Camelia, the Perl 6 bug

IRC log for #cdk, 2008-12-13

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

All times shown according to UTC.

Time Nick Message
02:16 CIA-16 cdk: rajarshi * r13550 /cdk/branches/mark_rynbeek-vflib/src/ (2 files in 2 dirs): Refactored so that constructor signature matches that of UIT (i.e., target, query)
02:55 CIA-16 cdk: rajarshi * r13551 /cdk/branches/mark_rynbeek-vflib/src/ (2 files in 2 dirs):
02:55 CIA-16 cdk: Refactored so that one can set up the class with no target and query and
02:55 CIA-16 cdk: manually specify a target and query. This is useful since you can now run
02:55 CIA-16 cdk: multipe different queries against a fixed target or multiple targets with a
02:55 CIA-16 cdk: given query.
02:59 CIA-16 cdk: rajarshi * r13552 /cdk/branches/mark_rynbeek-vflib/src/main/org/open​science/cdk/isomorphism/SubgraphIsomorphism.java: Added test class/method annotations
03:09 CIA-16 cdk: rajarshi * r13553 /cdk/branches/mark_rynbeek-vflib/src/ (2 files in 2 dirs): Updated Javadocs with usage example. Also added a getter for the method - better than grabbing the mappings directly
03:37 CIA-16 cdk: rajarshi * r13554 /cdk/branches/mark_rynbeek-vflib/src/main/org/open​science/cdk/isomorphism/SubgraphIsomorphism.java: Added some more Javadocs tags
06:11 slyrus_ joined #cdk
09:11 edrin joined #cdk
10:11 CIA-16 joined #cdk
10:35 carsten joined #cdk
12:08 edrin joined #cdk
12:27 Gpox joined #cdk
12:28 Gpox left #cdk
13:10 jbrefort joined #cdk
16:41 CIA-16 joined #cdk
19:07 egonw joined #cdk
19:50 rguha joined #cdk
19:50 rguha hi egonw
19:52 egonw hi rguha
19:53 rguha saw your mail, good excuse for some coding rather than writing slides :)
19:53 egonw :)
19:53 * egonw is playing with his new laptop
19:53 egonw read installing java/eclipse etc
19:53 egonw and copying git repositories from the old to the new laptop
19:53 rguha aah
19:54 rguha_ joined #cdk
19:54 rguha_ left #cdk
19:54 rguha_ joined #cdk
19:54 egonw how was the eScience session?
19:55 rguha_ so so
19:55 rguha_ just 4 people in the room
19:55 egonw :(
19:55 rguha_ the date change at the last minute messed things up
19:55 egonw that's not good
19:55 rguha_ yeah
19:55 CIA-16 joined #cdk
19:56 rguha_ so can you do a review of the vflib code?
19:57 egonw yes, but not now
19:57 egonw cdk1.2.x is my current priority
19:57 rguha_ ok
19:57 rguha_ whats left on 1.2.x?
19:57 egonw check the wiki
19:58 egonw not too much...
19:58 egonw I could use some help with getting the number of failing unit tests down
19:58 egonw in particular for the XLogP descriptor
19:58 egonw oh... btw
19:58 egonw got an email from a Italian group
19:59 rguha_ i plan to
19:59 egonw with a zip with extra descriptors
19:59 rguha_ one interview on Monday and then 3 week beak
19:59 rguha_ aah, nice
19:59 egonw have not checked it yet
19:59 rguha_ what types?
19:59 egonw but will set up a branch for it
19:59 rguha_ is it contributed for integration?
20:00 egonw "some very simple topological descriptors"
20:00 rguha_ aah
20:00 rguha_ it'll be good to see
20:01 egonw I'll check it next week
20:02 egonw something else
20:02 egonw Nightly is really nice
20:02 rguha_ :)
20:02 egonw I have a four core machine at my disposal
20:02 egonw so have been happily setting up nightlies for branches
20:02 rguha_ looking at the bugs, I see that 2192238 si still open
20:02 rguha_ cool!
20:02 rguha_ also, 2217061 can be closed
20:02 egonw what we now need, is a webpage what summarizes nightly states
20:03 egonw please go ahead
20:03 egonw so, I plan to have Nightly report some stuff in RDF
20:03 rguha_ not a bad idea
20:03 egonw and have a simple aggregator which summarizes states in a simple table
20:03 egonw with some basic info
20:03 rguha_ aah, I was thinking on similar lines - basically create an XML format along with the HTML output
20:03 egonw branch
20:03 egonw JDK version
20:03 egonw that sort of things
20:04 rguha_ yes, that'll be useful
20:04 rguha_ especially if nightly puts the XML file in a standard place
20:04 egonw yes, though I think using RDFa would be nice :)
20:04 rguha_ the aggregator will be able to ull info from any CDK nightly site
20:04 egonw if you want to go the XML way, let's do XML + XSLT
20:04 rguha_ RDFa sounds nice, I know nothing :) So I leave that to you
20:04 rguha_ yes, XML+XSLT sounds good
20:05 egonw ok, but this is a bit low priority too
20:05 egonw I also need to finish a book chapter and contribute to another
20:05 rguha_ right
20:05 egonw the first was a mistake... planning wise
20:05 rguha_ yes, book chapters are a little painful
20:05 rguha_ btw did you add anything to Camerons O'Reilly chapter?
20:05 egonw that is... unavoidable, unpredictable things came up, which very much interfered :(
20:06 egonw no, not yet...
20:06 egonw that's the other chapter
20:06 rguha_ do you know what book it's for?
20:06 egonw need to hack a bit on the RDF viewing in Bioclipse...
20:06 egonw looks nice... will be a nice section in that chapter, I think
20:06 egonw Cameron's chapter?
20:06 rguha_ yes
20:06 egonw Beautiful Data series
20:07 egonw which reminds me...
20:07 egonw really need to buy one of those processing.org books...
20:07 egonw oh, did you see this:
20:07 egonw http://chemhack.com/mx-gwt/demo​-molecule-structure-rendering/
20:07 zarah egonw's link is also http://tinyurl.com/5dt37m
20:07 egonw speaking of visualization :)
20:08 rguha_ nice!
20:08 rguha_ yes, processing is quite cool
20:11 rguha joined #cdk
20:11 rguha stupid vpn
20:12 egonw :)
20:12 egonw nxclient?
20:12 egonw assuming you have a linux machine at iu too
20:12 rguha no, this is on my macbook
20:12 rguha linux machine in the office
20:14 rguha btw, where is the check copyright code located?
20:14 egonw there might be nxclient for OS/X too
20:14 egonw src/main/net?
20:14 egonw not sure
20:15 CIA-16 cdk: egonw * r13555 /cdk/branches/cdk-1.2.x/src​/test/org/openscience/cdk/ (5 files in 4 dirs): Added a few missing test classes
20:19 rguha what's opinion on https://sourceforge.net/tracker2/?func=detail&am​p;aid=2192238&group_id=20024&atid=120024
20:22 egonw I think miguel fixed that
20:22 rguha hmm, I was thinking of the more general issue of PTElement vs PeriodicTable
20:22 rguha the discussion on the mailing list never got resolve
20:23 rguha my contention ws that PeriodicTable would be the users entry point to periodic table info
20:23 rguha I didn't see a role of rPeriodicTableElement (except for certain inernal usages)
20:32 egonw sorry...
20:32 egonw was afk
20:32 egonw ummm
20:32 egonw I understood the concensus was to remove the duplicate code
20:32 egonw and wasn't that what Miguel did?
20:35 rguha i didn't think so
20:35 rguha but since it will involve broad changes maybe its better for trunk
20:35 egonw ack
20:36 rguha but I just don't see the role of PeriodicTableElement and the factory and manually configuring it
20:36 egonw I think it is just the data model for the PT gui element
20:36 egonw used by JCP
20:37 rguha well I'll go ahead and update the code to address he curret bug, but IMO all new code should be useing PeriodicTable
20:37 rguha if I have one big desire for the CDK, is that it's API needs simplifying
20:37 egonw :)
20:38 egonw same here
20:38 rguha btw, some nice news, can you private message me
20:38 egonw though I want performance first
20:38 rguha I agree
20:38 rguha but from a users point of view, performance wont mater if they cant work out what they want to do :)
20:39 egonw true
20:43 rguha joined #cdk
20:43 rguha left #cdk
20:44 rguha joined #cdk
20:52 egonw ok, skype works too...
20:52 egonw including the laptop build in webcam :)
20:54 rguha :)
20:58 CIA-16 cdk: egonw * r13556 /cdk/branches/cdk-1.2.x/src/test/org/​openscience/cdk/AtomParityTest.java: IAtomParity extend IChemObject, so AtomParityTest should extend ChemObectTest
21:00 slyrus_ egonw: is their a git mirror of cdk around?
21:01 egonw slyrus_: how kind of you to ask :)
21:01 slyrus_ s/their/there/
21:01 egonw yes, there is ...
21:01 slyrus_ I find SVN and its damn branches to be rather annoying and space inefficient
21:02 egonw http://github.com/egonw/cdk/tree/master
21:02 zarah egonw's link is also http://tinyurl.com/3tpstn
21:02 slyrus_ thanks egonw!
21:03 rguha_ joined #cdk
21:04 slyrus_ egonw: btw, that InChI/mckay graph canonicalization algorithm looks like a PITA to reimplement
21:04 egonw just the other day depth-first complained about a clean inchi library
21:06 slyrus_ time to make a trip down to the local chem library to look for that Helson CCR structure drawing paper
21:06 slyrus_ clean, how so?
21:07 egonw read the blog
21:08 rguha joined #cdk
21:09 rguha left #cdk
21:09 rguha joined #cdk
21:09 rguha left #cdk
21:09 rguha joined #cdk
21:09 rguha left #cdk
21:10 rguha joined #cdk
21:13 slyrus_ ah, I see.
21:18 CIA-16 cdk: rajarshi * r13557 /cdk/branches/cdk-1.2.x/src/ (4 files in 4 dirs): Updated PeriodictableElement so that it does not allow invalid IUPAC group numbers to be set. Fixes bug #2192238
21:18 zarah bug #2192238 | http://tinyurl.com/6fa4a2
21:22 rguha ncie link from the commit message
21:22 rguha why does it go to Bugzilla at Uppsala?
21:22 rguha (and not SF)
21:22 rguha left #cdk
21:23 rguha joined #cdk
21:25 egonw because zarah is not smart enough to distinguish between bioclipse and cdk bugs :)
21:28 rguha aah
21:30 rguha any reason why the N in C#N=CC(CNC)N1CC1 has a null atom type?
21:33 slyrus_ bah. ant 1.7.1.
21:33 egonw yes, 1.7.0 seems to have a bug
21:34 egonw C#N=C ?
21:34 egonw Mm....
21:34 egonw if only they had charges....
21:35 egonw I think...
21:36 egonw or does the first carbon have a hydrogen?
21:36 egonw mmm... yeah, I can see why this fails...
21:36 egonw I don't think I ever considered this set up...
21:36 rguha the error is occuring when the test code is adding H's
21:37 egonw sure
21:37 egonw if no atom type is perceived...
21:38 egonw then the CDK does not know the atom's valency
21:38 egonw please add a unit test...
21:38 rguha to which class?
21:38 egonw and file a bug report
21:38 slyrus_ wouldn't it be easier just to fire the chemist who wanted to make that? :)
21:38 egonw CDKAtomTypeTest
21:38 egonw well, he likely already is
21:38 egonw given current financial affairs :)
21:39 rguha ok
21:42 CIA-16 cdk: rajarshi * r13558 /cdk/branches/cdk-1.2.x/src/test/or​g/openscience/cdk/CDKTestCase.java: Updated to use JDK 1.5 idiom
21:50 CIA-16 cdk: egonw * r13559 /cdk/branches/cdk-1.2.x/src​/test/org/openscience/cdk/ (7 files in 4 dirs): More missing unit test classes: for AdductFormulaTest and MolecularFormulaSet for nonotify and datadebug
21:51 CIA-16 cdk: rajarshi * r13560 /cdk/branches/cdk-1.2.x/src/test/org/openscie​nce/cdk/atomtype/CDKAtomTypeMatcherTest.java: add test case for a N atom type
21:59 CIA-16 cdk: rajarshi * r13561 /cdk/branches/cdk-1.2.x/src/main/o​rg/openscience/cdk/qsar/descriptor​s/molecular/ALOGPDescriptor.java: Updted to add explicit H's always - this should be cached, but we don't have that mechanism. But it fixes bug #2185475
22:00 zarah bug #2185475 | http://tinyurl.com/57c3b9
22:03 egonw I'm going offline
22:03 rguha bye
22:03 egonw bye
22:03 CIA-16 cdk: rajarshi * r13562 /cdk/branches/cdk-1.2.x/src/main/o​rg/openscience/cdk/qsar/descriptor​s/molecular/XLogPDescriptor.java: Updted to add explicit H's always - this should be cached, but we don't have that mechanism. But it fixes bug #2185475
22:03 zarah bug #2185475 | http://tinyurl.com/57c3b9
22:15 CIA-16 cdk: rajarshi * r13563 /cdk/branches/cdk-1.2.x/src/main/o​rg/openscience/cdk/tools/manipulat​or/AtomContainerManipulator.java: Correctly remove H's even for H2, fixes bug #2366528
22:15 zarah bug #2366528 | http://tinyurl.com/6yjhfd
23:06 jpansanel joined #cdk
23:06 Conrad joined #cdk
23:48 slyrus__ joined #cdk

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