go ahead. My kudos...er...karma could not possibly be affected :)
09:29 masak :)
09:30 jonalv guys we need to do somethign about the cdk manager...
09:32 jonalv it's a mess...
09:33 maclean oh?
09:33 masak olass: ping
09:33 shk3 in which respect?
09:35 maclean looks okay. some odd logging messages : "Rad CDK chemfile with format: "
09:36 maclean and "Hej, you are running OS/X! You just encountered a known bug"
09:36 egonw :)
09:37 egonw yes, correct
09:37 egonw that should say:
09:37 egonw "Hej, you are running OS/X! You just encountered a known bug in the OS/X JVM"
09:37 egonw "Hej, you are running OS/X! You just encountered a known bug in the OS/X JVM/Eclipse combo"
09:38 maclean did you report it?
09:38 egonw no
09:39 egonw no idea how to make a proper report for this
09:39 egonw it takes some effort to make a good bug report, particularly so for Eclipse...
09:39 maclean true. it's also quite a combination of things to test.
09:40 maclean for the "steps to reproduce"
09:40 egonw right
09:46 egonw so, should I put rednael to work here with Planet Bioclipse as RSS feed?
09:50 olass egonw: sounds good
09:50 masak egonw: does Bugzilla have an RSS feed?
09:50 olass interesting
09:50 egonw I'll check
09:50 egonw for all new reports
09:50 egonw not all changes I hope
09:51 olass would be really nice
09:51 olass to get them here
09:51 masak egonw: new reports should be enough, yes.
09:51 olass hmm, maybe
09:51 egonw masak: thanx for the Groovy book
09:51 egonw Groovy is groovy indeed
09:51 masak so they say.
09:51 masak you're welcome.
09:52 egonw it just removed the last reason why I was still using Perl
09:52 masak oh, really? :)
09:52 shk3 jonalv: I need some explanation on the job stuff.
09:52 egonw luckily for Perl, there is version 6
09:52 egonw which may just have some candy I cannot live without
09:52 masak egonw: Perl 5 is not so bad either, to be honest.
09:53 shk3 if I want a method to show a progressmonitor, what do I need to do?
09:53 shk3 add an IProgressMonitor monitor parameter to the method?
09:53 shk3 anything else?
09:53 shk3 what is BioclipseUIJob for?
09:53 shk3 is this related to the progress monitor?
09:56 masak shk3: jonalv just left to hunt lunch.
09:56 masak shk3: have you read the wiki page about the New World Order?
09:56 shk3 yes
09:56 masak oh, ok.
09:56 masak then you probably know more than I do. I've only read half.
09:57 shk3 it's not yet totally clear for me, though
10:05 rednael joined #bioclipse
10:08 rednael [master] Bioclipse: a powerful Jmol application  http://chemicalrcp.blogspot.com/2009/04/​bioclipse-powerful-jmol-application.html
10:08 zarah rednael's link is also http://tinyurl.com/d2gwbr
10:08 egonw rednael++
10:09 egonw wrong label, though
10:09 egonw egonw--
10:09 egonw zarah: no complaint about that, do you? :)
10:09 zarah egonw: I see.
10:09 masak hm, I'll get zarah to ignore rednael.
10:09 egonw why?
10:09 masak shouldn't I?
10:10 egonw no
10:10 masak I don't want bots to reply to bots.
10:10 egonw not Planet Bioclipse links at least
10:10 egonw it's nice to have those tiny too
10:10 masak ok, then I'll leave it as it is.
10:12 rednael [master] Better labels, instead of CDK branch name  http://github.com/egonw/rednael/commit/a​8866e99a7e7fe1d113aa15d5627d922ddbb3fc8
10:12 zarah rednael's link is also http://tinyurl.com/czb9e5
10:14 rednael joined #bioclipse
10:33 CIA-73 bioclipse: carl_masak * r9780 /bioclipse2/trunk/plugins/net.bioclipse.jsedito​r/src/net/bioclipse/jseditor/RhinoConsole.java:
10:33 CIA-73 bioclipse: [RhinoConsole] whitespace therapy
10:33 CIA-73 bioclipse: - introduced empty lines between methods
10:33 CIA-73 bioclipse: - shortened lines to be within 80 chars
10:33 CIA-73 bioclipse: - intented things consistently and without tab characters
10:33 CIA-73 bioclipse: - some vertical alignment
10:35 CIA-73 bioclipse: carl_masak * r9781 /bioclipse2/trunk/plugins/ (4 files in 3 dirs):
10:35 CIA-73 bioclipse: [net.bioclipse.jseditor] prevented Eclipse console from opening
10:35 CIA-73 bioclipse: Removed class RhinoConsole from the codebase. Added comment about the commit
10:35 CIA-73 bioclipse: where it can archeologically be found. A problem that needs to be addressed
10:35 CIA-73 bioclipse: now is whether we want to be able to print stuff from the js editor, and if so,
10:35 CIA-73 bioclipse: where.
10:37 shk3 @all:
10:37 shk3 In CDKPluginMangerTest, i get 28 errors.
10:37 shk3 Most (or even all, didn't check)
10:37 shk3 of these are due to UnsupportedOperationException
10:38 shk3 because they use loadMolecule.
10:38 shk3 Can somebody check this and confirm or tell me,
10:38 shk3 what I am doing wrong?
10:38 masak shk3: I'm sure that jonalv can, but he's on lunch.
10:39 masak he'll be back in half an hour or so.
10:40 shk3 can't you just run the test?
10:46 egonw shk3: yes, this is a know problem
10:46 egonw there is a bug report about it on PZ
10:47 olass shk3: jonalv has that bug
10:47 olass and he is working on it
10:48 shk3 olass: one question
10:49 olass sure
10:49 shk3 when do you think will there be a bc2 release?
10:49 jonalv shk3: yes I am on it. IT's a bit messy though... :(
10:49 olass RC1 in two weeks approx
10:49 shk3 I won't do anything on keeping things up to date to the current design till there is a release.
10:49 olass I think it's realistic to say 4 weeks until bc 2.0
10:49 olass shk3: ack
10:49 shk3 I can't afford spending most of my time on keeping in line with all your back and forth changes
10:49 olass but RC1 will be feature freeze
10:50 olass shk3: do as you like
10:50 shk3 this is escpecially bad if I tried to do something months ago
10:50 shk3 was declared an idiot for it
10:50 shk3 and now a similar change was done
10:50 shk3 and I am supposed to keep in sync with it.
10:50 olass shk3: Bioclipse is not stable
10:51 shk3 yes
10:51 olass you have to live with that
10:51 olass after 2.0 things will settle
10:51 egonw shk3: agree... the CDKManagerPluginTest should not have been broken in the first place
10:51 egonw olass: there is no need for two standards here
10:51 olass ?
10:51 egonw don't break things applies to all stuff...
10:51 olass right
10:52 egonw the CDKManagerPluginTest unfortunately got unnoticed
10:52 CIA-73 bioclipse: shk3 * r9782 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.business/src/net/bioclipse/cdk/business/ (CDKManager.java ICDKManager.java): tried to add a progress monitor to coordinate generation
10:52 olass yes
10:52 * jonalv ducks and hides *he is not here*
10:52 egonw shk3: be assured... the one who broke it has seen the ugly looks from others here
10:52 olass :)
10:52 egonw shk3: unfortunately, you have been witnessed that too
10:54 CIA-73 bioclipse: shk3 * r9783 /bioclipse2/trunk/plugins/net.bioclipse.cdk.bus​iness.test/src/net/bioclipse/cdk/business/test/ (CDKManagerPluginTest.java CDKManagerTest.java): adopted the tests
10:54 jonalv egonw, shk3, olass: I merged my changes to quick. I obviusoly wasn't ready. I am sorry for this.
10:55 olass jonalv: apologize accepted :)
10:55 olass now please fix it ;)
10:56 * jonalv dives down into rewriting the CDKManagerPluginTest now...
10:56 CIA-73 bioclipse: shk3 * r9784 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui.sdf​editor/src/net/bioclipse/cdk/ui/sdfeditor/editor/ (JCPCellPainter.java MoleculesEditorLabelProvider.java): adopted to changes in cdkmanager
10:56 maclean how does the cdk.smartsmatching plugin manifest itself? Er...I mean - how do you use as a bioclipse user?
10:56 shk3 jonalv: could you tell me something about jobs before you do that?
10:56 jonalv shk3: maybe
10:56 shk3 I added a progressMonitor to the generate coordinate methods
10:57 egonw olass: what was that about BioList again?
10:57 jonalv shk3: in cdk manager?
10:57 shk3 yes
10:57 maclean wait a minute, maybe its this cdk.smartsmatches method. hmmm.
10:57 shk3 if I now want to use it from a handler, what shall I do?
10:58 egonw olass: regarding use of IMolecule[]
10:58 shk3 pass a NullprogressMonitor?
10:58 olass egonw: BioList is recorded
10:58 olass or what was the question?
10:58 egonw BioList replaces IMolecule[] right?
10:58 egonw or
10:58 egonw does it replace List<IMolecule>?
10:58 olass the latter
10:58 olass well
10:59 olass ArrayList<IMolecule>
10:59 egonw why that?
10:59 olass BioList implements List
10:59 jonalv shk3: no, you call the method taking only an IFile which creates a jobn and calls the method taking an IFIle and that jobs progressmonitor
10:59 olass egonw: BioList is an implementation
11:00 olass List<IMolecule> olaslist = new BioList<ICDKMolecule>()
11:00 egonw but the interfaces is pure List, good
11:00 olass yes
11:00 jonalv egonw: BioList is an implementation of Javas List interface which allows for recording of calls to the lists methods
11:00 shk3 jonalv: so i need both methods in the manger, with and without ProgressMonitor?
11:00 jonalv egonw: it also tracks what bioobjects are in what biolist
11:00 olass so instead of instantiation ArrayList<IBioObject> we should use BioList, when dealing with IBioObjects
11:01 egonw shk3: can you make the coordinate generation methods use BioList instead of arrays?
11:01 jonalv shk3: yes... ugly I know. I am planning to come up with a nicer way of doing this after 2.0....
11:01 olass yes, that would be good
11:01 olass (to egonw's suggestion)
11:02 shk3 no
11:02 shk3 I will not do this.
11:02 shk3 I will make this progressmonitor stuff working
11:02 shk3 and then i will not do anything
11:02 shk3 ever
11:02 shk3 perhaps once bc2 is out
11:02 jonalv shk3: I don't like your attitude
11:02 olass shk3: if you write code in Bioclipse you take on supporting it
11:02 olass at least for now
11:02 olass you disagree with this?
11:03 * egonw is fine with seeing that patch post 2.0
11:03 egonw recording is post 2.0 too
11:03 olass indeed so
11:03 shk3 but then we need a sort of one-in-all policy
11:03 olass egonw: file a bug for the record
11:03 egonw I think shk3 just said that no, not before 2.0, but will look at it afterwards
11:03 egonw shk3: correct?
11:03 jonalv I am sorry that I am not perfect and can come up with the perfect solution at my first attempt. I need a few iterations to reach perfection. It's not so easy for a bear with a very small brain...
11:03 shk3 once we had the idea of external people writing for bc
11:03 * olass hopes so
11:04 shk3 if we want this, we cannot expect them to do changes every other day
11:04 olass shk3: exactly what are you angry for?
11:04 shk3 I am not angry
11:04 olass that Bioclipse does not have a stable API?
11:04 olass sounds like it
11:04 egonw shk3: correct, which is why we need it to be good the first time
11:04 shk3 I just want things to be a bit more economical
11:05 shk3 for me, right now
11:05 egonw shk3: can you elaborate on that?
11:05 shk3 but also for others in the long run
11:05 egonw what do you mean with 'more economical?
11:05 shk3 applying changes less piecewise
11:06 egonw you mean larger commits?
11:06 shk3 jonalv: do the to methods call each other automatically or do I need to do it?
11:06 shk3 egonw: larger commits for me, yes
11:07 masak joined #bioclipse
11:07 egonw but larger commits are more diffcult to review
11:07 shk3 once the api is stable, I am happy to apply changes
11:07 olass hi masak
11:07 jonalv shk3: I am guessing the answer to that question is that the proxy makes sure that the right method is called automatically
11:07 shk3 sorry, what's the proxy here?
11:07 egonw code quality and unit testing must be important to the Bioclipse econymo too?
11:07 shk3 don't understand that word here.
11:07 egonw economy
11:08 jonalv the proxy is the object that wraps the manager. The proxy object is constructed by Spring
11:08 shk3 sure, but shouldn't be a problem to review an update to current api
11:08 shk3 ah ok
11:08 jonalv shk3: which manager is it that you want to wait with updating until the api has settled down so to say?
11:09 shk3 but what means "making sure the right method is called"?
11:09 jonalv shk3: sry, calls the right method
11:09 shk3 I need to use the method from a handler.
11:09 shk3 which one shall I call?
11:09 shk3 in the handler, I mean
11:10 jonalv the one taking an IFile
11:10 olass shk3: I have not made my managers into New World Order
11:10 olass and I don't think they will be until post 2.0
11:10 olass but I am not sure
11:10 olass but it does reduce their features
11:10 olass I'll see what I can make time for
11:10 shk3 and do I need to call the method with progress monitor in there
11:11 shk3 or is this done automatically?
11:11 jonalv shk3: does this method return something?
11:11 shk3 like the Path->IFile translation (which doesn't work)
11:11 shk3 yes, it returns something
11:12 jonalv shk3: then how are you expecting to run it in a job if I may ask?
11:12 * masak eats home-baked bread
11:12 jonalv masak: sounds good
11:12 shk3 I don't know
11:12 jonalv shk3: okey
11:12 shk3 that's why I am asking you
11:13 shk3 or do i ned to use this BioclipseUIJob stuff instead?
11:13 jonalv shk3: I am sorry but could I trouble you to explain one more to me what it is that you want to do?
11:13 jonalv |one more time
11:14 jonalv shk3: YEs I think you want to use the BioclipseUIJob stuff. Gpox is the real experton that one but I hope I know enough... :)
11:14 shk3 I want to fix bug #830
11:14 shk3 in a nutshell
11:14 jonalv @pz 830
11:14 jonalv zarah?
11:14 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=830
11:15 jonalv shk3: ah, trixy. So the handler generates 3d and the triggers some refresh?
11:15 jonalv Gpox: are you around?
11:15 shk3 yes
11:16 shk3 it uses the manager method
11:16 Gpox yes
11:16 jonalv shk3: you want to use the BioclipseUIJob stuff.
11:16 jonalv Gpox: Do you think you can explain. You have used this right?
11:17 shk3 http://wiki.bioclipse.net/index.php?t​itle=How_to_make_a_manager#On_the_top​ic_of_running_manager_methods_as_jobs tells to use IProgressMonitor
11:17 zarah shk3's link is also http://tinyurl.com/dypvah
11:18 jonalv shk3: yes, but this is somewhat of a special case. Your method returns the answer and then you want to do something with the returned value....
11:19 shk3 ok
11:19 shk3 so in this case the BioclipseUIJob is the way to go?
11:20 jonalv shk3: yes, but you still need all the other stuff. The BioclipseUIJob is just something more that is used for updating the gui once your job has finnished...
11:20 masak joined #bioclipse
11:21 shk3 is there a wiki page for this?
11:21 jonalv Gpox?
11:21 jonalv shk3: no not yet... :/
11:22 jonalv shk3: there is only one proof of concept implementiation for this approach so far I think
11:22 jonalv and I think Gpox++ did it so I was hoping he could instruct you...
11:22 shk3 ok
11:22 shk3 perhaps you do the page instead of telling me?
11:22 shk3 would be more economical, I suppose
11:23 shk3 tell then I leave this bug open
11:23 shk3 ?
11:23 jonalv shk3: I guess that works...
11:23 jonalv shk3: could you file me a bug about that wiki text please?
11:24 shk3 sure
11:24 shk3 sorry for troubling you
11:24 jonalv perfect and make your bug depend on my one :)
11:25 jonalv I think I must talk a bit to Gpox before I can write that text but it's probably me or Gpox that should write it...
11:28 CIA-73 bioclipse: shk3 * r9785 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.business/src/net/bioclipse/cdk/business/ (CDKManager.java ICDKManager.java): removed monitor stuff
11:30 CIA-73 bioclipse: shk3 * r9786 /bioclipse2/trunk/plugins/net.bioclipse.cdk.bus​iness.test/src/net/bioclipse/cdk/business/test/ (CDKManagerPluginTest.java CDKManagerTest.java): removed monitor stuff
11:30 CIA-73 bioclipse: shk3 * r9787 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui.sdf​editor/src/net/bioclipse/cdk/ui/sdfeditor/editor/ (JCPCellPainter.java MoleculesEditorLabelProvider.java): removed monitor stuff
11:39 jonalv what pluigns has all deps needed for running Bioclipse?
11:39 jonalv |what plugin?
11:41 jonalv olass: help me pleasea
11:41 olass shk3: cdk.ui does not compile for me due to changes to generate2D
11:42 olass shk3: are you working on this?
11:48 olass shk3: are you there?
11:50 olass could someone else update and see if I'm correct that cdk.ui does not compile?
11:50 olass egonw?
11:51 olass shk3: now I have updated and still compile error
11:52 olass shk3: please do not commit code that does not compile
11:52 olass now I cannot run the latest Bioclipse
11:53 egonw olass: updating...
11:53 olass thanks
11:55 jonalv olass, egonw: Did we agree on something regarding cdk10? It's not compiling for the moment... :(
11:56 egonw jonalv: file a bug report against me, I'll maintain it
11:56 egonw BTW, what did it make broken?
11:56 egonw I'll check...
11:56 jonalv egonw: sry?
11:57 masak in standard English we usually just say "what did it break?" :P
11:57 * masak ducks
11:58 egonw masak: file a patch
11:58 masak :)
11:58 masak I'm only lurking on this channel right now so I can jump out and be difficult when the occasion arises.
11:59 jonalv masak: nice to meet you too
11:59 * masak bows
12:00 masak egonw: http://gist.github.com/95140
12:01 masak hm, s/difficult/a jerk/
12:02 egonw masak: http://delicious.com/egonw/english+funny
12:02 zarah egonw's link is also http://tinyurl.com/dzbfcw
12:02 masak :D
12:03 masak egonw: did you write 'Egon improving is its English' on purpose?
12:03 masak just checking.
12:03 egonw what made you think that?!?!
12:03 olass egonw: does cdk.ui compile for you after update?
12:03 egonw olass: hang on
12:03 * olass is just waiting
12:04 egonw sure, as if there is not enough documentation to be written
12:04 egonw cdk.ui is confirmed broken
12:04 egonw trice
12:07 olass @tell shk3 Please don't commit code that does not compile. cdk.ui is now broken on three occasions
12:07 zarah Consider it noted.
12:07 olass this is vary annoying
12:07 egonw masak: you have two funny jokes around?
12:08 masak egonw: "two funny jokes walk into a bar..."
12:08 olass egonw: please advice what to do with cdk.ui
12:08 olass I need it to compile *now*
12:08 egonw revert the patch that broke it?
12:08 olass which one is that?
12:08 egonw but this will likely only upset shk3 more
12:08 shk3 olass: working on that
12:08 zarah shk3: You have new messages. Write '/msg zarah @messages' to read them.
12:08 egonw shk3: thanx
12:08 shk3 one minute
12:10 shk3 committed
12:10 CIA-73 bioclipse: shk3 * r9788 /bioclipse2/trunk/plugins/net.biocli​pse.cdk.ui/src/net/bioclipse/cdk/ui/ (2 files in 2 dirs): adopted to changes in cdkmanager
12:15 olass shk3: please next time commit these changes in a single commit
12:16 shk3 sorry, was thinking about more changes in the handlers, but will wait till the job thing is ready
12:17 jonalv shk3: wasn't the second commit to undo the first commit? In that case I can sort of see why they wheren't as one commit... :)
12:17 olass lol
12:17 jonalv shk3: However perhaps a revert would have been b=more suiting in that case?
12:17 olass well, I just want trunk to compile at all times
12:17 shk3 wasn't a revert
12:17 shk3 yes, I want to
12:17 shk3 as well
12:17 shk3 shouldn't have done that
12:18 * jonalv feels that he has dug himself a nice big hole in the ground with all this fancy "New World Order" stuff and now stands at the bottom of it trying to figure out how he ended up there and how to get up again... :/
12:18 masak like in Matrix III!
12:37 jonalv masak: I ahve an array and need to call a method wth the ... signature. How do I do that?
12:37 masak is this pure Java we're talking about?
12:37 jonalv masak: yes
12:37 jonalv masak: java reflection that is
12:37 masak please clarify 'call a method with the signature'. not clear what you mean.
12:38 jonalv or no java
12:38 jonalv the method is invoe( Object o, Object... args )
12:38 jonalv I have the o object and an array of arguments I want to use for args
12:38 jonalv how do I do that?
12:40 masak what's the problem?
12:40 jonalv those are ...
12:40 jonalv I have an array
12:40 masak that syntax looks OK to me.
12:40 masak I'm sorry, I'm a bit distracted... :/
12:40 CIA-73 bioclipse: shk3 * r9789 /bioclipse2/trunk/cdk-externals/trunk/ (2 files in 2 dirs): better exception handling in model builder needed for fixing bug #860
12:40 zarah bug #860 | http://tinyurl.com/cnr6v2
12:40 masak I'm working with other things.
12:40 jonalv masak: so you don't know?
12:41 masak jonalv: I don't understand your problem.
12:41 CIA-73 bioclipse: shk3 * r9790 /bioclipse2/trunk/plugins/net.bioclip​se.cdk.ui/src/net/bioclipse/cdk/ui/ha​ndlers/Create2dHandlerWithReset.java: fixes bug #860 by explicitly handling the nusuchatomtype exception
12:41 jonalv :(
12:41 zarah bug #860 | http://tinyurl.com/cnr6v2
12:41 masak jonalv: may I recommend some of the nice books on Java? there's plenty.
12:41 masak the JLS might actually have the answer you need.
12:41 masak it's available online.
12:41 jonalv masak: I think I will try around a little. Trial and error you know...
12:42 masak aye.
12:42 masak good man.
12:42 * masak will soon disappear
12:52 maclean @tell TheInternet : the PDBReader in the CDK really sucks.
12:53 zarah Consider it noted.
12:53 maclean heh. glad that masak missed that.
12:53 jonalv yea...
12:53 maclean in my defence : I have a cold.
12:54 jonalv Things are sooo messy!
12:54 jonalv :(
12:55 maclean design-wise, or syntax-wise?
12:57 jonalv all-wise
12:57 maclean :(
12:58 maclean well, if it was easy, it would be tidy. it would also not be worth attempting...
12:58 maclean (or it would already have been done)
12:59 jonalv probably...
12:59 * jonalv needs water
12:59 jonalv brb
13:41 jonalv egonw: http://java.sun.com/j2se/1.5.0/d​ocs/guide/language/varargs.html it seems like it should work strangely enough...
13:41 zarah jonalv's link is also http://tinyurl.com/35zxnj
13:43 maclean so, you can say "Object[] args = { new Integer(1), new Char('C') }; invoke(o, args);" ?
13:44 jonalv maclean: I don't know. Bujt that is what they are claiming...
13:44 jonalv it doesnät seem to work for me though...
13:46 maclean oh :(
13:47 jonalv hm I had an Object that happend to be an array seems if I defined it as Object[] it worked better...
13:47 jonalv egonw: that solved it
13:48 egonw casting?
13:48 jonalv egonw: now I am down to 48 failing test out of 138 for the cdk plugin test
13:48 egonw cool
13:48 jonalv egonw: basically yes but \I changed the return type of the transformArgs ethod which was Object into Object[] (ooops)
13:49 jonalv or 43 errors and 5 failures actually
13:50 olass shk3: sorry, but I reopened bug 838 again
13:50 olass shk3: it is actually very important for my for the workshoip
13:50 jonalv egonw: I am gonna commit now. It's a least a bit better I think...
13:50 olass since I am to demo the scripting then
13:51 olass shk3: let me know
13:51 shk3 what's the problem now?
13:51 olass well..
13:51 olass you removed the old way
13:51 olass please see the bug comments
13:51 olass I did not want you to remove the generate3D for single mols
13:51 olass (which you did)
13:51 olass I want both single and multiple mols
13:52 shk3 so List<IMolecule> is the current method to use?
13:52 CIA-73 bioclipse: jonalv * r9791 /bioclipse2/trunk/plugins/ (3 files in 2 dirs): renamed CDKManagerPluginTest -> AbstractCDKManagerPluginTest
13:52 olass well, it has never been something else in Bioclipse2
13:52 olass however
13:52 olass if you strongly feel for IMolecule[] then it is your decision
13:52 CIA-73 bioclipse: jonalv * r9792 /bioclipse2/trunk/plugins/net.bioclipse.​cdk.business.test/src/net/bioclipse/cdk/ (4 files in 2 dirs): Seperated CDKMAnagerPluginTest into two test classes one for each manager flavour. Also added a test suite runnign all of them.
13:53 shk3 no problem with List<IMolecule>
13:53 CIA-73 bioclipse: jonalv * r9793 /bioclipse2/trunk/plugins/net.bioclipse.ui/sr​c/net/bioclipse/ui/jobs/CreateJobAdvice.java: translate from String to IFile even if not running as job...
13:53 olass shk3: great
13:53 shk3 I am just afraid it will be something else tomorrow
13:53 olass jonalv: the commit flooder
13:53 shk3 a third thing the day after tomorrow
13:53 olass shk3: :)
13:53 shk3 and IMolecule[] again next week
13:53 olass let's hope not
13:54 olass shk3: no, we've always used List
13:54 olass for good and for bad
13:54 * olass is struggling with Lists in Javascript
13:56 jonalv egonw: I get a lot of "this method should not be called and will look into these but also "File already exists: when I run the cdk plugin tests now. Do you have any ideas about that?
13:57 egonw no, those are all new
14:15 CIA-73 bioclipse: ospjuth * r9794 /bioclipse2/trunk/plugins/net.bioclips​e.data/data/Javascripts/qsar-multi.js: Added script to calculate list of descriptors for list of molecules, without parameters. Also available as gist 95194.
14:43 shk3 joined #bioclipse
15:15 CIA-73 bioclipse: shk3 * r9795 /bioclipse2/trunk/plugins/ (4 files in 2 dirs): more versions of the coordinate generation
15:15 CIA-73 bioclipse: shk3 * r9796 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui.sdf​editor/src/net/bioclipse/cdk/ui/sdfeditor/editor/ (JCPCellPainter.java MoleculesEditorLabelProvider.java): adopted this again to generatecoordinates
15:17 CIA-73 bioclipse: shk3 * r9797 /bioclipse2/trunk/plugins/net.biocli​pse.cdk.ui/src/net/bioclipse/cdk/ui/ (2 files in 2 dirs): adopted this again to generatecoordinates
15:48 egonw joined #bioclipse
16:05 jonalv shk3: sry, what it this?
16:05 jonalv |what is "this"?
16:05 jonalv (gah I can't spell!)
16:06 shk3 "this" is the plugin
16:06 jonalv oh
16:07 jonalv can someone point me to a strnage manager method? A one of those, "Heck I didn't know you could do that" kind of mehtod.
16:07 jonalv I am trying to figure out how to solve the current cdk manager (and all managers) crisis... :)
16:15 maclean not...really.
16:16 maclean what are the features/properties that this method should have - it should be surprising?
16:16 jonalv well, I have one use case where you are working with an IFIle and getting back an IBioObject
16:17 jonalv another where you are working with an IBIoObject and get back an IBIoObject
16:17 jonalv what else are there_
16:17 jonalv ?
16:17 maclean Ahh. So you want to know the diversity of the methods. hmmm.
16:18 jonalv the one working with an IFIle osrt of covers string as ell..
16:18 maclean well, I guess that there are methods that take no parameters, and return none - like jcp.updateView().
16:19 maclean and (recently) jcp.snapshot()...
16:20 maclean oh, no wait. not that one, sorry.
16:20 maclean that takes an IFile/String.
16:21 jonalv more spcifically I am intrested in long running methods...
16:21 jonalv methods that should be run in jobs
16:22 maclean well, at some point, there might be a structure generation manager, with methods that are very long running.
16:22 maclean Not sure what they would consume and produce.
16:23 maclean Probably best not to have them as scriptable methods I suppose. Hmmm.
16:25 jonalv hmm indeed :)
16:25 jonalv well I am calling this a day now
16:52 ilogger2 joined #bioclipse
17:00 ilogger2 joined #bioclipse
17:24 ilogger2 joined #bioclipse
17:50 CIA-73 bioclipse: goglepox * r9798 /bioclipse2/trunk/plugins/net.bioclipse.​cdk.ui.sdfeditor/src/net/bioclipse/cdk/u​i/sdfeditor/editor/MoleculesEditor.java:
17:50 CIA-73 bioclipse: Added check for biolist to open with ui.open(...)
17:50 CIA-73 bioclipse: Fixes bug #869
17:50 zarah bug #869 | http://tinyurl.com/d8y258
17:51 rajarshi joined #bioclipse
18:17 olass joined #bioclipse
18:19 egonw hi olass
18:19 olass hi
18:19 zarah hello olass, you fantastic person you
18:19 olass @botsnack
18:19 zarah ;)
19:15 CIA-73 bioclipse: egonw * r9799 /bioclipse2/trunk/plugins/net.bioclips​e.cdk.business.test/src/net/bioclipse/​cdk/business/test/CDKManagerTest.java: Added test for detecting .mdl file formats as MDL V2000 molfile
19:16 CIA-73 bioclipse: egonw * r9800 /bioclipse2/trunk/plugins/net.bioc​lipse.gist/src/net/bioclipse/gist/ (3 files in 2 dirs): Renamed method from load() to download() per #890
21:25 olass joined #bioclipse

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