Camelia, the Perl 6 bug

IRC log for #bioclipse, 2009-08-12

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

All times shown according to UTC.

Time Nick Message
00:49 miguelrojasch joined #bioclipse
06:34 Gpox joined #bioclipse
06:34 Gpoks joined #bioclipse
06:35 Gpox joined #bioclipse
06:46 masak joined #bioclipse
06:59 egonw joined #bioclipse
07:05 egonw moin
07:05 zarah ni hao egonw
07:06 mgerlich hi
07:06 zarah hello mgerlich, you fantastic person you
07:06 masak :)
07:14 egonw nice temperature outside
07:14 egonw there is oxygen again
07:14 * egonw likes Sweden
07:14 masak yes, we discovered oxygen!
07:15 masak before that, people had a very unprofessional approach to breathing.
07:16 egonw oh, you mean the swedes 'invented' oxygen?
07:16 masak :)
07:17 egonw wow... news today:
07:17 egonw "Microsoft cannot sell Word, the judge ruled"
07:17 egonw apparently, for real!
07:17 egonw http://blog.seattlepi.com/mi​crosoft/archives/176223.asp
07:17 zarah egonw's link is also http://tinyurl.com/npmmlc
07:18 masak weird.
07:36 egonw and Nokia signed a deal to provide Office on their phones
07:36 egonw why?!?! They have koffice just around the corner, nicely build in Qt !?!?!
07:37 masak I think I would be more enthusiastic about this if I were a consumer of Office-type products.
07:38 masak oh, actually, I've been using Calc for budget-making lately.
07:38 masak I have to admit that it's difficult to create a script with that kind of immediate flexibility.
07:42 egonw ah, shit
07:52 jonalv joined #bioclipse
07:53 jonalv egonw: joelib now?
07:54 * egonw does not understand the question
07:55 * egonw sees multiple interpretations, and does not know how to answer
07:55 * masak does a "joelib now" dance!
08:00 jonalv egonw: wasnät that the library that had everything that cdk had but not all that cdk had_ Or was that some other library_
08:00 egonw jonalv: yes
08:01 egonw it's just to make a point
08:01 jonalv egonw: ah
08:01 egonw "bioclipse stuff is not restricted to the CDK"
08:01 jonalv good point ;)
08:02 * jonalv had a throurough look at the Mychem handbook and wan't to build a StructureDB implementation based on that som time in the future...
08:04 egonw then better make sure to have a manager API with no references to the backend
08:05 egonw and make the data model used hy hibernate hidden
08:05 jonalv egonw: sry?
08:05 jonalv egonw: you lost me
08:05 egonw well, what we talked about
08:05 jonalv egonw: why?
08:05 egonw why the current RDF manager is not really working
08:05 egonw that the manager is relatively clean, except ...
08:06 egonw that it returns the data model oriented at the backend implementation
08:06 egonw that is ...
08:06 egonw the object that is serialized
08:06 egonw into the db
08:06 egonw DBMolecule that is
08:06 egonw DBMolecule should have zero to do with the DB backend used
08:06 jonalv egonw: uhm what are you trying to say?
08:07 jonalv ah
08:07 jonalv egonw: no
08:07 jonalv egonw: that is not the way to do things with Hibernate
08:07 jonalv egonw: but why do you want me to use Hibernate?
08:07 * jonalv don't like Hibernate
08:07 egonw ibatis
08:07 egonw sorry
08:07 jonalv egonw: ah same thing with iBatis
08:08 egonw and I don't want to see you use that ...
08:08 jonalv although I have heard of some other pattern...
08:08 jonalv but I can't remember what it was called and I only heard someone mention it...
08:08 egonw I'd rather see you go off to the EBI for a couple of months (6) and make structuredb interoperate with chemisql
08:08 egonw (hence MyChem, OrChem and PgChem)
08:09 * masak isn't sure he believes in ORM
08:09 egonw it's too slithering...
08:09 * jonalv isn't sure he believeves in ORM either but since that is the best he knows about...
08:09 masak egonw: your Swedish is improvind, I notice. :)
08:10 masak s/nd/ng/
08:10 jonalv egonw: I think the chemisql standard is so vague that just about just working with it...
08:10 jonalv egonw: but then again I ahven't really seen it just the Mychem one..
08:11 jonalv egonw: is there a chemisql standard out there btw?
08:11 egonw being worked on...
08:12 egonw ask jpansanel or steinbeck
08:14 * egonw spends an hour setting up a new manager
08:14 * egonw really wants the Bioclipse SDK plugin with "File -> New -> Bioclipse Manager Project"
08:15 jonalv egonw: yea that would be nice. We should organize a nice hacksession and build that... :)
08:16 egonw what about tomorrow ?
08:16 egonw we/I have a student soon
08:16 jonalv egonw: hackathon tomorrow?
08:16 egonw and the GROMACS people at BMC are gonna write a manager too
08:16 egonw jonalv: yes
08:16 jonalv egonw: hm I guess that is a possibility
08:17 egonw a Bioclipse SDK hackaton
08:17 jonalv I wonder how hard it can be..
08:17 egonw right
08:17 jonalv :)
08:17 egonw there should be enough code around to show how to write a New Project wizard
08:17 egonw we just need quite a few wizard pages
08:17 egonw with many setting...
08:17 jonalv egonw: yea...
08:18 egonw Author, his email, license, ...
08:18 egonw manager name
08:18 egonw plugin name
08:18 jonalv I have SWT-designer now so making wizards is like piece of cake
08:18 jonalv the problem is the models...
08:18 jonalv and autogeneration of code
08:18 jonalv and stuff
08:19 egonw which we all now do by hand
08:19 jonalv true
08:21 jonalv egonw: I would love some feedback on StructureDB if you have time. Primarily the UI. I know what you think about the architectural consecvences of how I am using the ORM... ;)
08:21 jonalv @pz jmol view
08:22 zarah [1333] 70 source files without copyri... http://tinyurl.com/nz9aff
08:22 zarah [688] Bring back the JmolView.           http://tinyurl.com/lbhwok
08:22 zarah [874] Jmol should show static propert... http://tinyurl.com/klzbsr
08:22 zarah [893] Jmol should show properties in ... http://tinyurl.com/mq9k88
08:22 zarah [686] Set help context ID on parts       http://tinyurl.com/kskpu2
08:22 zarah Entire list (13) at http://tinyurl.com/p539lu
08:22 egonw jonalv: please schedule a demo next Tuesday
08:22 egonw Ola should be back too then
08:22 egonw ah no
08:22 egonw not Tuesday
08:22 egonw next Thursday
08:22 jonalv egonw: as in not tomorrow I guess?
08:23 egonw right
08:23 egonw next, next Thu
08:23 egonw :)
08:23 jonalv :)
08:27 egonw 20 Aug
08:33 jonalv egonw: what is the ETA for bug 688 btw?
08:33 egonw @pz 688
08:33 zarah http://pele.farmbio.uu.se/cgi-b​in/bugzilla/show_bug.cgi?id=688
08:33 egonw dunno
08:33 egonw post august
08:34 egonw pre 2.2.0
08:34 jonalv it feels like one of the more important bugs we have left...
08:34 jonalv but I could be wrong...
08:35 egonw well... the *really* important bugs are not in pelezilla
08:35 egonw paper X is not written
08:35 egonw paper Y is not written
08:35 egonw paper Z is not written
08:35 jonalv I guess
08:36 CIA-63 bioclipse.joelib: Egon Willighagen master * r341d7bc / (16 files in 6 dirs): Added a (empty) JOELib manager - http://bit.ly/afiRY
08:39 jonalv egonw: When I am performing a SMARTS query and the AllringsFinder times out. How would you recommend me to handle that?
08:39 egonw mark it with an Error
08:39 jonalv egonw: I don't understand
09:18 jonalv joined #bioclipse
09:28 egonw masak: check my last tweet
09:28 egonw rest can try:
09:28 egonw http://friendfeed.com/egonw/68447ee0/mmm​-that-email-to-taverna-ml-is-interesting
09:28 zarah egonw's link is also http://tinyurl.com/qx3a3j
09:29 * egonw forsees a nice microblog manager hack in a nearby spot of free time
09:29 masak egonw: you're like those people who come around asking "did you get my latest email?" :P
09:30 egonw yeah, we really need a good bot here to mirror RSS feeds...
09:30 egonw leander is not working :(
09:30 masak :/
09:30 egonw crappy shit of code I wrote
09:30 masak @gg we hates software
09:30 zarah masak: http://we.hates-software.com/
09:30 egonw it does not properly remember what items it already did...
09:31 egonw and the cache I hacked up is not working either :(
09:31 egonw there is some bug, but no time to debug it :(
09:31 masak egonw: I'm sure there are CPAN modules out there which could assist you.
09:31 egonw yeah, maybe zarah could be extended
09:32 masak egonw: yah. problem with that is I don't grok POE.
09:32 egonw Poetry of Existence ?
09:34 egonw Poor old Equation?
09:35 masak egonw: http://search.cpan.org/dist/POE/lib/POE.pm
09:35 zarah masak's link is also http://tinyurl.com/pzrl85
09:37 egonw tack
09:38 egonw jonalv: @wbp.parser ?
09:39 jonalv egonw: yes it's for my gui builder tool... I can understand if you don't want it there...
09:39 egonw jonalv: can you please blog/email/foo about those two new dialogs, with screenshots ?
09:40 egonw jonalv: well, some info in the commit message would have been appreciated
09:40 jonalv egonw: yea I didn't think of that... :)
09:40 egonw "Added @wtf annotation to allow Foo Bar with http://..."
09:40 jonalv egonw: it's actually quite ugly...
09:41 jonalv but it needs that annotation to work when there are multiple constructors. It doesn't know which one to choose otherwise....
09:42 * masak could use a @WTF annotation sometimes...
09:43 egonw jonalv: no, it's fine to have it there...
09:43 jonalv egonw: okey good
09:49 CIA-63 bioclipse.cheminformatics: jonalv master * r9112825 / plugins/net.bioclipse.chemoinformatics/src/net/bi​oclipse/chemoinformatics/PickMoleculeDialog.java :
09:49 CIA-63 bioclipse.cheminformatics: Made PickMoleculeDialog truly resizable thanks to some inspiration from autogenerated code made by (RCP-developer)++
09:49 CIA-63 bioclipse.cheminformatics: Signed-off-by: Egon Willighagen <egon.willighagen@gmail.com> - http://bit.ly/NkKWy
10:23 CIA-63 bioclipse.cheminformatics: jonalv master * r1e09a35 / (2 files in 2 dirs):
10:23 CIA-63 bioclipse.cheminformatics: moved PickMoleculeDialog into a dialogs package
10:23 CIA-63 bioclipse.cheminformatics: Signed-off-by: Egon Willighagen <egon.willighagen@gmail.com> - http://bit.ly/UzDBG
10:23 CIA-63 bioclipse.cheminformatics: jonalv master * r248876a / plugins/net.bioclipse.chemoinf​ormatics/META-INF/MANIFEST.MF :
10:23 CIA-63 bioclipse.cheminformatics: export the net.bioclipse.chemoinformatics.dialogs package
10:23 CIA-63 bioclipse.cheminformatics: Signed-off-by: Egon Willighagen <egon.willighagen@gmail.com> - http://bit.ly/XCkoO
10:23 CIA-63 bioclipse.cheminformatics: jonalv master * r9b33cc9 / plugins/net.bioclipse.chemoinformati​cs/src/net/bioclipse/chemoinformatic​s/dialogs/PickMoleculesDialog.java :
10:23 CIA-63 bioclipse.cheminformatics: Created dialog for picking many molecules.
10:23 CIA-63 bioclipse.cheminformatics: A dialog for picking files containing one or many molecules from the workspace.
10:23 CIA-63 bioclipse.cheminformatics: Signed-off-by: Egon Willighagen <egon.willighagen@gmail.com> - http://bit.ly/2JxI1T
10:51 egonw arghh....
10:52 egonw ok, lunch time
11:03 shk3 joined #bioclipse
11:26 egonw I can't the bloody plugin to see the packages exported in net.sourceforge.joelib :(
12:11 egonw slowly making progress with the JOELib descriptor calculation
12:16 sneumann joined #bioclipse
12:46 CIA-63 bioclipse.joelib: Egon Willighagen master * r37a1535 / (4 files in 3 dirs): Fiddled to get it finally export the packages to be used in other plugins. - http://bit.ly/3V8igO
13:11 * egonw is puzzled why his extension point is now showing up...
13:17 egonw ummm
13:17 * egonw is puzzled why his extension is noT showing up...
13:32 egonw how can I debug why my extension is not found??
13:32 egonw @gg how can I debug why my extension is not found??
13:32 zarah egonw: http://www.nabble.com/looking-for-a-symboli​c-debugger-for-FF-extension-td15236611.html
13:33 egonw @gg how can I debug why my eclipse extension is not found??
13:33 zarah egonw: http://benramsey.com/archives/zend-​executable-debugger-eclipse-plugin/
13:35 egonw of course, no errors what so ever
13:35 egonw perhaps missing classes... who knows...
13:36 jonalv @pz bioclipse planet google reader
13:36 zarah jonalv: I couldn't find anything on 'bioclipse planet google reader' in the database.
13:36 jonalv @pz bioclipse planet google
13:36 zarah jonalv: I couldn't find anything on 'bioclipse planet google ' in the database.
13:36 jonalv @pz bioclipse planet
13:36 zarah jonalv: I couldn't find anything on 'bioclipse planet' in the database.
13:36 jonalv @pz planet
13:36 zarah jonalv: I couldn't find anything on 'planet' in the database.
13:36 egonw @pz bioclipse
13:36 zarah [761] JsEditor edits a file backend, ... http://tinyurl.com/np3zzr
13:36 zarah [1480] moltable cannot handle propert... http://tinyurl.com/p4xokk
13:36 zarah [1440] Update BODO web page to be cle... http://tinyurl.com/pspo8f
13:36 zarah [1445] XMPP service to calculate CDK ... http://tinyurl.com/ojb6yx
13:36 zarah [1017] XMPP commands to list services... http://tinyurl.com/ncqmz7
13:37 zarah Entire list (426) at http://tinyurl.com/mlmbmr
13:37 egonw mmm...
13:37 egonw @pz rss
13:37 zarah egonw: I couldn't find anything on 'rss' in the database.
13:38 egonw argghh.... !!!!
13:38 egonw stupid eclipse
13:40 egonw it's difficult to stay calm now...
13:40 jonalv egonw: what's up?
13:40 masak @pz rss
13:40 zarah masak: I couldn't find anything on 'rss' in the database.
13:40 masak @pz rss
13:40 zarah masak: I couldn't find anything on 'rss' in the database.
13:41 egonw I wrote an extension for net.bioclipse.qsar.descriptorProvider
13:41 egonw but it is not loaded...
13:41 egonw and no messages about missing plugins, or what ever...
13:41 jonalv egonw: oh I hate that...
13:41 egonw no ClassNotFound... nothing
13:42 jonalv egonw: and you _are_ ofcourse actually running the plugin?
13:42 egonw I hope so
13:42 egonw the feature is in the joelib.product
13:42 egonw and the plugin in the feature
13:42 jonalv egonw: add a Sysout to hte start method of the plugins Activator
13:42 jonalv egonw: then at least you know it is running...
13:42 egonw did that
13:43 egonw but does indeed does not show up
13:43 jonalv egonw: and it is running...
13:43 jonalv o
13:43 jonalv so it is not running then
13:43 egonw but why ?!?!?
13:43 jonalv egonw: well does anything make it start?
13:43 egonw I double checked it to be in the feature, and the feature in the .product
13:44 egonw I was hoping the extension did that...
13:44 * jonalv is nto so sure about how that works with Eclipse plugins but OSGI bondles have to be started by something...
13:46 egonw it does not show up in the list of plugins in About Bioclipse Plug-ins either...
13:47 jonalv egonw: start Bioclipse with flag -console and write ss in the console
13:49 egonw oi, that's a long list...
13:49 egonw no, not in that list either...
13:49 egonw what is going on?!?!
13:49 egonw what stupid mistake am I making ?!?!
13:54 jonalv egonw: is it not even listed in that list?
13:54 jonalv egonw: then you are probably not runnign your plugin
13:55 egonw but why not?
13:55 egonw I checked for the 100th time now it is listed in the featue.... :(
13:57 egonw jonalv: http://gist.github.com/166518
13:57 egonw do you spot anything weird there?
13:58 egonw the extension is on net.bioclipse.joelib.qsar
13:58 jonalv egonw: no I am not used to reading feature xml... :(
14:04 masak people: when you write a comment, don't put into the comment what the code _does_. I can see what the code does. I am capable of reading code. the code already tells me what the code does. there's no need to repeat in the comment what the code does. I already know what the code does. thank you.
14:06 egonw masak: I disagree
14:06 egonw it's prob in the details..
14:06 egonw but comments must describe what the code is supposed to do
14:07 egonw what else must the do?
14:07 egonw comment on the weather
14:07 egonw I know you like rain...
14:07 egonw but don't overdo it :)
14:07 masak I think comments should add something to the code itself.
14:07 egonw yes ...
14:07 masak preferably by explaining something, by saying things that are not in the code.
14:08 masak I instinctively delete comments that aren't more informative than the surrounding code itself.
14:08 jonalv egonw: I think commetns should comment on what the code is doing not repeating what the code already states that it is doing.
14:08 masak jonalv: yes, something like that.
14:08 masak people tend to say that comments should say _why_ and not _what_.
14:08 egonw then please rephrase your statement
14:08 masak but maybe that's too narrow.
14:09 egonw it should not describe what the code is doing, or is should...
14:09 egonw please state the difference
14:09 jonalv masak: I think a before and after example would be educative...
14:09 masak egonw: ok. I find comments on the form /* add one to i */ utterly useless.
14:09 egonw masak: ah, right... why not what
14:09 egonw that's more clear already...
14:09 egonw go on
14:09 masak egonw: but there are many variants of this.
14:09 egonw what about summarizing ?
14:10 masak egonw: for example /* initialize the object */
14:10 egonw // now find all instances
14:10 masak egonw: it depends. usually that indicates that you can create a method with a good name instead.
14:10 egonw ah, masak ++
14:10 egonw mistake number 1
14:10 * jonalv sometimes add commetns like that into _BIG_ files in order to make it easier to find his way through them when browing. masak: is he evil when he does that?
14:10 egonw reason why we have so many code blobs fail PMD
14:10 egonw jonalv: yes you are
14:11 egonw typical result of not using enough methods to split up what you are doing
14:11 masak jonalv: I've learnt to accept that you do that. I don't, myself, but we have different ways of browsing files.
14:11 egonw good code should fit on one computer screen
14:11 masak and that's ok.
14:11 masak egonw: I think I used to believe that, but I'm not sure I do anymore.
14:12 jonalv egonw: no not for splitting up big methods but for handling big files
14:12 masak I can think of cases where a well-refactored method is 200 lines long.
14:13 egonw jonalv: /me thinks Google Reader forgot to update their cache
14:14 jonalv egonw: oh ios there anything we can do about that if so?
14:14 egonw not that I am aware of :(
14:14 * jonalv removed his Bioclispe planet and wants it back...
14:14 egonw the HTML really and clearly points to:
14:14 egonw http://pele.farmbio.uu.se/planetbioclipse/atom.xml
14:14 zarah egonw's link is also http://tinyurl.com/m2akof
14:14 egonw but google reader still reports:
14:14 egonw http://pele.farmbio.uu.se/atom.xml
14:14 zarah egonw's link is also http://tinyurl.com/n4m6y5
14:14 egonw (google reader)--
14:15 egonw jonalv: try adding the URL in 'Add a subscription':
14:15 egonw http://pele.farmbio.uu.se/planetbioclipse/atom.xml
14:15 zarah egonw's link is also http://tinyurl.com/m2akof
14:15 jonalv egonw: besides what posts go to bioclipse.planet?
14:15 jonalv egonw: yea that worked
14:21 egonw and why it is loading plugins that I do *not* have in the feature???
14:22 jonalv egonw: sounds very strange
14:22 egonw yeah, I'm quite pissed...
14:23 * jonalv has evil merge conflicts instead
14:23 CIA-63 bioclipse.core: jonalv master * rc10fa8f / plugins/net.bioclipse.core/src/net/bi​oclipse/core/util/TimeCalculater.java :
14:23 CIA-63 bioclipse.core: Added generate time-remaining-String method.
14:23 CIA-63 bioclipse.core: To be used in long runnign jobs for generating a time remaining estimation to show in the text field for a monitor using subTask method. - http://bit.ly/12UCLy
14:23 CIA-63 bioclipse.core: jonalv master * r4ce690e / plugins/net.bioclipse.core/src/net/bio​clipse/core/domain/RecordableList.java :
14:23 CIA-63 bioclipse.core: Thread safeing.
14:23 CIA-63 bioclipse.core: I ran into ConcurrentModification exceptions when running multiple threads. This should fix those problems. - http://bit.ly/8beJ
14:23 CIA-63 bioclipse.structuredb: jonalv master * r5a3b411 / plugins/net.bioclipse.structuredb/src/net/biocli​pse/structuredb/business/StructuredbManager.java :
14:23 CIA-63 bioclipse.structuredb: Use TimeCalculater.generateTimeRemainEst instead.
14:23 CIA-63 bioclipse.structuredb: Changed to using the new method doing all of it instead. - http://bit.ly/ZzeBu
14:23 CIA-63 bioclipse.structuredb: jonalv master * rd35b6f7 / (4 files in 3 dirs):
14:23 jonalv that's strange as well becasue when I loo at then the files to be merged are equal...
14:23 CIA-63 bioclipse.structuredb: Made SMARTS querying a bit nicer.
14:23 CIA-63 bioclipse.structuredb: Added time left estimation and care taking of CDKException because of AllRingFinder timeouts. - http://bit.ly/JnyVd
14:38 jonalv hm the actual SMARTS matching only take about 25% of the time when doing a SMART query. The rest is in retrieving the molecules...
14:39 jonalv With Mychem I think we could get structureDB quite fast actually...
14:43 egonw and now it works...
14:43 egonw finally
14:43 egonw headache...
14:43 jonalv egonw: what did you do wrong?
14:43 egonw no idea
14:43 jonalv o
14:44 egonw I know made the joelib feature include the platform too...
14:44 egonw and everything else
14:44 egonw so, I guess, it's missing a dependency somewhere...
14:44 jonalv you might want to try to export that sooner than later
14:44 * jonalv is afraid that there might be more trouble around...
14:45 egonw yeah...
14:45 egonw that's what we have olas for ...
14:45 jonalv hehe
15:03 micky joined #bioclipse
15:04 micka joined #bioclipse
15:07 CIA-63 bioclipse.joelib: Egon Willighagen master * r00e5e4d / (9 files in 6 dirs): Added a plugin extension the QSAR descriptor calculate extension point with JOELib descriptors - http://bit.ly/eeEs8
15:07 CIA-63 bioclipse.joelib: Egon Willighagen master * r335460e / (2 files): Added the JOELib-QSAR plugin to the feature - http://bit.ly/13iHHx
15:10 CIA-63 bioclipse.joelib: Egon Willighagen master * rde351ec / features/net.bioclipse.joe​lib_feature/joelib.product : Added features to product too - http://bit.ly/DQ3xV
15:19 micky joined #bioclipse
15:28 micky joined #bioclipse
15:38 micky joined #bioclipse
16:23 egonw joined #bioclipse
16:24 micky joined #bioclipse
17:05 egonw joined #bioclipse
17:23 shk3 joined #bioclipse
17:59 sneumann joined #bioclipse
18:41 egonw joined #bioclipse
19:04 egonw joined #bioclipse
19:56 shk3 joined #bioclipse
20:04 sneumann joined #bioclipse
21:11 egonw joined #bioclipse

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