Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-04-28

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

All times shown according to UTC.

Time Nick Message
05:03 egonw joined #bioclipse
05:58 CIA-44 bioclipse: egonw * r5496 /bioclipse2/trunk/plugins/net.bioclipse.cdk/ (.classpath META-INF/MANIFEST.MF): Synched dir name (jars -> jar), to make life easier with a git checkout, where svn:imports is not automatically used, and I have to do a manual svn co
05:58 CIA-44 bioclipse: egonw * r5497 /bioclipse2/trunk/plugins/net.bio​clipse.core/META-INF/MANIFEST.MF: Made plugin 'singleton', which is required by Eclipse3.4 (please email if this gives trouble in 3.3)
05:58 CIA-44 bioclipse: egonw * r5498 /bioclipse2/trunk/plugins/net​.bioclipse.cdk10/.classpath: Fixed CP problems
05:58 CIA-44 bioclipse: egonw * r5499 /bioclipse2/trunk/plugins/net​.bioclipse.cdk10/.classpath: Reordering.
05:59 CIA-44 bioclipse: egonw * r5500 /bioclipse2/trunk/plugins/net.bioclipse.cd​k10.jchempaint/src/net/bioclipse/cdk10/jch​empaint/ui/editor/MDLMolfileEditor.java: Added class JavaDoc
06:50 olas joined #bioclipse
06:50 olas hi
06:50 egonw hi olas
06:51 * egonw is checking out bioclipse/trunk
06:51 olas egonw: work or holiday today?
06:51 egonw to allow copy/paste of code...
06:51 egonw work
06:51 olas aah
06:51 * olas too
06:51 egonw with an hour to travel, and then work from my father in laws place
06:51 olas great
06:52 egonw I got BC2 running this morning...
06:52 egonw solved that class load problem
06:54 olas good
06:56 egonw ready to C/P from BC1 n.b.cdk.jchempaint...
07:01 egonw olas: btw -> http://chem-bla-ics.blogspot.com/2008/04/c​omments-on-rethinking-software-access.html
07:06 olas Ilike that blog
07:07 egonw hoped so :)
07:07 egonw crap :(
07:08 olas I will do some minor work on the multi-structure table and also support your work today
07:08 olas so let me know what I can do to assist
07:08 egonw a classpath clean up for cdk10 each time messes up the .classpath by removing all jars... :(
07:08 egonw yes, sure
07:08 egonw I'm in the process of copying in old BC1 JCP code
07:08 olas ok
07:08 egonw because Stefan did quite some magic to keep things in sync etc...
07:09 olas yes
07:09 olas but it's quite dirty code
07:09 egonw not sure how that will translate into proper IEditor talk..
07:09 egonw yes, I know
07:09 olas it really needs some cleanup
07:09 egonw but that's partly because the JCP code is ugly
07:09 olas but let's start
07:09 egonw the new editor will solve that...
07:09 egonw right
07:09 egonw we really need an student to work on the JCP editor...
07:10 egonw don't think I will participate in the Programmeerzomer again
07:10 egonw too crappy organization...
07:10 egonw costs me more time, than that it delivers results...
07:10 egonw well.. not entirely...
07:10 egonw but close to break even
07:11 olas yes
07:11 olas I hope Christoph can assist with a student on JCP
07:12 egonw yes, that would be nice
07:17 olas plugin singleton is required if you have a plugin.xml file in the plugin
07:18 olas at least that's what I think
07:19 egonw 3.4 was complaining about that with an error
07:21 olas so doeas 3.3
07:21 olas (I think)
07:22 CIA-44 bioclipse: egonw * r5501 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/src/net/bio​clipse/cdk10/jchempaint/ui/editor/ (4 files): Copied in some code snippets from BC1; now about to merge JCPEditor bits...
07:32 rojasm joined #bioclipse
07:34 egonw hi rojasm
07:39 jonalv joined #bioclipse
07:42 jonalv hi
07:47 egonw oi, this is one major big blob of ugly code
07:47 olas hi jonalv
07:47 * egonw is suprised if it will work at all
07:47 egonw I know remember why I did not want to work on the old JCP anymore !!!!
07:47 jonalv egonw, have you produced a blob of code?
07:47 jonalv ah the old one...
07:48 jonalv So is there a new one in the works now?
07:48 olas yes
07:48 egonw well... sort of...
07:48 olas egonw: is it that bad?
07:48 egonw I am working on it a couple of hours a month... mm... per half year
07:48 egonw check out this patch:
07:49 egonw hang on ...
07:50 CIA-44 bioclipse: egonw * r5502 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/src/net/bio​clipse/cdk10/jchempaint/ui/editor/ (4 files): OK, this should show up a JCP editor... not tested.. actually, I don't think the data->editor link is set up yet
07:50 egonw btw, there are many BC1 files apparently without any copyright info!!!
07:50 egonw which I therefore can't copy either...
07:50 egonw very bad practice...
07:51 olas if no copyright info, then no copyright
07:51 olas so please copy
07:51 egonw you wish
07:51 egonw that's not how it works...
07:51 olas is there a Jar for JAXB binding?
07:51 olas or is it in Java?
07:51 egonw ok, seems to JCP bit shows up...
07:52 egonw olas: now I could use your help, with the editor input 2 jcpmodel link...
07:52 olas hmm ok
07:52 olas place FIXME
07:52 olas and commit
07:53 olas or
07:53 olas tell me where
07:53 egonw well, that would have been *my* first question :)
07:53 egonw JCPPage.java
07:53 egonw .init()
07:53 olas does not compile now
07:53 egonw that would be the first thing to implement, I think
07:54 olas right?
07:54 egonw for me it does...
07:54 egonw what error?
07:54 egonw maybe I forgot something to add...
07:54 olas aah
07:54 olas jars gone
07:54 olas can I add them?
07:55 jonalv cdk is exporting things that aren't there
07:55 egonw olas: no
07:55 egonw use svn:imports please
07:55 olas ??
07:55 olas CDK10 external jars are not imported in cdk10
07:55 egonw right
07:55 egonw use svn:imports
07:55 olas so?
07:56 olas how resolve?
07:56 egonw hang on
07:58 * egonw is updating his svn checkout of bc2
07:58 egonw which takes a while...
07:58 egonw will then check the svn:ignores for cdk10
07:58 olas ok
07:59 * jonalv is just done doing the same :)
08:00 jonalv Since CDK isn't compiling in trunk I might as well not work structuredb so what can I do to help you? :)
08:02 olas CDK is
08:02 olas CDK10 is not
08:02 olas right?
08:02 jonalv is CDK actually supposed to export all thos packages?
08:02 jonalv Neither one works for me...
08:02 olas egonw: please help
08:02 olas no CDK compiling :-(
08:03 jonalv it's exporting things that it can't find. But it's strange things like com.sun.media.jai and org.apache.batik
08:04 jonalv CDK can't reasonably be supposed to export those
08:04 jonalv or can it?
08:05 egonw back
08:07 CIA-44 bioclipse: egonw * r5503 /bioclipse2/trunk/plugins/net.bioclipse.cdk10/: Fixed this externals
08:07 egonw problem was that I had to changed jars/ to jar/
08:07 egonw but forgot to update the svn:externals
08:07 egonw pleaes update
08:07 olas cdk10 or cdk or both?
08:08 egonw cdk10
08:08 olas k
08:08 olas jonalv: does cdk compile for you?
08:09 jonalv hang on
08:09 * olas is updating cdk10
08:09 * jonalv is updateing bc2
08:10 jonalv building now, my Pan isn't the fastest computer around... :)
08:10 jonalv okey, cdk10 compiles but not cdk
08:11 egonw jonalv: ack
08:11 CIA-44 bioclipse: egonw * r5504 /bioclipse2/trunk/plugins/net.bioclipse.cdk/: Fixed this externals
08:11 egonw another jars/ jar renaming problem
08:11 olas my cdk10 does not compile
08:11 egonw olas: what error
08:11 * egonw would love to get ant compiling online
08:12 olas many
08:12 egonw give me the first
08:12 olas import net.sf.jniinchi.INCHI_RET;
08:12 jonalv egonw, yea that would be great
08:12 olas next: import com.ozten.font.JFontChooser;
08:12 egonw olas: yeah, well I was already complaining about the .classpath to be reset upon 'Clean classpath'
08:13 egonw make sure you have the .classpath from SVN
08:13 olas import Jama.EigenvalueDecomposition; import Jama.Matrix;
08:13 egonw (or add many jars from jar/)
08:13 olas ok
08:13 egonw olas: please confirm that you have these jars:
08:13 egonw jar/jfontchooser.jar
08:13 egonw jar/jama.jar
08:13 egonw jar/jniinchi-0.4.jar
08:13 egonw to name those related to the errors you posted here on IRC
08:14 olas my minifest does not contain imports of any jars at all
08:14 olas why is not this the case?
08:14 egonw no idea
08:14 egonw maybe that's why 'clean classpath'
08:14 egonw messes up my .classpath all the time
08:14 egonw hang on...
08:14 olas fixing it
08:14 olas yup
08:14 olas solves the problem
08:14 egonw that's Bundle-ClassPath:
08:14 egonw right?
08:15 olas ?
08:15 egonw the relevent lines in MANIFEST.MF
08:15 olas Classpath on Runtime tab
08:15 olas yes
08:15 egonw ok, please commit
08:15 CIA-44 bioclipse: ospjuth * r5505 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10/META-INF/MANIFEST.MF: Added jars to manifest.
08:16 olas compiles here now
08:16 olas should we export anything to JCP?
08:16 olas cdk10.jcp complains about vecmath missing
08:17 egonw yeah, cdk10 should export javax.vecmath
08:17 egonw thought I had commited that :(
08:17 olas will do that
08:17 egonw well, there is a reason why egonw wrote in the Bioclipse wiki the page Classpath Horror
08:17 olas or you?
08:18 egonw please do
08:20 olas hmm
08:21 olas restriction on vecmath...
08:21 olas what does that mean?
08:21 olas point2D is not available due to restriction on package vecmath
08:22 olas help?
08:22 CIA-44 bioclipse: ospjuth * r5506 /bioclipse2/trunk/plugins/net.bioclipse.cdk10/ (META-INF/MANIFEST.MF build.properties): Exported vecmath but still does not work.
08:22 egonw not sure
08:22 jonalv sounds like the vecmath is not exported from the jar file
08:23 egonw sounds like it is not exported?
08:23 egonw jonalv: :)
08:23 olas hmm, but I did
08:24 jonalv olas: did you hack the jar file?
08:24 olas javax.vecmath is exported
08:24 olas hack?
08:24 olas I added a line to the manifest
08:24 olas please checkout and see for yourselves
08:24 jonalv I am on it
08:25 jonalv which jar file is supposed to contain it?
08:25 egonw jar/vecmath1.2.jar
08:29 jonalv okey you didn't update the classpath
08:29 jonalv or did you?
08:31 CIA-44 bioclipse: jonalv * r5507 /bioclipse2/trunk/plugins/net​.bioclipse.cdk10/.classpath: updated classpath. jchempaint compiles for me after this
08:31 CIA-44 bioclipse: ospjuth * r5508 /bioclipse2/trunk/plugins/net.bioclipse​.cdk10.jchempaint/src/net/bioclipse/cdk​10/jchempaint/ui/editor/JCPPage.java: Added file accessing, placed FIXME for egonw to continue parsing stream into JCPModel.
08:32 olas same errors
08:32 olas yes, I updated CP
08:32 jonalv it worked for me... :)
08:32 * jonalv is updating now
08:33 * olas will make coffee
08:33 olas egonw: placed FIXME for you
08:34 egonw ack
08:34 jonalv yey everything compiles for me now
08:35 jonalv so does the jchempaint do anything yet? Can I test it?
08:35 jonalv :)
08:35 olas jonalv: there's a LOT OF work before that
08:35 jonalv k
08:36 olas you can test the structureWidget
08:36 jonalv okey, how?
08:36 olas thats not CDK10 but CDK
08:36 egonw jonalv: what I'd really like to see, it a "SDF import" for the structuredb
08:36 olas expand an SDF-file with the new view open and click an IMolecule
08:36 egonw I'd even rather see that than the SDF editor you worked on a bit
08:37 olas egonw: yea, that's what jonalv is working on
08:37 olas I worked on the SDFEditor
08:37 jonalv egonw: Yea how hard can that be? wouldn't CDK do all the job for me?
08:37 olas jonalv will do SDF import
08:37 olas jonalv: what's the status?
08:37 olas with structureddb?
08:38 * jonalv is fighting a bit with HSQLDB but that might just be trouble when running JUnit tests. I have not tried anything from the GUI
08:38 egonw jonalv: yes, indeed... all I need to know where to put the code...
08:39 egonw jonalv: if your provide me a template,
08:39 egonw ... then I'll fill in the CDK bits
08:39 jonalv I would love a method in the cdk manager that takes an sdf file and returns a set of IMulecules
08:39 olas hmm
08:39 jonalv perhaps an ieterator to them is better
08:40 jonalv on second thought :)
08:40 egonw yes, indeed
08:40 olas that would be trivial for egonw
08:40 egonw you don't want to open a PubChem SDF in that way
08:40 jonalv we don't wanan read them all into memery at the same time I guess
08:40 egonw correct
08:41 CIA-44 bioclipse: egonw * r5509 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10/META-INF/MANIFEST.MF: Export the root CDK java package too
08:41 jonalv egonw: Why not? That should contain all I store in the database (I think).
08:41 egonw jonalv: meaning... we should use the iterator...
08:41 jonalv ah yes
08:41 egonw and one by one load them, then stored them in the db, then delete them
08:42 egonw the iterator can do milions of SDF entries
08:42 jonalv yea
08:42 jonalv we just have to hope the garbage collector does it job :)
08:44 egonw right
08:45 egonw ok, me patched JCPPage to implement the InputStream -> ChemModel
08:45 egonw and gets a NPE :)
08:45 jonalv brb
08:47 * egonw implemented a STUB
08:47 egonw solving the NPE
08:47 olas let me know if probs
08:47 egonw and got a new one :)
08:48 egonw but, I get a white JCP background now... so making progress :)
08:48 egonw well... one problem is that JCP seems only to take half of the widget...
08:49 egonw which you might have a look at?
08:49 egonw while I get banging out those exceptions...?
08:49 jonalv back
08:51 egonw implemented a few more stubs
08:51 egonw yes!
08:52 egonw I see a structure !
08:52 egonw cool!
08:52 jonalv w00t
08:56 egonw joined #bioclipse
08:56 egonw oops... wrong close window :)
08:57 jonalv :)
08:59 egonw http://chem-bla-ics.blogspot.com/200​8/04/jchempaint-for-bioclipse2.html
09:01 CIA-44 bioclipse: egonw * r5510 /bioclipse2/trunk/plugins/net.bioclipse​.cdk10.jchempaint/src/net/bioclipse/cdk​10/jchempaint/ui/editor/JCPPage.java: Basic JCP online; not sure it even edits yet, but display works: http://chem-bla-ics.blogspot.com/200​8/04/jchempaint-for-bioclipse2.html
09:02 olas egonw: can there be a name in an SDF file?
09:03 olas and how do I retrieve it from the AtomContainer?
09:03 egonw getProperty(CDKConstants.TITLE)
09:04 olas ok
09:04 CIA-44 bioclipse: ospjuth * r5511 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui/src/net/bioclipse/cdk/ui/editors/sdf/ (4 files): The SDFEditor now displays the parsed SDF file as SMILES. Will continue to update using StructureWidget.
09:08 jonalv olas++, egonw++ //Things are happening! :)
09:12 CIA-44 bioclipse: jonalv * r5512 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui/src​/net/bioclipse/cdk/ui/editors/sdf/SDFEditor.java: Stay within the 80 cols marginal
09:15 thomas_ku joined #bioclipse
09:18 olas jonalv: please do not edit my code at this point
09:18 olas it is under heavy development
09:19 egonw olas: a template for JCPPage.doSaveAs() would be appreciated too
09:19 olas jonalv: please work on the structuredb instead
09:19 olas hmm
09:19 olas egonw: here is what you do
09:19 jonalv sry I couldn't help myself. I was locking at how you where doing things and couldn't help myself :)
09:19 olas 1) sync model so that texteditor is updated
09:19 jonalv I am on that SDF import track
09:20 olas 2) delegate doSaveAs to texteditor.soSaveAs()
09:20 olas egonw: ok?
09:20 egonw actually... no
09:21 egonw it's more complex...
09:21 egonw because the tabs need synching...
09:21 egonw ah, step 1)
09:21 egonw ok, never mind :)
09:22 malick joined #bioclipse
09:33 CIA-44 bioclipse: egonw * r5513 /bioclipse2/trunk/plugins/net.bioclipse​.cdk10.jchempaint/src/net/bioclipse/cdk​10/jchempaint/ui/editor/JCPPage.java: Some cleanup, and implementation of other methods (by C/P from BC1 code)
09:33 CIA-44 bioclipse: egonw * r5514 /bioclipse2/trunk/plugins/net.bioclipse​.cdk10.jchempaint/src/net/bioclipse/cdk​10/jchempaint/ui/editor/JCPPage.java: Fixed copyright information
09:33 CIA-44 bioclipse: egonw * r5515 /bioclipse2/trunk/plugins/net.bioclipse​.cdk10.jchempaint/src/net/bioclipse/cdk​10/jchempaint/ui/editor/JCPPage.java: Fixed half screen mode of JCP tab
09:35 CIA-44 bioclipse: jonalv * r5516 /bioclipse2/trunk/plugins/ (4 files in 3 dirs): Wrote failing test for createMoleculeIterator in cdk manager
09:36 * jonalv would really like a automagic build and test runner
09:44 olas jonalv: please write to bioclispe-devel and ask rklancer and grantsparks
09:44 olas both wanted to jump on that I think
09:44 * jonalv is not sure they are left in the game..
09:45 olas please write anyway!
09:45 olas I'd appreciate it
09:49 egonw jonalv: about that last commit
09:50 egonw where do I start hacking?
09:50 * egonw will read chapter 19 in the car to my farther in law
09:50 jonalv net.bioclipse.cdk.business.CDKManager
09:50 egonw ok
09:50 jonalv but maybe you should start by having a look at the test...
09:51 jonalv brb
09:53 jonalv back
09:53 egonw ha, me run BC2 unit tests!
09:53 egonw for the first time!
09:53 egonw that was fun :)
09:53 jonalv w00t
09:54 egonw oi, you understand the implementation of that method, right?
09:55 jonalv hm that's always a trixy wuestion to asnwer...
09:56 egonw hehehe
09:56 egonw ok, here's the impl:
09:56 jonalv If you mean that I know what CDK components are needed the answer is: no
09:56 egonw public Iterator<IMolecule> creatMoleculeIterator(InputStream instream) {
09:56 egonw return new IteratingMDLReader(instream, NoNotificationChemObjectBuilder.getInstance());
09:56 egonw }
09:57 egonw will commit that
09:57 jonalv uhm that's not all I am afraid...
09:58 jonalv the IMolecule interface is the Bioclipse one... extending IBIoObject
09:58 jonalv or should be anyway...
09:58 egonw ah...
09:58 egonw different IMolecule :)
09:58 jonalv yup
09:58 egonw sorry.. did not notice that :)
09:58 egonw ok, point taken
09:58 jonalv we need a few Bioclipse spcific things on it :)
09:58 egonw will work on that
09:58 egonw but not now...
09:59 egonw because my ride is about to leave...
09:59 jonalv okey
09:59 egonw so, will go offline, and will be back around 14:00, I hope
09:59 jonalv cya
09:59 egonw after drive and subsequent lunch
09:59 egonw will hack up a wrapper...
09:59 egonw so, is there a IMolecule impl?
09:59 jonalv yea a few I think
10:00 jonalv you will proabbly want the CDKMolecule
10:00 egonw ok, cu later
10:19 steinbeck joined #bioclipse
10:22 * jonalv is off for food somewhere. cya
11:07 masak joined #bioclipse
11:07 olas hi masak
11:07 masak hi olas
11:07 * masak backlogs
11:17 masak whoa, seems you already have tangible results
11:20 masak mad props to all involved: egonw++; olas++; jonalv++
11:24 * jonalv is back
11:24 masak wb jonalv
11:24 jonalv any news?
11:28 masak only that the bank business is taken care of :)
11:28 jonalv okey
11:28 jonalv I am fighting against IOExceptions
11:29 jonalv handling files isn't so easy
11:29 jonalv you know how to do it?
11:29 masak programming in general would be easier without I/O
11:29 masak jonalv: I don't know
11:29 masak I've done it, but not extensively
11:29 jonalv you wanna help me?
11:30 masak sure
11:30 jonalv structuredb.tests   StructuredbMnanagerTEst
11:31 masak on my way
11:31 olas masak: I'm in need of help too
11:32 masak olas: tell me more
11:32 * jonalv is going to check if Erica is here. masak: keep an eye on the log as well there is more stacktraces there...
11:32 jonalv brb
11:32 masak jonalv: k
11:34 olas weehoow!
11:34 jonalv back
11:34 jonalv she wasn't here yet
11:34 * olas just created a StructureTable
11:34 jonalv *there
11:34 masak olas: gratki
11:34 olas WEEHOOW!!
11:35 jonalv okey, what works?
11:35 jonalv :)
11:35 jonalv ooh
11:35 jonalv how do I test it?
11:35 jonalv =)
11:35 olas and it's fast!
11:35 olas WOW
11:35 olas that SWT.VIRTUAL is really good
11:36 jonalv nice
11:36 olas plz await my commit
11:36 * jonalv is awaiting comit
11:36 masak I've heard good things about it, too
11:36 * olas is VERY happy
11:36 * jonalv is awaiting commit ;)
11:36 olas will be a blog about this :-)
11:37 masak olas: please call blog posts blog posts, not blogs :)
11:38 jonalv masak, HAve you managed to run the tests?
11:38 masak just about to
11:38 masak hm, both fail
11:38 jonalv yup
11:39 jonalv timeout?
11:39 masak looking
11:39 olas jonalv--
11:39 jonalv What is it now?
11:39 masak olas: hey, failing tests are good!
11:39 masak jonalv: no, not that I can see
11:39 olas I have conflicts
11:39 masak ah
11:40 olas due to jonalvs messing with my code
11:40 masak I see mostly SQL expressions
11:40 jonalv lol, keep the 80 cols marginal then :)
11:40 masak I was going to step in in olas' defense
11:40 jonalv it can't be that bad thugh. I only achagned two lines
11:40 masak but now that I know that it's about the 80 columns...
11:40 CIA-44 bioclipse: ospjuth * r5517 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui/src/net/bioclipse/cdk/ui/editors/sdf/ (7 files): The SDFEditor now displays the structure inline the table. Really fast with the SWT.VIRTUAL option set! Double-click an SDFile to test.
11:42 masak jonalv: the first one gives an EOFException
11:42 masak the second one gives a so-called BioclipseException
11:42 masak about not being able to create a reader in CDK
11:42 masak what's a BioclipseException?
11:42 jonalv ask olas
11:43 masak olas: what's a BioclipseException?
11:43 olas jonalv: I want results presented with similarity displayed as a bar chart, see http://dev.eclipse.org/viewcvs/index.cgi​/org.eclipse.swt.snippets/src/org/eclips​e/swt/snippets/Snippet228.java?view=co
11:43 jonalv huh what resutls? what similiratity?
11:43 olas looks good IMO
11:44 olas :-)
11:44 olas chew on that
11:44 jonalv you lost me
11:44 olas of a similarity search
11:44 jonalv in structuredb?
11:44 masak olas: yes, it does look good :)
11:44 olas compounds with high ranking should come out on top
11:44 jonalv aha
11:44 masak olas: what's a BioclipseException?
11:44 olas what you like it to be
11:45 masak hm
11:45 masak I'd like it to be a RuntimeException
11:45 jonalv "a class should do one thing and one thing only"?
11:45 olas feel free to change it
11:45 masak jonalv: as long as I'd only like it to be one thing... :)
11:45 masak olas: oki
11:45 olas Maybe set up a wiki page about excpetions in BC2?
11:46 masak sure
11:46 jonalv I think we wanted it to be checked one
11:46 masak it's on my TODO list
11:46 masak jonalv: ah
11:46 jonalv masak: why do you want a runtimeexception?
11:46 masak jonalv: why have it checked if it's punted all the way up to top level?
11:46 jonalv We should probably have IRL meeting about this sometime and have another look at the exceptions in bc2
11:46 masak seems a waste
11:47 masak yup
11:47 masak a checked exception is an exception you can do something about, right?
11:47 jonalv olas: about that diagram, is there a screendump somewhere?
11:47 masak what can you do about not being able to create a reader in CDK?
11:48 jonalv not much I think.
11:48 olas ??
11:48 masak jonalv: so, why checked?
11:48 jonalv masak, I don't think I wrote it and anyway I don't rememebr the thoughts around that
11:49 * masak changes
11:50 masak jonalv: btw, did I help you with that structuredb thing?
11:50 jonalv masak: no
11:50 masak can I do anything else before I leave it?
11:50 jonalv uhm help me fix it?
11:50 masak jonalv: am I right in assuming we get different errors?
11:50 jonalv it seems so
11:51 jonalv try running all the tests in the plugin
11:51 masak eh, ok
11:51 jonalv I have a feeling they effect eachother... :(
11:52 masak might be
11:53 jonalv I am making a commit with some logging code which might help us
11:53 CIA-44 bioclipse: jonalv * r5518 /bioclipse2/trunk/plugins/ (2 files in 2 dirs): Wrote some logging in structuredb and it's tests
11:53 masak jonalv: yup, now I get an out-of-time exception
11:53 jonalv update and run it again and check the log
11:53 masak k
11:54 * jonalv is gonna test that sdf viewer :)
11:54 jonalv uhm editor
11:55 egonw joined #bioclipse
11:56 egonw jonalv, olas: had a traffic jam, and throwing up kid...
11:56 egonw now going for lunch, I'll be back in about 30 minutes
11:57 masak olas: do you know someone who might be interested in a SAS programmer? a friend of mine is asking. she has a masters degree in med.sci.
11:59 masak jonalv: what am I looking for in the logs?
11:59 masak I see 2 IOExceptions
11:59 jonalv a stack trace
11:59 masak yes, those
11:59 jonalv yup one for each test
11:59 masak both are saying "No such file or directory"
12:00 jonalv olas: I tried with a BIG sdf file and bioclipse is somewhat totally locked now even before I had the time to open the editor :(
12:00 masak jonalv: not the same as that reported by the test
12:00 jonalv no because they are catched and printed only
12:00 jonalv ah bioclipse is back, don't know what it was doing...
12:01 jonalv olas: The editor works great! :)
12:01 jonalv this makes me wanan move brunn into bc2
12:03 jonalv masak: Can you understand why it thorws those exceptions?
12:04 edrin joined #bioclipse
12:04 jonalv hi edrin
12:04 edrin hi
12:04 edrin http://xkcd.com/
12:04 masak jonalv: I can try
12:04 edrin haha
12:05 jonalv lol
12:07 jonalv masak: It would be higly appredciated this si a bit of a show stopper...
12:07 * masak is on it
12:07 jonalv what troubles me is that sometimes it just works
12:07 jonalv like now for example... :/
12:08 masak seems a timing problem, then
12:08 jonalv perhaps
12:10 steinbeck joined #bioclipse
12:12 masak jonalv: have you tried adding a second-long pause at the beginning of the failing test?
12:12 CIA-44 bioclipse: jonalv * r5519 /bioclipse2/trunk/plugins/net.bioclipse​.structuredb/src/net/bioclipse/structur​edb/business/StructuredbManager.java:
12:12 CIA-44 bioclipse: minor details
12:12 CIA-44 bioclipse: switched to unsopprtedoperationexception and added a few whitespaces in log message
12:12 jonalv no
12:12 jonalv masak: no I haven't
12:12 * masak tries
12:13 jonalv but why would it work? I think the problem is with files not with hsaldb
12:13 masak hm
12:13 jonalv you don't?
12:13 masak are we still talking about the timeout?
12:13 jonalv yea
12:14 masak how can a timeout be about files?
12:14 masak file access doesn't time out
12:14 jonalv I think the timeout exception is a second hand error after the file error
12:14 masak might be
12:14 masak what do you base that on?
12:14 jonalv mm
12:15 masak sometimes we get the timeout, sometimes the file error
12:15 masak why not always just one of them?
12:15 jonalv no I always get both
12:15 jonalv you don't?
12:15 masak olas: do you know someone who might be interested in a SAS programmer? a friend of mine is asking. she has a masters degree in med.sci.
12:15 masak jonalv: no, I don't think do
12:16 masak I got the file error when I ran the test class on its own
12:16 jonalv I get the IOExcpetion in the logs and the timeout in the JUnit view
12:16 masak I got the timeout when I ran all tests
12:16 masak ah
12:16 masak yes
12:16 masak I'm with you
12:16 jonalv good
12:17 masak so
12:17 masak why are the files missing?
12:17 jonalv wanna skype instead?
12:17 masak if you prefer to talk to my beard, sure
12:18 masak like it's not crowded enough here with just one screen... :)
12:29 steinbeck_ joined #bioclipse
12:34 egonw joined #bioclipse
12:35 egonw olas: to what SDF size does it scale?
12:37 jonalv theeditor wroked with a file of 6MB but try to open the subelemtns
12:37 jonalv and it fails
12:37 jonalv (for me)
12:43 egonw jonalv: iterator test works now
12:43 egonw wrote the wrapper to create CDKMolecule, using the a Bioclipse IMolecule impl
12:43 egonw being CDKMolecule
12:44 jonalv nice
12:44 jonalv me and masak is stuck in some problem with hsqldb :(
12:44 olas egonw: not tried it
12:44 olas just tried 2112 mols
12:45 CIA-44 bioclipse: egonw * r5520 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/business/CDKManager.java: Added support for reading SDF
12:45 olas works like a charm
12:45 egonw excellent
12:45 egonw will try it here too
12:45 olas egonw: already read sSDF with loadMolecules()
12:46 CIA-44 bioclipse: ospjuth * r5521 /bioclipse2/trunk/plugins/net.bioclipse.cd​k.ui/src/net/bioclipse/cdk/ui/editors/sdf/ (4 files): Set init size to 100x100. Display name if exists.
12:47 * egonw is googling to learn how to fake his MAC
12:58 stain ifconfig
12:58 stain ifconfig eth0 hw ether de:ad:be:ef:..
13:00 EskilA joined #bioclipse
13:02 egonw stain: ack, that worked
13:12 egonw_ joined #bioclipse
13:14 edrin egonw_: ?
13:15 edrin egonw_, i have a cell extract. can you tell me how much alpha-1-2-mannoproteins is in there?
13:15 edrin or anyone else?
13:15 egonw_ theoretically, yes
13:16 edrin heh theoretically?
13:16 edrin :)
13:16 edrin how would I measure this?
13:17 edrin in fact I need to measure it..
13:18 egonw_ I'd guess a proteomics like experiment could do...
13:22 egonw_ Uppsala: I have to say that BC2 feels much more performant!
13:22 egonw_ not sure which event loop it no longer has...
13:22 egonw_ but the difference is nice :)
13:23 masak egonw_: good to hear
13:23 masak that's nothing we've given much thought to, as far as I know
13:23 masak but it might very well be that it falls out of the design...
13:24 masak or you just had more in your 1.x installation :)
13:36 * egonw_ gives up on getting wired internet to work
13:37 edrin egon: maybe try this computer: http://xkcd.com/
14:06 olas http://bioclipse.blogspot.com/2008/04/sdfe.html
14:06 olas egonw: you like it?
14:06 egonw_ olas: I'm struggling with the mouse events...
14:06 egonw_ mom, reading it
14:07 olas I'm totally not into mouse events at all :-(
14:07 * jonalv has given up mastering hsqldb and goes back to reading the manual
14:07 jonalv it's complicated when running multiple instances
14:07 jonalv sometimes it works sometimes it doesn't...
14:08 egonw_ olas: yes, looks great!
14:08 masak olas++ # screenshot
14:08 olas thx
14:08 masak hehe, "OlaProject" :)
14:08 olas whateve
14:09 masak no, I like it
14:09 egonw_ olas: you can propagate the editor cell size to the JCP SWT widget
14:09 olas not yet
14:09 olas have not found out that
14:09 egonw_ not sure how to resize the height in the GUI, though
14:09 olas my point
14:09 egonw_ the View had code for that
14:09 egonw_ did you try a SDF from PubChem FTP?
14:09 olas no
14:10 olas don't use your new loading yet
14:10 olas uses loadMolecules()
14:10 olas must wrap up a presentation for tomorrow now
14:10 egonw_ ok
14:10 olas will present for Uppsala HPC group
14:10 olas for Grids in Bioclipse :-)
14:10 egonw_ ok...
14:11 * egonw_ is away for some tea
14:11 olas very interesting project
14:11 olas egonw: how is progress with JCP?
14:12 egonw_ mouse events are offline...
14:12 egonw_ trying to get those going now
14:12 egonw_ but tea first now
14:12 * olas @ coffee
14:13 CIA-44 bioclipse: ospjuth * r5522 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui/ (4 files in 2 dirs): Read and display properties in SDFileEditor.
14:30 CIA-44 bioclipse: jonalv * r5523 /bioclipse2/trunk/plugins/ (4 files in 4 dirs): Work with trying to get multiple databases to work. No success though. This code mostly consists of cleanups and logging.
14:43 CIA-44 bioclipse: jonalv * r5524 /bioclipse2/trunk/plugins/net.bioclips​e.hsqldb.tests/tests/log4j.properties: added log4j.properties
14:51 * jonalv gets tea
15:10 edrin hey olas, how was your talk at Embrace?
15:11 jonalv anyone wanna help me with hsqldb?
15:13 olas edrin: it was really good
15:13 olas jonalv: not really...
15:13 olas Derby?
15:13 jonalv anyone know how Darvy works?
15:14 jonalv I had a look on it and the first I ran into was something about global path variables. It sounded dangerous.
15:14 edrin olas: did you talk about the XMPP services?
15:14 jonalv And then instructions for how to set them in all different sytems. How would that work with bioclipse?
15:15 egonw joined #bioclipse
15:17 olas edrin: I mentioned it
15:17 olas but had little time
15:17 olas and my talk was the last, everyone was tired
15:17 olas anyway, they were informed
15:18 olas and I said that they should contact me or you if they were interested
15:18 jonalv egonw, you started a hsqldb plugin back in the days right? Have you used it before?
15:18 * jonalv needs help :/
15:19 egonw jonalv: yes, I did use it... not much though...
15:20 egonw won't be of any help, I'm afraid
15:20 * egonw just knows that is was a pain in the a**
15:20 jonalv okey
15:20 jonalv yea but derby seemed stragne
15:20 jonalv they talked alot about some global path stuff
15:21 jonalv I have no idea what that is or how it would effect me if I where to go with derby...
15:30 egonw yeah, I know that feeling...
15:32 jonalv I am honestly beginning to get close to giving this up. How bad would it be if we settle with only being able to run one local database instance?
15:34 egonw and good import for other dbs?
15:37 olas jonalv: very strange
15:37 olas I honestly can't believe we are the only ones trying to set up two HSQLDB instances
15:37 olas have you asked on their ML?
15:38 egonw ah, finally... I got mouse events :)
15:38 jonalv I am trying on theri forum but
15:38 jonalv I don't know
15:38 jonalv egonw, good imports?
15:38 egonw now I'm going to try to propagate them into JCP
15:39 egonw no, could not find where to set the listener...
15:39 egonw remember it's ugly code
15:39 jonalv The plan was for using mysql to set up public databases and write code to be able to use them too
15:39 egonw jonalv: oh... that import
15:39 egonw good imports -> e.g. biomart
15:39 egonw does multiple local dbs work?
15:39 jonalv sometimes they do and sometimes they don't
15:40 jonalv I don't see any reason for why though...
15:40 jonalv but mostly it doesn't
15:48 * egonw is reconsidering the multipage editor for JCP...
15:48 egonw synching makes it unrealistically more complicated...
15:48 egonw and also adds an extra layer of widgets...
15:48 jonalv it does?
15:48 jonalv too bad
15:48 egonw yeah, because there are two data models...
15:48 jonalv ah
15:49 egonw a plain text one... and a CDK-based one
15:49 egonw also, I have no idea where events go or should be catched
15:49 jonalv how do you mean?
15:50 egonw well... where the heck should my mouse events propagate too?!?!?
15:51 jonalv I don't think I undestand the problem...
15:52 egonw never mind
15:52 egonw well...
15:53 egonw the BC1 JCPEditor only has a MouseMotionListener...
15:53 jonalv yea
15:53 egonw so, where does it catch the MouseClick events??
15:53 jonalv hm didn't you write it?
15:53 egonw no, Tobias and Stefan did
15:54 jonalv k
15:55 jonalv but how does that complicate the multipage part?
15:55 egonw because it has more core to browse through
15:55 egonw I mean... some 10 classes for just a JCP editor?
15:56 egonw excluding all CDK logic ... !
15:56 jonalv I don't find that strange. After all it's GUI we are talking about here...
15:57 egonw ah...
15:57 jonalv you found something?
15:57 egonw not 10 classes used in an EditorPage class...
15:57 egonw but 10 classes to define the code of one EditorPage
15:58 egonw but, yes that too...
15:58 egonw just found a second place where a DrawingPanel is constructred...
15:58 egonw now with a MouseListener :(
15:58 jonalv okey...
15:58 egonw apparently...
15:59 egonw there is a PopupController2D in between...
15:59 jonalv I am thinking you have some work there :)
15:59 egonw no
16:00 egonw it's not my code
16:00 egonw I won't maintain this code
16:00 egonw don't have time for that...
16:00 egonw the code needs cleaning up...
16:00 jonalv ah I mean in reading it and reproducing the functionality
16:00 egonw that's a FTE
16:00 jonalv FTE?
16:00 egonw no, won't be able to reproduce all functionality at all
16:00 egonw full time equivalent
16:00 egonw 1 manyear
16:01 egonw I'll try to reproduce *some* functionality...
16:02 jonalv good :)
16:02 egonw no idea when I'll end up...
16:02 egonw I hope that some editing will work
16:02 egonw then I'm going to give up...
16:02 egonw Tobias and/or Stefan is needed to get this porting finished
16:02 egonw any Java coder would do for a rewrite based on the new renderer
16:03 egonw the latter would even be better
16:03 jonalv isn't Tobias out in the business world?
16:03 jonalv And Stefan is probably very bussy right?
16:03 egonw OMG
16:03 jonalv ?
16:04 egonw it's the universal call, for HELP, OR JUST KILL WAY RIGHT AWAY
16:04 egonw (oh, my god!)
16:04 jonalv yea
16:04 masak usually a sign of surprise
16:04 jonalv but why*?
16:04 egonw fortunately, one can safely say that, as he hardly ever responds (if at all)
16:05 egonw or she
16:05 egonw or it
16:05 jonalv did your kid break something or what happened? :)'
16:05 masak egonw: the usual response to that is that you're just not listening properly
16:05 jonalv hi masak are you still here? :)
16:05 masak but I would agree that that's a fairly lame response
16:05 masak jonalv: seems so
16:05 masak just not working :)
16:06 masak writing a very belated document
16:06 egonw the OMG refered to a pop up controller with a constructor which modeled about the world...
16:06 jonalv but please egonw we (at least me) is very curious about that OMG call. What's up?
16:07 egonw I just hate constructors with more than 3 parameters...
16:07 masak who doesn't? :)
16:07 egonw this one has 6
16:07 jonalv uhm...  me?
16:07 jonalv even me finds that a little bit too many...
16:19 * jonalv is giving up now. olas: I think we should reevalutate how important multiple local instances are tomorrow...
16:25 jonalv cya
17:03 CIA-44 bioclipse: egonw * r5525 /bioclipse2/trunk/plugins/net​.bioclipse.cdk10.jchempaint/ (17 files in 3 dirs): Ported more stuff from BC1, no idea what it all does, and still no editing functionality... :(
17:03 egonw enough for today...
17:03 egonw dinner time
17:36 rojasm joined #bioclipse
17:44 rojasm left #bioclipse
18:19 edrin left #bioclipse
18:34 CIA-44 bioclipse: egonw * r5526 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/src/net/bio​clipse/cdk10/jchempaint/ui/editor/ (3 files in 2 dirs): OK, some editing is starting to work... highlighting is working... actions or update of model is not yet working though; also icons are missing
18:52 olas left #bioclipse
20:00 CIA-44 bioclipse: egonw * r5527 /bioclipse2/trunk/plugins/net.bioclipse.c​dk10.jchempaint/src/net/bioclipse/cdk10/j​chempaint/ui/editor/resources/small-bin/ (42 files): Added missing icons
20:19 CIA-44 bioclipse: ospjuth * r5528 /bioclipse2/trunk/plugins/net.bioclipse.cdk.ui/src​/net/bioclipse/cdk/ui/editors/sdf/SDFEditor.java: Added a simple progressbar for parsing large SDFiles.

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