Camelia, the Perl 6 bug

IRC log for #bioclipse, 2008-06-02

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

All times shown according to UTC.

Time Nick Message
04:22 malick joined #bioclipse
05:07 egonw_ joined #bioclipse
06:56 Gpox joined #bioclipse
07:09 CIA-46 bioclipse: goglepox * r5769 /bioclipse2/trunk/plugins/net.bioclipse.​cdk10.jchempaint/src/net/bioclipse/cdk10​/jchempaint/ui/editor/DrawingPanel.java: Changed paint to paintComponent
07:12 CIA-46 bioclipse: goglepox * r5770 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/src/net/bio​clipse/cdk10/jchempaint/ui/editor/ (JCPMultiPageEditorContributor.java JCPPage.java): Fix for SDFEditor, creates new BCJCPPopupController when the ChemModel is changed.
07:19 EskilA joined #bioclipse
07:25 jonalv joined #bioclipse
08:14 CIA-46 bioclipse: jonalv * r5771 /bioclipse2/trunk/plugins/net.bioclipse.core/ (3 files in 3 dirs):
08:14 CIA-46 bioclipse: molecule manager now get's published. Thanks egonw++ for fixing my error reporting so that finding the error was a walk in the park!
08:14 CIA-46 bioclipse: Also switched from BioclipseException to UnSupportedOperationException for getCML on Molecule.
08:25 egonw joined #bioclipse
08:26 jonalv hi egonw
08:26 CIA-46 bioclipse: jonalv * r5772 /bioclipse2/trunk/plugins/net.bioclipse.structure​db/src/net/bioclipse/structuredb/persistency/dao/ (LabelDao.java StructureDao.java): Fixed cascade persisting from Label to Structure
08:27 thomas_ku joined #bioclipse
08:28 egonw hi jonalv
08:28 egonw hi others
08:29 jonalv I fixed the problem with the molecule manager
08:29 jonalv egonw: thanks for the help
08:29 egonw don't have much time, really...
08:29 egonw but will update and test if it solves booting BC2
08:29 jonalv yea, it shouldn't effect you booting of bioclipse though...
08:30 egonw I hope it does ...
08:30 jonalv but it's worth a try :)
08:30 olas joined #bioclipse
08:30 olas hi
08:30 jonalv hi olas
08:30 egonw hi olas
08:30 olas egonw: did you see my bug regarding cdk10-> cdk conversion?
08:30 olas hi btw :-)
08:30 egonw yes, I did
08:30 olas comment?
08:30 egonw not fixed though...
08:31 egonw no comment at this moment...
08:31 olas semms like I need the MassNumber
08:31 egonw bloody important project deadline today...
08:31 olas I know
08:31 egonw URL?
08:31 olas to bug?
08:31 egonw ye
08:31 egonw +s
08:31 olas mom
08:31 olas http://sourceforge.net/tracker/inde​x.php?func=detail&aid=1979421&a​mp;group_id=150681&atid=778609
08:32 olas If I can just set the MassNumber, I think it would work
08:33 egonw commented
08:33 olas is this lost in the CML conversion or is it something that was added between cdk10 and cdk?
08:33 olas ok
08:33 egonw no, is lost in translation
08:33 egonw do you want to fix it?
08:34 egonw I won't have time until tonite
08:34 egonw I can explain how
08:34 olas hard?
08:34 egonw not really
08:34 olas got 30 mins
08:34 egonw and rather educational
08:34 olas maybe 45
08:34 egonw should work
08:34 olas go on then, I'll give it a try
08:34 egonw open the org.openscience.cdk project
08:34 egonw go into src/test
08:35 egonw cdk.io.cml
08:35 egonw CMLRoundTripTest
08:35 olas ok
08:35 olas I'm there
08:35 egonw add a unit test there...
08:35 egonw round trip one atom
08:35 olas how create the atom with MN?
08:35 egonw use atom.setMassNumber(120
08:35 egonw 12)
08:35 egonw for a carbon or so
08:35 olas Atom.setMassNumber?
08:35 egonw yes
08:35 olas ok
08:36 olas I can do that
08:36 olas but how do I fix it?
08:36 egonw then assertEquals(atom.getMassNumber(), roundTrippedAtom.getMassNumber();
08:36 egonw ok...
08:36 egonw ok, go to:
08:36 egonw CML2WriterTest
08:37 egonw testHydrogenCount
08:37 egonw write a unit test like that for MassNumber
08:37 egonw use the writer.toString() (e.g. on line 113) to introspect the CML cerated
08:37 olas ok
08:37 egonw and test for that using indexOf() (as in line 115)
08:38 egonw ok, then two alternatives:
08:38 egonw 1. it made the CML
08:38 egonw but reading fails
08:38 egonw 2. it never was stored in the CML
08:38 olas right
08:38 egonw (well, 3: both are not working)
08:38 olas ok, will do that
08:38 egonw ok, so, one more unit test:
08:38 egonw CML23FragmentsTest.java
08:39 egonw where you input a CML string (no whitespace)
08:39 egonw parse the CML
08:39 olas how should it look like?
08:39 olas (in case it doesn't make the CML)
08:39 egonw and then assertEquals(12, mol.getAtom(0).getMassNumer());
08:39 egonw after a assertNotNull(mol.getAtom(0).getMassNumer());
08:39 egonw I will hack up the CML now
08:39 egonw hang on
08:39 olas ok
08:41 olas fo I need to explicitly set massnumber or just add an atom "C"?
08:41 egonw set it explicitly
08:44 egonw mmm... not 100% sure... but <atom @isotopeNumber> seems the way to do it...
08:45 olas should I do:
08:45 olas Molecule mol = new Molecule();     Atom atom = new Atom("C");     atom.setMassNumber( new Integer(12) );
08:45 olas or
08:45 olas IMolecule molecule = new Molecule();     molecule.addAtom(molecule.getBuil​der().newAtom(Elements.CARBON));
08:45 olas ?
08:45 egonw first is fine
08:45 olas ok
08:49 CIA-46 bioclipse: jonalv * r5773 /bioclipse2/trunk/plugins/net.biocli​pse.structuredb.tests/tests/net/bioc​lipse/structuredb/persistency/dao/ (LabelDaoTest.java StructureDaoTest.java): Added two failing tests catching broken sql related to the many-to-many relation between Structure and Label
08:49 egonw olas: [10:49] <CIA-31> cdk: egonw * r11251 /cdk/trunk/src/test/org/openscience​/cdk/io/cml/cml23TestFramework.xml: Added mass number CML example
08:49 egonw olas: now it's up to you
08:50 olas <atom id="a1" elementType="C" formalCharge="0" hydrogenCount="0"/>
08:50 olas after writing CML
08:50 egonw ok, so it is not written...
08:51 egonw ok, hang on
08:51 egonw will patch
08:51 egonw listin in:
08:51 egonw go to cdk.libio.cml
08:51 egonw in src/main
08:51 egonw Convertor.java
08:51 egonw cdkAtomToCMLAtom
08:52 egonw lines 631
08:52 egonw ff
08:52 egonw there is the code which is supposed to write it
08:53 egonw mass number=12 is lost, because it's the MN of the major isotope
08:53 olas hmm
08:53 egonw let's consider that a bug
08:53 olas I made the roundtriptest, and that actually passes!!
08:53 olas very very strange
08:54 egonw indeed...
08:54 egonw please commit that unit test
08:54 olas will do
08:55 olas committed
08:56 olas does the Roundtripper really write to file?
08:56 egonw no
08:57 egonw just to the CML format
08:57 egonw using a StringWriter, IIRC
08:57 olas ok
08:57 olas egonw: I'll leave it to you to fix the bug
08:57 olas I will commit my other unit tests too
08:58 CIA-46 bioclipse: jonalv * r5774 /bioclipse2/trunk/plugins/net.bioc​lipse.structuredb/src/sqlMap.xml: fixed broken sql for many-to-many relation between Structure and Label. All structuredb tests pass. Time to write some more! :)
08:59 olas jonalv++
09:04 egonw olas: please verify if conversion is working now
09:04 egonw btw, mass numbers might be unset...
09:04 egonw so, please commit the exception to that bug report too...
09:05 egonw because there seems to be other code that needs updating too
09:06 olas where should it be set?
09:09 egonw the MN?
09:09 olas yup
09:09 olas passes now
09:09 egonw not required to be set
09:09 olas what is the bug then?
09:09 egonw so, conversion is fixed?
09:09 olas hmm
09:09 olas will test in Bioclipse
09:09 olas in CDK tests, it passes now
09:09 egonw please add the stacktrace to the bug report
09:10 olas no
09:10 olas still no luck
09:11 olas ok
09:11 olas bug will be in bioclipse or cdk?
09:11 olas it's a bioclispe-problem IMO
09:11 egonw please try to get a stacktrace...
09:11 olas oh
09:11 olas I need a fix for CDK10
09:12 olas I convert with the writer in cdk10 and trhe reader in cdk
09:12 egonw ah... ok, so please apply my Convertor.java patch to the cdk10 branch then
09:13 * egonw really needs to go into deadline mode
09:13 egonw bbl
09:21 malick left #bioclipse
09:22 malick joined #bioclipse
09:35 CIA-46 bioclipse: goglepox * r5775 /bioclipse2/trunk/plugins/net.bioc​lipse.cdk10.jchempaint/src/net/bio​clipse/cdk10/jchempaint/ui/editor/ (JCPMultiPageEditorContributor.java JCPPage.java):
09:35 CIA-46 bioclipse: Copied registerModel() from JCPMultiPageEditorContributor to JCCPage.
09:35 CIA-46 bioclipse: Not nice but it seams to get the work done.
09:36 CIA-46 bioclipse: ospjuth * r5776 /bioclipse2/trunk/plugins/net.bioclipse.cdk.busine​ss/src/net/bioclipse/cdk/business/CDKManager.java: Use CML first, and secondly XMILES when createing CDKMolecule from IMolecule.
09:43 thomas_ku joined #bioclipse
09:49 olas egonw: FYI: solved the bug
09:50 CIA-46 bioclipse: ospjuth * r5777 /bioclipse2/trunk/plugins/ (4 files in 3 dirs): Conversion between CDK10 and CDK via CML fixed. Solves 1979421.
09:51 egonw olas: perfect
09:51 olas :-)
09:51 olas it looks good
09:51 egonw umm... your patch to cdk10
09:51 egonw should read:
09:51 egonw massNumber > 0
09:51 olas oh
09:51 egonw 0 is the default for unset
09:52 olas ok
09:52 olas will correct
09:52 egonw thanx
09:55 olas now: lunch
10:41 thomas_ku_ joined #bioclipse
11:18 masak joined #bioclipse
11:21 egonw bbl
11:22 Gpox_ joined #bioclipse
11:40 egonw joined #bioclipse
12:24 masak has anyone here extended EditorPart directly before?
12:59 olas joined #bioclipse
13:03 olas egonw: can I add labels to atoms in JCP?
13:05 egonw not sure
13:11 CIA-46 bioclipse: jonalv * r5778 /bioclipse2/trunk/plugins/ (4 files in 2 dirs): Added delete structure and delete label
13:11 olas like a LabelDecorator?
13:11 olas would be great to be able to present numbers next the atom, based on some calculation
13:50 olas egonw: does CDK support writing of mol2?
13:51 olas and does cdk support generation of Sybyl atom types?
13:53 egonw ah, you can show atom numbers
13:53 egonw that's a simple option in Renderer2DModel
13:53 egonw no Sybyl atom type perception yet
13:53 egonw bbl
13:53 olas save as mol2?
13:53 egonw does not save atom type
13:54 olas ok
13:54 olas reading of mol2 works with atom types?
13:55 olas what reader should I use?
13:55 olas MDLV2000Reader?
13:55 olas egonw?
14:03 masak anyone know how to add scrollbars to an editor?
14:04 masak (i.e. an EditorPart which is not a TextEditor)
14:07 * olas is using Mol2Reader, but with problems
14:12 egonw our mol2reader is not very good
14:12 egonw we cannot support mol2 format at this moment
14:16 malick left #bioclipse
14:22 masak ok, a more specific question:
14:23 masak SWT.H_SCROLL in a Canvas constructor gives that canvas a horizontal scrollbar
14:23 masak ...except it doesn't scroll the content
14:23 masak why?
14:24 CIA-46 bioclipse: jonalv * r5779 /bioclipse2/trunk/plugins/ (8 files in 4 dirs): wrote and updated test related to labeling and unlabeling structures. 3 tests doesn't pass. more work :)
15:29 CIA-46 bioclipse: jonalv * r5780 /bioclipse2/trunk/plugins/ (5 files in 4 dirs): All tests passes. Now removing of labels from structures get persisted again. Hopefully that was the last trouble related to the Structure - Label relation. (Probably not.)
15:47 thomas_ku joined #bioclipse
15:52 shk3 joined #bioclipse
15:53 shk3 hi all
15:53 jonalv hi shk3. How are things?
15:53 shk3 just started at the ebi
15:53 masak hello shk3!
15:53 shk3 got ubuntu running on my new laptop
15:53 jonalv shk3, ubuntu is nice :)
15:53 shk3 doing apt-get install eclipse right now
15:53 jonalv ah
15:54 shk3 hope to have more time for bioclipse soon
15:59 shk3 which eclipse version should I have for Bc2 development?
15:59 shk3 seems ubuntu has 3.2.2 - is that ok?
16:01 olas no
16:01 olas you need 3.3.x
16:01 shk3 :-(
16:01 olas welcome back btw!
16:01 olas use 32-bits
16:01 olas is a recommendation
16:02 shk3 any ubuntu expert here? Is a 3.3 installation possible via apt?
16:07 jonalv shk3: I usually just download the latest from eclipse's website
16:08 shk3 I will do that as well
16:08 jonalv it's nice to have things installed with apt but the tools that it's important to have the latest version of I have found one might have to do manually
16:11 shk3 no problem with eclipse really, just unpack...
16:14 CIA-46 bioclipse: carl_masak * r5781 /bioclipse2/trunk/plugins/net.bioclipse.ui/ (3 files in 2 dirs):
16:14 CIA-46 bioclipse: [net.bioclipse.ui]
16:14 CIA-46 bioclipse: * rudimentary alignment editor
16:14 CIA-46 bioclipse: ** reads fasta files with multiple sequences
16:14 CIA-46 bioclipse: ** no scroll (couldn't figure out how)
16:14 CIA-46 bioclipse: ** no zoom (yet)
16:14 masak and that concludes the day
16:14 masak ta ta
16:51 shk3 ok, will leave for today. hope I can start soon working on stuff
16:57 olas left #bioclipse
21:24 edrin joined #bioclipse
21:27 thomas_ku joined #bioclipse

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