Camelia, the Perl 6 bug

IRC log for #cdk, 2010-07-27

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

All times shown according to UTC.

Time Nick Message
03:01 azeem joined #cdk
03:19 s9asad joined #cdk
04:02 s9asad joined #cdk
04:02 azeem joined #cdk
04:02 slyrus joined #cdk
04:02 bag joined #cdk
04:02 s_wolf joined #cdk
04:02 mgerlich joined #cdk
04:02 Conrad joined #cdk
04:02 CIA-52 joined #cdk
04:02 jpansanel joined #cdk
04:02 stain joined #cdk
04:05 s_wolf joined #cdk
04:05 jpansanel joined #cdk
04:05 stain joined #cdk
04:08 s_wolf joined #cdk
04:08 jpansanel joined #cdk
04:08 stain joined #cdk
04:14 azeem joined #cdk
04:14 s_wolf joined #cdk
04:14 jpansanel joined #cdk
04:14 stain joined #cdk
04:18 azeem joined #cdk
04:18 s_wolf joined #cdk
04:18 jpansanel joined #cdk
04:18 stain joined #cdk
05:24 sneumann_ joined #cdk
05:34 bag_ joined #cdk
05:39 egonw joined #cdk
05:59 jbrefort joined #cdk
06:59 sneumann_ joined #cdk
08:00 s9asad joined #cdk
08:05 s9asad joined #cdk
10:19 s_wolf joined #cdk
10:32 mgerlich joined #cdk
10:40 maclean joined #cdk
13:10 egonw joined #cdk
13:50 slyrus joined #cdk
14:05 maclean egonw: all StereoTool methods now tested, have annotations, etc. no complaints in report/result-standard.txt (for Stereo, at least :)
14:05 maclean shall I submit a patch?
14:05 egonw cool
14:06 egonw is it CIP-independent too now?
14:06 maclean yup
14:06 egonw nice
14:06 egonw yes, please go ahead
14:06 maclean ok
14:12 s9asad joined #cdk
14:12 egonw hi s9asad
14:12 s9asad Hi Egon
14:14 s9asad CDK has a function to generate canonical SMILES? IAtomContainer => Canonical SMILES
14:14 egonw yes
14:15 s9asad Can we get IAtomContainer to Canonical IAtomContainer without converting it to SMILES?
14:16 egonw mmm... cannot find how to do it, though
14:16 s9asad This function might help to layout similar molecules in similar manner or speed the substructure search
14:17 s9asad I think there must be something is the code where we generate canonical SMILES
14:18 s9asad I will play around the code to see if it can be done :-)
14:18 s9asad you suggestions are welcome
14:18 s9asad your* (CDK developers/users)
14:19 s9asad Hi maclean, is it possible to do something like this with signatures?
14:23 maclean s9asad : sorry, was afk
14:23 maclean mom
14:24 maclean oh, right. yes it is very easy : String canonicalSignatureString = new MoleculeSignature(atomConta​iner).toCanonicalString();
14:24 maclean perhaps that's not quite "easy", but it is one-line...
14:26 maclean Oh, no wait. That's not what you want. sorry.
14:27 maclean You want the canonical labelling, and then apply that permutation to the container.
14:28 maclean So, uhm...
14:29 maclean int[] permutation = new MoleculeSignature(atomConta​iner).getCanonicalLabels();
14:29 maclean then, well, some code that applies that permutation.
14:31 maclean Equally, you could do that with the CanonicalLabel(l)er, but obviously in an inferior way :)
14:34 egonw obviously :)
14:58 asad_ joined #cdk
15:01 maclean This is bizarre. Public superclass methods should be callable from a subclass, right?
15:04 maclean Like class BassClass { public int getX() {...}} and class DerivedClass extends BaseClass { ..} and then d = new DerivedClass().getX()...
15:09 maclean Ahhhh. I see the problem. signatures jar not exported from CDK...
15:09 maclean Sorry, never mind all.
15:19 s9asad joined #cdk
15:36 s9asad joined #cdk
15:38 maclean s9asad : http://github.com/gilleain/labelling
15:38 maclean (sample code for labelling)
15:40 s9asad Thanks for the code maclean
15:40 maclean n/p
15:41 s9asad so after canonical labelling if we rewrite the atomcontainer in a mol file the index will change?
15:41 maclean It should, yes.
15:42 maclean The order of the atoms in the container's atom[] will change.
15:42 maclean The bonds stay in the same order, but they refer to atoms with new positions.
15:42 maclean I think that's right, isn't it?
15:43 s9asad say mol1 is o-p=p-p and mol2 is o-p-p=p but after canonization they both are stored as say o-p=p-p
15:44 s9asad ?
15:44 maclean what? no - those are different molecules, aren't they?
15:45 maclean mol1 = O1.P2.P3.P4/1-2.2=3.3-4 and mol2 = O1.P2.P3.P4/1-2.2.3.3=4
15:45 s9asad I mean if you take phosphate for example
15:45 s9asad it can be written in many way
15:46 s9asad but after canonization it atom iterator should be same?
15:46 maclean yes, it should.
15:47 s9asad great
15:47 maclean Different canonizers will produce a different ordering, of course.
15:47 s9asad but if I just use you tool, it should be consistent ;-)
15:48 s9asad your*
15:48 maclean or just the CanonicalLabeler (note one 'l') - I've changed the interface name to ICanonicalLabeller
17:02 jbrefort joined #cdk
17:26 maclean joined #cdk
17:51 egonw joined #cdk
17:59 s9asad joined #cdk
18:17 s9asad left #cdk
19:42 sneumann_ joined #cdk
20:17 sneumann__ joined #cdk
20:26 sneumann__ joined #cdk
20:26 s_wolf joined #cdk
20:26 azeem joined #cdk
20:26 jpansanel joined #cdk
20:26 stain joined #cdk
20:28 sneumann__ joined #cdk
20:28 s_wolf joined #cdk
20:28 azeem joined #cdk
20:28 jpansanel joined #cdk
20:28 stain joined #cdk
20:30 sneumann__ joined #cdk
20:30 s_wolf joined #cdk
20:30 azeem joined #cdk
20:30 jpansanel joined #cdk
20:30 stain joined #cdk
21:02 jbrefort joined #cdk
23:21 bag_ joined #cdk
23:34 CIA-52 cdk: Egon Willighagen cdk-1.2.x * r4ac7111 / src/test/org/openscience/cdk/BondTest.java :
23:34 CIA-52 cdk: Updated test to assume new Bond() creates a bond with zero atoms
23:34 CIA-52 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/dkmm5I
23:34 CIA-52 cdk: Egon Willighagen cdk-1.2.x * r916ab96 / src/main/org/openscience/cdk/Bond.java :
23:34 CIA-52 cdk: Fixed Bond() constructor to create a bond with zero atoms. Also fixed setAtom(IAtom, int) to increase the atom count if a null entry is filled with a non-null IAtom.
23:34 CIA-52 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/9sNdTH
23:34 CIA-52 cdk: Egon Willighagen cdk-1.2.x * r23d9685 / (2 files in 2 dirs):
23:34 CIA-52 cdk: Also fix the new NNBond() == 0 atoms for the nonotify module
23:34 CIA-52 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/aFJV8O
23:34 CIA-52 cdk: Egon Willighagen cdk-1.2.x * r6cf95db / (2 files in 2 dirs):
23:34 CIA-52 cdk: Additional patch to reduce atom count on setAtom(null, int) and unit tests for the setAtom(IAtom, int) behavior.
23:34 CIA-52 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/cl3scE
23:34 CIA-52 cdk: Egon Willighagen cdk-1.2.x * rce9b1bd / src/test/org/openscience/cdk/in​terfaces/AbstractBondTest.java :
23:34 CIA-52 cdk: Backport patch, to make the patches compile with cdk-1.2.x
23:34 CIA-52 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/9RuE3x

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