Camelia, the Perl 6 bug

IRC log for #cdk, 2008-05-19

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

All times shown according to UTC.

Time Nick Message
05:53 ilbot2 joined #cdk
05:53 Topic for #cdkis now The Chemistry Development Kit (http://cdk.sf.net) | JUnit: 6043, 64 fails + 59 errors (2.04%) | CDK 1.0.x TODO: see http://cdk.sf.net/wiki/ | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News paper online on pharmacophores: http://www.cdknews.org/
06:16 jbrefort joined #cdk
06:17 carsten joined #cdk
06:27 rojasm joined #cdk
06:44 jpansanel moin
06:49 malick joined #cdk
06:58 rojasm moin
07:04 CIA-31 cdk: miguelrojasch * r11020 /cdk/branches/miguelrojasch-IP/ (2 files in 2 dirs): added citation
07:30 thomas_ku joined #cdk
07:35 CIA-31 cdk: miguelrojasch * r11021 /cdk/branches/miguelrojasch-IP/src/ (8 files in 5 dirs): if containers are not found in resonance search return null
07:35 thomas_ku moin
07:42 CIA-31 cdk: miguelrojasch * r11022 /cdk/branches/miguelrojasch-IP/src/main/org/opensc​ience/cdk/tools/StructureResonanceGenerator.java: Removed smiles import.
07:45 jonalv joined #cdk
08:53 rojasm joined #cdk
09:12 rojasm joined #cdk
09:54 rojasm joined #cdk
09:58 rojasm joined #cdk
10:16 rojasm joined #cdk
10:34 malick joined #cdk
10:34 rojasm joined #cdk
11:56 CIA-31 cdk: miguelrojasch * r11023 /cdk/branches/miguelrojasch-IP/src/ (2 files in 2 dirs): added method to extract the resonance container given a bond
12:35 egonw joined #cdk
13:00 JimDowning joined #cdk
13:00 JimDowning Hi there.
13:00 JimDowning Is there a list of the jar versions around anywhere?
13:01 jonalv joined #cdk
13:10 egonw ha JimDowning
13:10 egonw of cdk/jar's, I assume?
13:10 egonw if so: no
13:10 egonw though the svn commit messages often describe which version
13:11 JimDowning Hmmm.
13:11 JimDowning Can you remember many?
13:11 egonw no, but some should work :)
13:11 egonw ok, which ones do you want to know?
13:12 JimDowning Baysmith IO?
13:12 egonw not used
13:12 JimDowning Ah. That's easy then :-)
13:12 egonw I think
13:12 JimDowning Batik?
13:12 egonw JimDowning: regarding being used or not:
13:13 egonw grep baysmith-io.jar src/META-INF/*
13:13 * egonw is checking batik
13:13 egonw Batik: 1.5.1
13:13 egonw quite ancient
13:14 JimDowning Dadml?
13:14 egonw not sure if I ever gave that a version number :)
13:14 egonw neat concept, never really took off
13:14 egonw dadml: 1.2
13:14 egonw (using svn log)
13:15 JimDowning jai classes not used, it seems?
13:16 egonw seems so...
13:16 JimDowning jfontchooser v?
13:16 egonw not used currently
13:16 egonw because JChemPaint 2.x has been removed from trunk
13:16 JimDowning Listed as dep in applications...
13:16 JimDowning Ah.
13:17 JimDowning JRI v?
13:17 egonw yeah, I really need to clean up those deps...
13:17 egonw ah, JRI...
13:17 egonw let me check
13:17 JimDowning I've guessed at 0.2.4
13:17 egonw comes from rJava 0.5
13:18 JimDowning Xindice v?
13:19 egonw cool :) you know, that's YY's work...
13:19 egonw you're looking at some old stuff from Peter's group :)
13:19 JimDowning :-) Yes, I saw his id on the last version!
13:19 egonw you've seen Bob?
13:19 JimDowning No?
13:19 JimDowning WHich bob?
13:19 egonw Bob was the WWMM server, I think...
13:19 egonw maybe the client :)
13:20 egonw one of the components of the earlier WWMM
13:20 egonw CDK had client code for talk to 'Bob'
13:20 egonw I think it's dead
13:20 JimDowning :-) if it isn't it probably needs finishing off
13:20 egonw ah, those good old days...
13:20 egonw I really had fun working with YY and Juergen
13:21 egonw yes, agreed
13:21 egonw so, don't worry about the version number
13:21 * egonw started cleaning up jar/ and src/META-INF/*.libdepends
13:23 JimDowning Our usual hack is to add them as "[foo]-expediency"
13:23 * egonw gets a EN-NL dictionary :)
13:23 JimDowning Egon, thanks for the help. I'll update tomorrow to pick up your changes to lib-depends and get back in touch.
13:23 JimDowning ttfn
13:23 JimDowning left #cdk
13:35 Andreas1 joined #cdk
13:35 Andreas1 hi
13:36 egonw hi Andreas1
13:37 Andreas1 Hey egonw, did you have the chance to review my new patch to the atom typer?
13:37 CIA-31 cdk: egonw * r11024 /cdk/trunk/ (11 files in 5 dirs): Removed SVG writer and Batik libraries: version of the last one was ancient; someone suggested a better library; and the code was using the deprecated 2D renderer anyway; the functionality will return
13:37 egonw Andreas1: no
13:37 egonw good of you to ping me
13:37 egonw must have missed it...
13:38 Andreas1 I posted it as a comment to the bug report.
13:38 egonw mmm... can't find it in my inbox
13:38 egonw ah...
13:38 Andreas1 For some reason my university e-mail server has a problem with sourceforge.net. This is way I can't post to the mailinglists right now... :-(
13:38 Andreas1 way=why
13:38 egonw oh :(
13:39 egonw int electronCount = ringSize + 1
13:39 egonw that assumes all rings have a pz
13:39 egonw that should include a check for that...
13:39 egonw I think there is a method along the lines: allRingAtomsAreSP2
13:39 egonw or so...
13:40 egonw a second thing...
13:40 egonw && atom.getHydrogenCount() !=
13:40 egonw CDKConstants.UNSET && atom.getHydrogenCount() == 1) {
13:40 egonw that does not take into account an explicit hydrogen
13:40 egonw yet
13:41 Andreas1 ok, let me explain
13:42 egonw sure
13:42 Andreas1 isHueckelNumber(ringSize + 1) detects 5-rings, 9-rings, etc.
13:43 Andreas1 if it is a 5-ring AND a nitrogen AND the nitrogen has a hydrogen attached THEN it must be planar3
13:43 Andreas1 otherwise sp2
13:43 Andreas1 explicit hydrogens are accounted for as well
13:44 Andreas1 here: if (atomContainer.getConnectedAtomsCount(atom) == 3) {
13:44 Andreas1 this comes above my code and is part of your original code
13:44 egonw ok, good
13:45 egonw is there also already a check above for all-ring-atoms-are-sp2?
13:45 egonw which might, since we discriminate between planar and sp2 only
13:45 * egonw is opening the .java file :)
13:46 Andreas1 you refer to the case where the N is sp2, but not all ring atoms are sp2 as well... I
13:46 Andreas1 I'll check
13:46 carsten joined #cdk
13:46 egonw correct...
13:46 CIA-31 cdk: egonw * r11025 /cdk/trunk/ (4 files in 2 dirs): Removed obsolete library
13:47 egonw so, I should have said instead:
13:47 egonw is there also already a check above for all-other-ring-atoms-are-sp2?
13:47 Andreas1 I don't think so, right now there is this: if (isRingAtom && bothNeighborsAreSp2(atom, atomContainer)) {
13:47 Andreas1 so only the neighbours are checked
13:47 egonw ok, good enough for me
13:47 egonw yeah, less expensive
13:48 egonw ok, final question
13:48 Andreas1 I see
13:48 egonw does it make unit tests fail? :)
13:48 egonw if not, please apply to trunk/
13:48 egonw at some point, this code needs to be replaced with better code...
13:49 egonw Martin Ott, formerly at Radboud University Nijmegen, now in Sheffield (IIRC), had a really nice approach...
13:49 egonw but never took the time to implement it...
13:49 Andreas1 uh, this is a hard question to answer. How do I check whether tests fail due to the code change?
13:49 egonw ummm...
13:49 egonw ant -logfile ant.log test-all
13:49 egonw and compare the results with those of before the patch
13:49 egonw (or, against Nightly)
13:50 Andreas1 ok, will do. Will take some time though...
13:50 egonw I know
13:51 Andreas1 the patch actually fixes another failing test case of the smiles parser test
13:51 egonw yeah, and I expect some fixes for the SMARTS module too
13:52 egonw which actually uses some structures involving hetero ring systems with nitrogens
13:54 Andreas1 there is another thing with atom types
13:54 Andreas1 there is no "N.minus.planar3"
13:55 egonw two singel-bonded neighbors, a LP as third orbital in the plane, and a LP in the pz?
13:55 Andreas1 probably not a frequent thing to encounter, but necessary for pyrrole anion "[n-]1cccc1"
13:55 egonw right
13:56 Andreas1 yeh
13:56 egonw ok, let me check...
13:57 egonw yeah, seems missing
13:58 Andreas1 now then we would need to add aromatic ring checking code in the "} else if (atom.getFormalCharge() == -1) {" block as well...
13:58 egonw yes, indeed
13:58 egonw lot's of room for optimization
13:58 egonw but that's not the important bit now...
13:59 egonw the functionality is at this stage of trunk/
13:59 egonw Andreas1: would you like to add this atom type?
13:59 Andreas1 I don't get it?
13:59 Andreas1 sure
14:00 Andreas1 I think getting those aromaticity things working is quite important as it would fix some SmilesParser issues
14:00 egonw agreed
14:01 Andreas1 and I still believe that SMILES is the most handy molecular format around
14:01 egonw it's handy indeed
14:01 Andreas1 that's why I'm pushing forward a little to fix those things
14:01 egonw with OpenSMILES, it becomes easier to implement too
14:02 egonw because it's implicit things are better defined
14:02 Andreas1 what is OpenSMILES, never heard of it?
14:02 egonw opensmiles.org
14:02 egonw a Blue Obelisk project to accurately define the SMILES standard
14:04 carsten Moin egonw
14:04 egonw hi carsten
14:04 carsten Avogadro 0.8 is out, very nice improvements.
14:04 egonw yeah, would still love to see ubuntu .deb's
14:05 Andreas1 Interesting egonw, good to know about OpenSMILES
14:05 carsten depends on OB 2.2 so I guess it will take until OB 2.2-final is out
14:07 CIA-31 cdk: egonw * r11026 /cdk/trunk/src/main/org/openscienc​e/cdk/applications/swing/editor/: Removed more old JChemPaint 2.x code
14:08 CIA-31 cdk: egonw * r11027 /cdk/trunk/ (jar/jfontchooser.jar src/META-INF/applications.libdepends): Removed another obsolete library
14:19 CIA-31 cdk: egonw * r11028 /cdk/trunk/ (build.xml jar/jai_codec.jar jar/jai_core.jar): Removed unused library
14:31 CIA-31 cdk: egonw * r11029 /cdk/trunk/ (5 files in 4 dirs): Removed more dead wood: DADML, a library for accesing web resource, before the days of SOAP/BioMoby/etc; Xindice code for WWMM support, which is no longer functional
14:31 egonw bbl
14:31 egonw doing some shopping
14:32 CIA-31 cdk: egonw * r11030 /cdk/trunk/jar/biojava-1.4.jar: Removed unused and outdated library
14:33 CIA-31 cdk: egonw * r11031 /cdk/trunk/jar/ (mysql-2.0.4.jar postgresql7.0-1.2.jar): Removed unused and outdated JDBC libraries
14:55 CIA-31 cdk: thomaskuhn * r11032 /cdk-taverna/trunk/src/main/java/org/opensc​ience/cdk/applications/art2aClassification/ (5 files): Add an implementation of the ART2A Classification Algorithm
14:56 CIA-31 cdk: thomaskuhn * r11033 /cdk-taverna/trunk/src/test/java/org/opensc​ience/cdk/applications/art2aClassification/ (. TestArt2aClassificator.java TestArt2aCollection.java): Add unit test for the implementation of the ART2A classification algorithm
14:57 CIA-31 cdk: thomaskuhn * r11034 /cdk-taverna/trunk/src/main/java/or​g/openscience/cdk/applications/tave​rna/qsar/model/ART2AClassificator/ (. ART2AClassificator.java): Add a worker which uses the new implementation of the ART2A classificator
14:58 CIA-31 cdk: thomaskuhn * r11035 /cdk-taverna/trunk/src/test/java/or​g/openscience/cdk/applications/tave​rna/qsar/model/ART2AClassificator/ (. ART2AClassificatorTest.java): Add an unit test for the worker which uses the new implementation of the ART2A classificator
14:59 CIA-31 cdk: thomaskuhn * r11036 /cdk-taverna/trunk/src/test/r​esources/data/IrisFlowerData/ (6 files): Add test data for the ART2A classification
15:01 CIA-31 cdk: egonw * r11037 /cdk/trunk/src/main/org/openscience/cdk/libio/ (biojava/ openbabel/ weka/): Mmmm... apparently, BioJava was used, but the library was outdated (was 1.4 in our trunk/, while upstream is at 1.6); therefore, still removing the Convertor
15:01 CIA-31 cdk: thomaskuhn * r11038 /cdk-taverna/trunk/src/main/java/org/openscienc​e/cdk/applications/taverna/LocalWorkerCDK.java: Add a fingerprint item to the possible input and output port typse
15:02 CIA-31 cdk: thomaskuhn * r11039 /cdk-taverna/trunk/src/main/java/or​g/openscience/cdk/applications/tave​rna/atomtype/PerceiveAtomType.java: Rename the AtomType perceiver
15:02 CIA-31 cdk: thomaskuhn * r11040 /cdk-taverna/trunk/src/main/java/org​/openscience/cdk/applications/tavern​a/atomtype/AtomTypePerceptor.java: Remove after renaming
15:03 CIA-31 cdk: miguelrojasch * r11041 /cdk/branches/miguelrojasch-IP/src/ (14 files in 8 dirs): Added dependencies, added model IP prediction for bonds.
15:05 CIA-31 cdk: thomaskuhn * r11042 /cdk-taverna/trunk/src/main/java/org/op​enscience/cdk/applications/taverna/data​base/pgchem/GetQSARVectorFromDB.java: Add a worker which loads qsar results from a database and converts them to a fingerprint item. This can be used as input for the ART2A classification worker
15:06 CIA-31 cdk: thomaskuhn * r11043 /cdk-taverna/trunk/src/main/java/org/openscience/​cdk/applications/taverna/io/TextFileReader.java: Add a text file reader worker
15:06 CIA-31 cdk: thomaskuhn * r11044 /cdk-taverna/trunk/src/main/java/org/​openscience/cdk/applications/taverna/​io/action/TextFileReaderAction.java: Add an action for the text file reader worker
15:08 CIA-31 cdk: thomaskuhn * r11045 /cdk-taverna/trunk/src/main/java/org/opensc​ience/cdk/applications/taverna/tools/chebi/ (3 files):
15:08 CIA-31 cdk: Add two worker to extract the structures from the CHEBI database. This database
15:08 CIA-31 cdk: provides its data as tsv (tab separated value) file from the website:
15:08 CIA-31 cdk: ftp://ftp.ebi.ac.uk/pub/databases/chebi​/Flat_file_tab_delimited/structures.tsv
15:09 CIA-31 cdk: thomaskuhn * r11046 /cdk-taverna/trunk/src/test/java/org/opensc​ience/cdk/applications/taverna/tools/chebi/ (. ExtractDataFromChebiStructureTSVTest.java): Add a unit test for the extraction of structures of the CHEBI database
15:10 CIA-31 cdk: thomaskuhn * r11047 /cdk-taverna/trunk/src/test/resources/data/tsv/ (. chebistructures.tsv): Add a test file which represents a subset of the CHEBI database.
15:11 CIA-31 cdk: thomaskuhn * r11048 /cdk-taverna/trunk/src/main/resources/ (3 files in 2 dirs): Add the new created worker to the configuration files
15:12 CIA-31 cdk: thomaskuhn * r11049 /cdk-taverna/trunk/pom.xml: Set the used Java version to 1.6 because the ART2A algorithm uses some of its new functionality
15:20 CIA-31 cdk: egonw * r11050 /cdk/trunk/jar/bcel-5.1.jar: Unused library
15:20 CIA-31 cdk: egonw * r11051 /cdk/trunk/src/ (2 files in 2 dirs): Obsolete class
15:22 CIA-31 cdk: egonw * r11052 /cdk/trunk/jar/xmldb.jar: Unused library
15:26 CIA-31 cdk: egonw * r11053 /cdk/trunk/.classpath: Updated .classpath for library clean up
15:48 malick left #cdk
15:50 CIA-31 cdk: egonw * r11054 /cdk/trunk/src/META-INF/ (6 files): Some more deps clean ups
16:04 CIA-31 cdk: egonw * r11055 /cdk/trunk/jar/libio/: Removed an empty and obsolete dir
16:06 CIA-31 cdk: archvile18 * r11056 /cdk/trunk/src/main/org/openscience/c​dk/atomtype/CDKAtomTypeMatcher.java: Fix for bug [1960990] "Imidazole N.sp2 is typed N.planar3".
16:13 Andreas1 egonw, why is the cdk:piBondCount for N.planar3 zero?
16:16 egonw because it does not have any double bonds
16:16 Andreas1 ah, ok. So I set the piBondCount for N.minus.planar3 to zero as well
16:16 egonw right
16:17 Andreas1 ok, thanks
16:17 egonw I'
16:18 egonw I'm aware that I have not used the most perfect terminology...
16:18 egonw e.g. neighbor count should really be coordination number
16:18 * egonw is away doing the dishes
16:18 Andreas1 hehe, no problem, as long as we use it the same way
16:19 CIA-31 cdk: archvile18 * r11057 /cdk/trunk/src/main/org/openscience​/cdk/config/data/cdk_atomtypes.xml: Added atom type "N.minus.planar3" as in pyrrole anion [n-]1cccc1.
16:26 egonw Andreas1: please make sure to add a unit test too
16:35 Andreas1 on it's way :)
16:49 Andreas1 egonw: what's the difference between charge and formalCharge?
16:50 egonw the former is a partial charge
16:50 Andreas1 I see
16:51 * egonw has to leave and catch a train
16:52 Andreas1 cu later
17:16 CIA-31 cdk: archvile18 * r11058 /cdk/trunk/src/ (2 files in 2 dirs): Added test case for new atom type "N.minus.planar3".
17:45 CIA-31 cdk: archvile18 * r11059 /cdk/trunk/src/test/org/openscience/cdk​/atomtype/CDKAtomTypeMatcherTest.java: Replaced uses of "Molecule" with "IMolecule".
20:07 rojasm joined #cdk
20:44 egonw joined #cdk
20:49 CIA-31 cdk: egonw * r11060 /cdk/trunk/jar/: Ignore InChI .so libs
20:51 CIA-31 cdk: egonw * r11061 /cdk/trunk/src/main/org/openscience​/cdk/renderer/Java2DRenderer.java: Deal with the new null defaults
20:52 CIA-31 cdk: egonw * r11062 /jchempaint/trunk/src/main/org/openscience/ (11 files in 3 dirs): Close to having a basic Swing editor online
20:53 CIA-31 cdk: egonw * r11063 /cdk/trunk/src/test/org/openscience/cdk/ (20 files in 8 dirs): Java 1.5 seems to complain about assertEquals(Object, Object), which can be solved by forcing a assertEquals(native, native), e.g. assertEquals(int, int)
20:56 CIA-31 cdk: egonw * r11064 /cdk/trunk/ (5 files in 2 dirs): Reworked Eclipse settings, to define a org.openscience.cdk plugin, on which other plugins can extend (viz. org.openscience.jchempaint, org.openscience.cdk.gek (GPL Extension Kit), etc)
20:57 CIA-31 cdk: egonw * r11065 /jchempaint/trunk/ (5 files in 2 dirs): Defined the org.openscience.jchempaint plugin for Eclipse, based on the org.openscience.cdk plugin
21:09 CIA-31 cdk: egonw * r11066 / (2 files in 2 dirs): Moved application class
21:17 CIA-31 cdk: egonw * r11067 /jchempaint/trunk/src/main/​org/openscience/jchempaint/ (JCPPropertyHandler.java SomeToolBar.java): The new editor now runs, see http://chem-bla-ics.blogspot.com/
22:11 CIA-31 cdk: rajarshi * r11068 /cdk/trunk/src/test/org/openscience/cdk/sm​iles/smarts/parser/SMARTSSearchTest.java: Added test cases for bug 1967468, which is reporting a worng match for ring methylenes in an acyclic urea derivative
22:22 CIA-31 cdk: rajarshi * r11069 /cdk/trunk/src/test/org/openscience/cdk/sm​iles/smarts/parser/SMARTSSearchTest.java: Fxied some file loading issues. Also added a version of the SMARTS that does seem to solve the acyclic urea issue. However the original problem still remains, since Daylight parses it OK
22:24 CIA-31 cdk: rajarshi * r11070 /cdk/trunk/src/test/org/openscience/cdk/sm​iles/smarts/parser/SMARTSSearchTest.java: Minor update to the alternative SMARTS pattern to make it more specific
23:27 CIA-31 cdk: miguelrojasch * r11071 /cdk/branches/miguelrojasch-IP/src/ (4 files in 3 dirs): added dependencies with IPBondLearningDescriptor
23:42 CIA-31 cdk: miguelrojasch * r11072 /cdk/branches/miguelrojasch-IP/ (69 files in 29 dirs): svn commit -m "Synchronized with trunk the branches.miguelrojasch-IP,Rev=10996:11076"

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