Camelia, the Perl 6 bug

IRC log for #cdk, 2008-02-03

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

All times shown according to UTC.

Time Nick Message
05:49 jbrefort joined #cdk
07:21 egonw joined #cdk
07:21 egonw moin
07:21 egonw around, but busy...
07:42 egonw joined #cdk
10:27 carsten joined #cdk
10:43 CIA-18 cdk: egonw * r10004 /branches/miguelrojasch/react​ion/src/org/openscience/cdk/ (4 files in 2 dirs): Added comments with 'FIXME: REACT'
10:56 carsten egonw: didn't know launchpad can do such things
10:56 egonw yeah, pretty cool
10:56 egonw and damn useful
10:56 egonw brings OS development to a more professional level
10:56 egonw beter insight in what is going on..
10:57 egonw and what the state is of things...
10:57 egonw should recommend this to jos too...
10:58 carsten indeed
10:58 carsten nice nice nice
10:58 carsten also nice: Git. I am now using Git to develeop KDE stuff
10:58 carsten no longer touching SVN :-)
10:59 egonw :)
10:59 egonw git-svn?
10:59 egonw I assume...
11:00 carsten I am developing with Git, the real thing
11:00 carsten I push my changes into kde's svn using git-svn dcommit
11:03 egonw right
11:03 egonw yes, used it for a while...
11:03 carsten and git is seriously cool
11:03 egonw it's great for on laptops in trains..
11:04 egonw but synching was a bit painful...
11:04 egonw sometimes my git checkout could not longer be synched...
11:04 egonw whcih is a killer...
11:04 egonw I'm sure I did something wrong...
11:04 carsten git-svn rebase
11:04 egonw yeah, I know...
11:04 egonw see my blog...
11:04 carsten so far it always worked here
11:05 carsten But I had huge issues with old verfsion of git
11:05 egonw yeah, not sure what went wrong...
11:05 egonw I've used it succesfully
11:05 carsten Git 1.5.4 is out.
11:05 carsten 165 contributers resulting in 684 changed files, included 70,435 insertions and 28,984 deletions.
11:05 carsten that is the diff to 1.5.3 :-)
11:07 rojasm joined #cdk
11:07 carsten 'git commit' has been rewritten in C;          <--- very very cool
11:07 carsten important for Windows!
11:07 carsten 'git clean' has been rewritten in C       <--- same
11:08 egonw yeah, might try it again...
11:11 egonw on the other hand...
11:11 egonw using mercurial, svn, cvs and since yesterday bzr too...
11:12 * carsten tried bzr about a year ago. I HATED it
11:13 carsten And I am no longer using CVS, all projects I contribute to switched finally to SVN :)
11:13 * carsten read that bzr improved a lot in the last few month
11:13 egonw so far I've been happy with it...
11:14 egonw but quite happy with LP anyway :)
11:14 egonw not just only their code.launchpad.net using bzr bit
11:14 egonw OK, lunch time
11:14 egonw bbl
11:22 rojasm hi all
11:23 carsten moin rojasm
11:24 rojasm egonw: question! as allways sorry. is there some method using AllRingsFinder which doesn't find the superring. I mean the ring which group two rings together
11:24 CIA-18 cdk: egonw * r10005 /branches/egonw/jchempaint​/src/org/openscience/cdk/ (17 files in 12 dirs): Removed the GeometryTools that worked with rendering coordinates
11:30 rojasm moin all
11:42 egonw hi rojasm
11:42 egonw still having lunch
11:42 egonw shoudl be back in some 15 minutes or so
11:42 egonw rojasm: left comments in the branch, using FIXME: REACT
11:42 egonw rojasm: about those rings...
11:43 egonw check SSSRFinder
11:43 egonw which finds you all rings...
11:43 egonw then it is fairly straighforward to detect if rings are merged...
11:58 rojasm ok thanks
12:02 egonw will be out to catch some sun
12:02 egonw bbl
12:28 CIA-18 jmol: nicove * r8947 /trunk/Jmol-FAH/projects/ (p2584.xyz.gz p2588.xyz.gz): Folding@Home
12:36 carsten joined #cdk
12:48 * egonw is back
12:48 rojasm :)
12:49 egonw it's fairly cold outside
12:54 egonw rojasm: got some minutes to discuss the blueprint tree?
12:55 rojasm Yes exactly I am preparing the lunch. in 30 min?
12:55 egonw yes, fine
12:55 rojasm Sorry. I am preparing the lunch. Is it ok in 30 min
12:55 egonw yeah, really no problem :)
13:43 rojasm joined #cdk
13:53 rojasm joined #cdk
14:05 egonw rojasm: btw, did you see: http://chem-bla-ics.blogspot.com/2008/02/​defining-development-goals-launchpad.html ?
14:17 rojasm I am here
14:18 rojasm Nice description
14:35 CIA-18 cdk: miguelrojasch * r10008 /branches/miguelrojasch/react​ion/src/org/openscience/cdk/ (4 files in 4 dirs): added some FIXME: REACT
14:35 kamevar joined #cdk
14:36 egonw hi kamevar. Welcom!
14:36 egonw +e
14:36 kamevar Hi
14:38 rojasm egonw: the resonance structure products are correctly generated for benzene. But fails with is applied the isIsomorph method which find it is the same as the reactant
14:40 rojasm Some Idea how I could create a queryAtomContainer with Symbol,Charge,Bond,LonePair,SingleElectron distinction
14:41 egonw let me check the API
14:42 egonw kamevar: btw, what brings you here? you're using the CDK, I assume?
14:44 carsten joined #cdk
14:45 carsten egonw: ping
14:45 kamevar I'm interested in something and simply looking through this channel....
14:45 egonw rojasm: I don't think this can be done with the UIT
14:45 carsten could you please paste the URL      svn info spectra.xml gives you?
14:45 egonw kamevar: something?
14:45 kamevar :)
14:46 egonw carsten: Revision: 12
14:46 carsten no, the URL please :)
14:46 egonw https://bodr.svn.sourceforge.net/svnro​ot/bodr/trunk/bodr/spectra/spectra.xml
14:46 carsten I want to svn cat $URL
14:47 egonw kamevar: seriously... what are you interesting in?
14:47 carsten egonw: ok, I wonder what the e1 stands for to be honest...
14:47 carsten e1 to e56
14:48 egonw element?
14:48 egonw no idea, really...
14:48 carsten possibly, yes
14:48 egonw please ask on the ML
14:48 * carsten is working on Kalziums Specturm stuff
14:48 kamevar i'm interesting in vitamins.. and trying to understand: is this resource can be useful for me
14:52 egonw kamevar: only you want to predict some property of them, better check out wikipedia
14:52 egonw kamevar: this channel is about chemoinformatics software
14:52 egonw cdk.sf.net in particular
14:52 kamevar i see
14:53 egonw rojasm: the problem is that the current UIT architecture is to compare IAtom and IBond separately...
14:53 egonw separately from IAtomContainer at least...
14:54 egonw however, LP and SE are not accessible from IAtom :(
14:54 kamevar ok, good luck for your work! :)
14:55 rojasm That means any solution for the problem
14:55 rojasm ?
14:56 kamevar left #cdk
14:57 egonw no, a workaround is to have the UIT do the matching on Symbol,Charge,Bond
14:57 egonw and let it return with too many solutions...
14:57 egonw for each solution it has the mapping...
14:58 egonw so you can then use that to check the LPs and SEs...
14:59 rojasm Ok, but other ways for bond it is possible to difference two benzene with double bonds distribution different?
14:59 egonw benzene not...
15:00 rojasm :(
15:00 egonw because the two kekule structure *are* isomorphic
15:00 egonw which is why I added two different side chains...
15:00 egonw which breaks the symmetry
15:00 rojasm yes!
15:01 rojasm but It finds are similar reactant and product
15:01 rojasm The test fails because in StructureResonanceGenerator.existAC() return true
15:02 rojasm which means reactant and product are symilars
15:02 egonw for benzene yes
15:02 egonw use para-xylene
15:02 egonw sorry...
15:02 egonw ortho-xylene
15:03 egonw then the two resonance structures are not isomorfic
15:03 egonw oh, and note that you must not detect aromaticity...
15:03 egonw because the UIT is happy when one is single and other is double, bot both are aromatic...
15:04 egonw but did not see aromaticity detection in the unit tests anyway...
15:09 rojasm what method I should use
15:10 CIA-18 cdk: miguelrojasch * r10009 /branches/miguelrojasch/reaction/sr​c/org/openscience/cdk/test/tools/St​ructureResonanceGeneratorTest.java: added Test 12DimethylBenezene
15:11 egonw not sure
15:11 egonw I would go for the workaround...
15:11 egonw have the UIT just match on symbol, charge, bond order
15:12 egonw and do a post check for LPs and SEs
15:13 rojasm egonw: if in the line 414 I put
15:13 rojasm SmilesGenerator sg = new SmilesGenerator();
15:13 rojasm System.out.println(sg.createSMILES((IMolecule) atomContainer));
15:13 rojasm System.out.println(sg.createSMILES((IMolecule) ac));
15:13 rojasm and let run the last test of the structureResonanceGenerator
15:14 rojasm it says are the same :(
15:14 egonw the SMILES are not, you mean?
15:14 rojasm not
15:14 rojasm one has the double bond between the chain and the other not
15:15 rojasm which shows clear are not the same
15:15 rojasm http://pubchem.ncbi.nlm.nih.gov/edit/index.html
15:15 egonw indeed
15:15 rojasm I look here
15:16 egonw OK
15:16 egonw ummm...
15:16 egonw might you please first do this:
15:17 egonw create a new class cdk.test.atomtype.ResonanceStructuresTest
15:17 egonw with a test method for the expected  product
15:17 egonw ac or atomContainer...
15:17 egonw btw, please name the correct something like expectedStructure
15:18 rojasm ok, clar
15:22 rojasm left #cdk
15:22 rojasm joined #cdk
15:23 egonw and please add a @see in the JavaDoc of the method of the whatever unit test this product is expected in, pointing to the new method in ResonanceStructuresTest
15:23 rojasm egonw could you please write again the last 3 lines
15:23 egonw [16:17] <egonw> with a test method for the expected  product
15:24 egonw [16:17] <egonw> ac or atomContainer...
15:24 egonw [16:18] <egonw> btw, please name the correct something like expectedStructure
15:24 egonw [16:18] <rojasm> ok, clar
15:24 egonw [16:22] <-- rojasm has left this channel.
15:24 egonw [16:22] --> rojasm has joined this channel (n=mrc@s55927d53.adsl.wanadoo.nl).
15:24 egonw [16:23] <egonw> and please add a @see in the JavaDoc of the method of the whatever unit test this product is expected in, pointing to the new method in ResonanceStructuresTest
15:24 egonw [16:23] <rojasm> egonw could you please write again the last 3 lines
15:24 rojasm perfect
15:57 egonw rojasm: got to go make dinner
15:57 CIA-18 cdk: miguelrojasch * r10010 /branches/miguelrojasch/react​ion/src/org/openscience/cdk/ (2 files in 2 dirs): added new test in StructureResonanceGenerator to prove if the isomorph test can distinguish between isomers
15:59 rojasm ok
16:00 egonw just continue asking questions adn so...
16:00 egonw will visit the laptop every now and then...
16:22 jpansanel moin
16:28 carsten joined #cdk
16:28 jpansanel egonw : the Junior Jobs listed on http://sourceforge.net/projects/cdk
16:28 jpansanel are up to date ?
16:28 jpansanel I could do some of them next week
16:42 egonw yes
16:42 egonw make sure to leave a note in the tracker
16:42 egonw indicating that you start working on it...
17:03 CIA-18 cdk: miguelrojasch * r10011 /branches/miguelrojasch/react​ion/src/org/openscience/cdk/ (2 files in 2 dirs): added name tests corresponding to the method according junit4
17:11 rojasm I put what you said! And for this case the test doesn't fail
17:11 rojasm Strang
17:21 egonw dinner now
17:21 egonw bbl
17:24 rojasm ok, the test for testIsomphrXilene fails when I call before of the AsserTrue the SmilesParser
17:47 CIA-18 cdk: miguelrojasch * r10012 /branches/miguelrojasch/reaction/sr​c/org/openscience/cdk/test/tools/St​ructureResonanceGeneratorTest.java: Added Inchi's for those real structure
19:30 rojasm left #cdk
19:41 edrin joined #cdk
19:41 egonw joined #cdk
20:41 rojasm joined #cdk
20:41 rojasm left #cdk
21:21 jbrefort joined #cdk

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