Camelia, the Perl 6 bug

IRC log for #cdk, 2007-09-28

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

All times shown according to UTC.

Time Nick Message
05:50 egonw joined #cdk
06:08 egonw bbl
06:19 jbrefort joined #cdk
07:21 egonw Oh, today will be the day for the 9000th commit to the CDK repository...
07:34 CIA-31 bioclipse: jonalv * r3858 /trunk/playground/plugins/net.bioclipse.keyring/ (.classpath META-INF/MANIFEST.MF): Added JUnit to manifest
07:35 jonalv joined #cdk
07:36 CIA-31 bioclipse: jonalv * r3859 /trunk/playground/plugins/​net.bioclipse.structuredb/ (4 files in 4 dirs): Removed ',' in Structure.sql and changed around with classpath stuff
07:37 egonw jonalv: about r3858
07:38 egonw you should better make a net.bioclipse.keyring.test plugin for the JUnit tests
07:39 jonalv egonw, hm that's an idea. Currebntly it's in its own source map. so there is source and tests
07:39 jonalv Then I can have the same pavkage name and access package protected stuff from the tests. But maybe that can be done from another project aswell?
08:04 egonw yes, guess so
08:06 jonalv but that wouldn't be very nice?
08:08 egonw why not?
08:10 jonalv oh maybee it would
08:11 jonalv But why so manu plugins?
08:11 egonw because it will make the downloads smaller
08:11 jonalv which downloads?
08:12 egonw update site, bioclipse core
08:12 egonw moreover...
08:12 jonalv Ah but the tests should not be in the build
08:12 egonw it makes it more clear which plugins have tests
08:12 jonalv that's why they are in their own package..
08:12 egonw no, but the plugin would still depend on JUnit
08:12 jonalv yes that's evil.
08:13 jonalv But it's som much refactoring chaning it now... :(
08:13 jonalv :)
08:14 egonw just copy/paste the plugin and delete stuff
08:15 jonalv Yes I will stat with structuredb and then do Keyring when I do the redesign of it that is due...
08:15 jonalv wait I can copy paste the entire plugin?
08:16 egonw in SVN terms...
08:16 egonw sure
08:16 egonw mom... shall I do that?
08:17 egonw or shall I explain you what to do?
08:17 egonw that would be more informative
08:17 jonalv yea explain
08:17 egonw ok, got a command line open?
08:17 egonw cd bla/bioclipse/trunk
08:17 jonalv copy paste in Eclipse didn't really do the trick because how do I repoint the svn stuff for the neew project?
08:17 jonalv wow
08:18 jonalv okey I will need to do an update to my trunk checkout
08:18 egonw your machine does have a command line, correct?
08:18 egonw what OS are you running?
08:18 jonalv my eclipse svn and my command prompt svn isn't the dsame version...
08:18 jonalv Ubuntu
08:18 egonw feisty?
08:19 jonalv yes I think so
08:19 jonalv but I have a checkout of trunk which I used for some time travelling earlier on... :)
08:19 egonw OK, do this:
08:20 egonw svn co -N bla/trunk
08:20 jonalv Needed to revert a commit and eclipse was very broken...
08:20 jonalv what's -N?
08:20 egonw non-recursive
08:20 egonw works also for update...
08:20 egonw mom... brb
08:20 jonalv but I am updated on enterire trunk now
08:21 egonw OK, if your inet links allows that
08:21 jonalv Well it took a while to download all of it the first time but just updating is quite ast...
08:22 egonw indeed
08:22 jonalv but I guess I can simple do cp -R for the plugin and then commit it right?
08:22 egonw I have a full check out too
08:22 egonw no
08:22 egonw better to:
08:22 egonw cd plugins
08:22 jonalv or something like that
08:23 egonw umm
08:23 egonw playground right...
08:23 egonw cd ../playground/plugins/
08:23 egonw svn copy net.bioclipse.keyring net.bioclipse.keyring.test
08:23 egonw then, *very important* ...
08:23 jonalv ah yes
08:23 egonw cd net.bioclipse.keyring.test
08:24 egonw edit the .project file
08:24 egonw and change the <name> field
08:24 jonalv hold on I fumbled...
08:26 jonalv yes what do I do in the .project file?
08:26 egonw change the <name> field
08:26 jonalv change name...
08:26 egonw right
08:26 jonalv woulf that be all that needs to be changed in it?
08:26 egonw also edit the META-INF/MANIFEST.MF
08:26 egonw to add the dependency on net.plugins.keyring
08:27 egonw then commit to SVN, import in Eclipse
08:27 egonw and the deleting etc can then be done in Eclipse
08:27 jonalv that's under required bundle right?
08:27 egonw yes, think so
08:28 jonalv hm there is a lot of things in this file that should be changed
08:28 egonw yes, but nothing that cannot be done in Eclipse
08:29 jonalv but the dependency can be addad in Eclipse aswelll right?
08:29 egonw yes
08:30 jonalv what's up with net.bioclipse.cdk.progz/src/org?
08:30 jonalv It turns up as ? in svn st
08:30 egonw that is/was a svn import...
08:30 egonw nifty stuff...
08:30 jonalv mhm
08:30 egonw automatically retrieved from CDK SVN...
08:30 egonw it's a SVN trick
08:31 egonw learned that from alexandr...
08:31 egonw neksa.blogspot.com
08:32 jonalv sounds nice, I think masak is doing something like that with his svn repo of his work catalogue when some work related stuff shuold be in another svn reop...
08:33 CIA-31 bioclipse: jonalv * r3860 /trunk/playground/plugins/n​et.bioclipse.keyring.tests/ (. .project): created net.bioclipse.keyring.tests to house the keyring tests
08:34 jonalv let's see if this worked correctly...
08:40 olas joined #cdk
09:12 CIA-31 joined #cdk
09:15 jpansanel moin
09:15 masak joined #cdk
09:20 CIA-31 cdk: egonw * r8978 /trunk/cdk/src/org/openscience/cdk/test/appl​ications/undoredo/AddHydrogenEditTest.java: Fixed test, after HydrogenAdder removal commit: use private methods to track changes.
09:21 CIA-31 bioclipse: jonalv * r3863 /trunk/playground/plugins/net​.bioclipse.structuredb.tests/ (. .project): copied net.bioclipse.structuredb to net.bioclipse.structuredb.tests
09:36 CIA-31 bioclipse: jonalv * r3864 /trunk/playground/plugins/​net.bioclipse.structuredb/ (5 files in 4 dirs): Removed all test related stuff from net.bioclipse.structuredb
09:38 CIA-31 bioclipse: jonalv * r3865 /trunk/playground/plugins/net​.bioclipse.structuredb.tests/ (8 files in 4 dirs): Removed all non test related stuff from net.bioclipse.structuredb.tests
09:48 thomas_ku joined #cdk
09:48 thomas_ku moin
10:49 neksa joined #cdk
11:05 egonw he neksa!
11:11 neksa hi egonw!
11:12 olas joined #cdk
12:22 jonalv Why does CMLReader.read take an IChemObject? I want to read a chemobject from the inputStream how would I do that?
12:30 jonalv egonw, would you know?
12:30 carsten joined #cdk
12:50 masak jonalv: CMLReader.read takes as a parameter the IChemObject that it should provide with information
12:52 steinbeck joined #cdk
12:56 jonalv hmpf
12:56 CIA-31 cdk: rajarshi * r8979 /trunk/cdk/src/org/openscience/cdk/d​ict/data/descriptor-algorithms.owl: Updted to include an entry for the CTYPES descriptor
12:58 CIA-31 cdk: rajarshi * r8980 /trunk/cdk/src/org/openscience/cdk/qsar/des​criptors/molecular/ChiPathDescriptor.java: Fixed the names, since the VP descriptors were going from 8 to 15 rather than 0 to 7
13:15 jonalv I am trying to read CML but get: org.openscience.cdk.exception.CDKException: Only supported is reading of ChemFile objects.
13:15 jonalv at org.openscience.cdk.io.CMLRea​der.read(CMLReader.java:200)
13:15 jonalv at net.bioclipse.structuredb.domain.Struct​ure.setMoleculeCML(Structure.java:222)
13:15 jonalv What am I doing wrong?
13:17 egonw you need to pass it a IChemObject... an empty one
13:17 CIA-31 bioclipse: jonalv * r3866 /trunk/playground/plugins/ne​t.bioclipse.structuredb/src/ (3 files in 3 dirs): Attempts to persist the molecule as CML fails strangely. I suspect I am using the CDK-api in the wrong way...
13:17 jonalv but I pass it a molecule
13:17 jonalv Isn't that an IChemobject?
13:18 jonalv IchemObject is an interface right?
13:18 * jonalv finds this confusing...
13:18 egonw sorry... pass it a IChemFile, for example a ChemFile
13:19 jonalv will I get an atomcontainer then?
13:20 jonalv I find it strange to pass things into something that I would expect creates things...
13:20 egonw no, you will get a IChemFile
13:21 egonw yes, but the IO classes need to know which implementation of the core classes...
13:21 jonalv hm sounds complicated
13:21 egonw it would be better to pass them a IChemObjectBuilder...
13:21 egonw but I need to change the API for that...
13:22 jonalv yes or ahve different readMethods producing the different implementations that are supported
13:22 egonw would make more sense... but that did not yet exist when the IO API was designed
13:22 jonalv instead of throwing excpetions for those that are not supported...
13:22 egonw jonalv: no, that would be ugly
13:22 jonalv why?
13:22 egonw one method for each implementation would be very ugly:
13:22 jonalv I find it more ugly with Exceptions that get's thrown
13:22 egonw it would require a fix list
13:22 egonw does not allow custom impls
13:23 egonw would add dependencies on all those implementations
13:23 jonalv yes that's true of course...
13:23 egonw jonalv: that's what the accepts() method is for
13:23 jonalv there is an accept method?
13:23 egonw yes
13:24 CIA-31 cdk: rajarshi * r8981 /trunk/cdk/src/org/openscience/cdk/ (3 files in 3 dirs): Added the CTYPES descriptor associatd unit test
13:24 jonalv I think what troubles me the most is that when using net.bioclipse.cdk I don't get any javadoc...
13:24 egonw that's available from:
13:24 egonw http://cheminfo.informatics.indiana.​edu/~rguha/code/java/nightly-1.0.x/
13:24 jonalv the nightly build?
13:25 egonw for the 1.0.x branch, yes
13:25 jonalv But that is not the same version is it?
13:25 egonw it is more or less
13:25 egonw only bug fixes go into the branch
13:25 egonw the API should be 99.99% the same
13:25 jonalv hm is it impossible to get it automagicly from with the Eclipse plugin?
13:26 egonw no idea
13:29 jonalv no I get a Found well-formedness error in line 1 error instead
13:30 egonw on what?
13:30 CIA-31 bioclipse: ospjuth * r3867 /trunk/bioclipse2/:
13:31 jonalv When I try reading the cml I wrote
13:32 CIA-31 bioclipse: ospjuth * r3868 /trunk/bioclipse2/testing/:
13:32 egonw ic
13:32 egonw that's a problem with your CML file...
13:34 jonalv It looks like this:
13:34 jonalv <?xml version="1.0" encoding="ISO-8859-1"?>
13:34 jonalv <molecule id="m1" xmlns="http://www.xml-cml.org/schema"/>
13:34 CIA-31 bioclipse: ospjuth * r3869 /trunk/bioclipse2/testing/testCNF/: Initial import of an RCP application for 3.3 with a Common Navigator Framework (CNF) implementation. Utilizes ui.ide, and ui.editors for now.
13:34 egonw jonalv: that's all?
13:34 CIA-31 bioclipse: ospjuth * r3870 /trunk/bioclipse2/testing/testCNF/ (22 files in 6 dirs): Initial import of an RCP application for 3.3 with a Common Navigator Framework (CNF) implementation. Utilizes ui.ide, and ui.editors for now.
13:34 jonalv egonw, yes strange isn't it?
13:35 egonw yes, it looks fine
13:35 jonalv egonw, would you like to have a look at my code?
13:35 egonw sure
13:35 CIA-31 bioclipse: ospjuth * r3871 /trunk/bioclipse2/testing/testCNF/: Added bin to svn.ignore
13:35 jonalv I'll just commit the last...
13:36 CIA-31 bioclipse: jonalv * r3872 /trunk/playground/plugins/net.biocl​ipse.structuredb/src/net/bioclipse/​structuredb/domain/Structure.java: Should use ChemFile instead of Molecule when reading CML
13:37 jonalv so it's net.bioclipse.strucutredb and net.bioclipse.structuredb.tests
13:37 jonalv You will also need net.bioclipse.hsqldb
13:37 jonalv All in playground
13:41 jonalv egonw?
13:42 egonw mom
13:42 CIA-31 cdk: rajarshi * r8982 /trunk/cdk/src/org/openscience/cdk/test/​smiles/smarts/parser/RecursiveTest.java: Updated comments since everything passes with the exception of the first one, since the atom type for Po is undetected
13:47 egonw olas: why not put bioclipse2/ in branches/ instead of trunk/ ?
13:53 CIA-31 bioclipse: egonw * r3873 / (branches/bioclipse2/ trunk/bioclipse2/): Moved the new Bioclipse branch (v2) into branches/
13:54 CIA-31 bioclipse: biocoder * r3874 /trunk/playground/plugins/net.bioclipse.chart/ (14 files in 7 dirs):
13:54 CIA-31 bioclipse: Functionality for making scatter plots from MatrixGridEditor now working.
13:54 CIA-31 bioclipse: Select two or more columns and choose scatter plot in the context menu.
13:55 CIA-31 bioclipse: biocoder * r3875 /trunk/plugins/net.bioclipse.statistics/ (2 files in 2 dirs): Made an action for scatter plotting
14:09 CIA-31 bioclipse: ospjuth * r3876 /branches/bioclipse2/testing/testCNF/ (4 files in 2 dirs): Fixed product so that plugin runs from generated launch config.
14:14 olas left #cdk
15:12 CIA-31 bioclipse: egonw * r3877 /trunk/playground/plugins/net.biocl​ipse.structuredb/src/net/bioclipse/​structuredb/domain/Structure.java: Fixed reading of CML content.
17:06 CIA-31 bioclipse: egonw * r3878 /trunk/playground/plugins/net.bioclipse.cha​rt/src/net/bioclipse/chart/ChartUtils.java: Cleaned up imports, removing one sun.x.y import not found in my classpath.
22:25 CIA-31 joined #cdk
22:46 neksa_ joined #cdk
23:45 CIA-31 cdk: rajarshi * r8983 /trunk/cdk-qa/projects/070928-0001/: Added projcet dir for atom type qa
23:58 CIA-31 cdk: rajarshi * r8984 /trunk/cdk-qa/projects/070​928-0001/AtomTypeQA.java: Added atom type qa code

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