Camelia, the Perl 6 bug

IRC log for #cdk, 2010-02-03

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

All times shown according to UTC.

Time Nick Message
05:02 Gpox joined #cdk
06:23 sneumann__ joined #cdk
07:53 egonw joined #cdk
08:15 bag joined #cdk
08:37 bag joined #cdk
09:12 sneumann__ joined #cdk
10:00 jbrefort joined #cdk
11:45 bag joined #cdk
12:11 CIA-50 cdk: shk3 * r15364 /detgen/:
12:12 CIA-50 cdk: shk3 * r15365 /detgen/trunk/:
12:14 CIA-50 cdk: shk3 * r15366 /detgen/trunk/: will come with checkin
12:15 CIA-50 cdk: shk3 * r15367 /detgen/trunk/: Initial import.
12:19 CIA-50 cdk: shk3 * r15368 /detgen/trunk/ (116 files in 8 dirs): Initial import.
13:34 CIA-50 cdk: shk3 * r15369 /detgen/trunk/ (. build.xml): better build.xml, added build to svnignore
13:38 PeterOdeus joined #cdk
13:45 egonw hi PeterOdeus!
13:45 PeterOdeus Greetings, egonw :-)
13:47 PeterOdeus I've cloned http://pele.farmbio.uu.se/git/cdk-jchempaint.git/ and now I'm on branch origin/13-bioclipse2.1.x...
13:47 zarah PeterOdeus's link is also http://tinyurl.com/ydn95m8
13:47 egonw good
13:48 PeterOdeus But still I cannot see any "Patch 2913324A" as stated at http://sourceforge.net/tracker/?func=detail&amp​;aid=2912627&group_id=20024&atid=120024
13:48 zarah PeterOdeus's link is also http://tinyurl.com/yatxucv
13:49 PeterOdeus How can I get hold of that patch?
13:50 egonw that's a bit tricky
13:50 egonw but I am working on a rebase on CDK 1.3.2
13:50 egonw but waiting for one patch
13:51 egonw the activities (2.4M) are in
13:51 egonw assays are being inserted
13:51 egonw but molecules, targets and docs still have to be done
13:52 egonw done == started
13:52 egonw oh, wrong channel :)
13:52 egonw regarding the activities :) that rdf chembl talk :)
13:53 egonw PeterOdeus: this is what you can do:
13:53 egonw 1. add the GitHub cdk repository as remote, with git remote add foo bar
13:53 egonw foo being an alias, bar being the URL to the github repos
13:53 egonw 2. git fetch foo
13:55 egonw 3. git checkout 13-bla
13:55 egonw 4. git cherry-pick bbc8f60caf6d5a9f5d067bba61e046e16d9e4b21
13:55 egonw that will attempt to port the single atom UIT fix patch to the branch you are on
13:56 egonw oh... :(
13:58 carsten_ joined #cdk
14:06 olass joined #cdk
14:08 PeterOdeus joined #cdk
14:12 egonw PeterOdeus: I'm pushing that commit to 13-foo as we speak
14:15 egonw PeterOdeus, olass: please pull from pele
14:15 egonw patch should be there now
14:15 olass will do
14:15 egonw http://pele.farmbio.uu.se/cgi-bin/gi​tweb.cgi?p=cdk-jchempaint.git;a=shor​tlog;h=refs/heads/13-bioclipse2.1.x
14:15 zarah egonw's link is also http://tinyurl.com/yzsn8fk
14:17 olass egonw: will confirm if it works in 1 min
14:17 egonw ack
14:20 olass egonw: thatdid not solve the MCSS problem
14:20 olass egonw: can you run the unit test against 13-foo?
14:20 olass the new one I sent you smiles for...
14:20 egonw new SMILES?
14:20 egonw I only saw those in the bug report from Lars
14:20 olass ok, that one indeed...
14:21 olass sry
14:21 egonw ok, np
14:21 egonw just making sure :)
14:21 olass if you can confirm that this test succeeds with 13-foo then problem solved
14:22 egonw booting eclipse ...
14:22 egonw test applied...
14:22 olass thx
14:28 sneumann In 1.3.x I can't anymore IMolecularFormula molecularFormula = new MolecularFormula();
14:28 sneumann what will return a new MolecularFormular thing ?
14:29 egonw DefaultChemObjectBuilder.getIn​stance().newMolecularFormula()
14:29 egonw or, if you have a Molecule or so around:
14:29 sneumann merci
14:29 egonw mol.getBuilder().newMolecularFormula()
14:29 egonw latter is prefered...
14:29 egonw any IChemObject has getBuilder()
14:31 egonw olass, PeterOdeus: ported the test case, and that unit test runs fine
14:31 egonw see the 13- branch
14:31 olass ok thx
14:31 olass I will dig deeper then
14:34 sneumann No, don't get this: what is mol. in mol.getBuilder().newMolecularFormula() ?
14:36 sneumann av, ic: an object of type IMolecularFormular.
14:43 Guest57794 egonw: Moin, you made heise.de :)
14:44 egonw Guest57794: yeah, I know :)
14:44 egonw well... Bioclipse has
14:44 egonw sneumann: indeed
14:44 carsten joined #cdk
14:45 sneumann but this will give me an NPE, wont it ?
14:45 sneumann IMolecularFormula molecularFormula = null;
14:46 sneumann molecularFormula = molecularFormula.getBuilde​r().newMolecularFormula();
14:46 egonw yes, that will give a NPE
14:46 egonw but you can use any IChemObject
14:47 egonw but if you have none around, then you need to use the DefaultChemObjectBuilder.getInstance()
14:49 PeterOdeus joined #cdk
14:50 PeterOdeus egonw: The junit test involving mcss/single atoms now succeeds. Thanks.
14:51 egonw ok, great !
15:04 CIA-50 cdk: Egon Willighagen master * rea5f0b7 / (4 files in 2 dirs): (log message trimmed)
15:04 CIA-50 cdk: Added a new IO API for reporting file format errors.
15:04 CIA-50 cdk: * introducing a IChemObjectReaderErrorHandler
15:04 CIA-50 cdk: * handleError() methods for IChemObjectReader's to report problems
15:04 CIA-50 cdk: * these handleErrors() method throw an CDKException when
15:04 CIA-50 cdk:  in STRICT mode
15:04 CIA-50 cdk: * line and column number support, useful for MDL V2000 molfile text
15:04 CIA-50 cdk: Egon Willighagen master * rfd81efc / src/main/org/openscience/c​dk/io/MDLV2000Reader.java :
15:04 CIA-50 cdk: Use the new error reporting IO API
15:04 CIA-50 cdk: Signed-off-by: Rajarshi Guha <rajarshi.guha@gmail.com> - http://bit.ly/cGsuCk
15:21 PeterOdeus joined #cdk
15:29 PeterOdeus left #cdk
15:38 egonw joined #cdk
16:41 sneumann cu later
17:06 CIA-50 cdk: Egon Willighagen master * rf7cc552 / (build.props changelog.txt): Release 1.3.2 - http://bit.ly/dwEpY3
17:22 egonw joined #cdk
17:22 olass joined #cdk
17:22 bag joined #cdk
17:22 jbrefort joined #cdk
17:22 zarah joined #cdk
17:22 mgerlich joined #cdk
17:22 s_wolf joined #cdk
17:22 Conrad joined #cdk
17:22 CIA-50 joined #cdk
17:22 jpansanel joined #cdk
17:36 sneumann joined #cdk
19:09 Conrad joined #cdk
19:23 egonw_ joined #cdk
19:45 olass left #cdk
19:52 egonw joined #cdk
19:56 CIA-50 cdk: Egon Willighagen master * rc4b4c48 / build.props : Master is open for patches... - http://bit.ly/9KgStm
20:05 bag_ joined #cdk
20:08 egonw sneumann: ok, my system is sort of back to normal now...
20:08 egonw going to import your plugin, and do a small bit of clean up
20:14 egonw sneumann: ping
20:14 egonw sneumann: I got two commits lined up
20:15 egonw sneumann: check http://github.com/sneumann​/de.ipbhalle.msbi/network
20:15 zarah egonw's link is also http://tinyurl.com/yk5e2hm
20:17 sneumann yes, I see that, and looked at the diffs.
20:18 sneumann so these changes are living in your (cloned) repo
20:18 egonw in two clones, really
20:18 egonw the one at my HD
20:18 egonw and the copy on github
20:18 sneumann Ack. would I pull them, or do you push them.
20:19 sneumann You have access to my repo
20:19 sneumann @github
20:20 egonw you can pull them via the web interface
20:20 sneumann why not command line ?
20:20 egonw fun?
20:20 egonw command line:
20:21 egonw git remote add egonw git://github.com/egonw/de.ipbhalle.msbi.git
20:21 egonw git fetch egonw
20:21 egonw git checkout master
20:21 egonw git merge egonw/master
20:32 egonw ok, that did not work out too well
20:32 egonw but not uncommon :)
20:32 egonw the art of not getting "merge commits" is not trivial :)
20:32 egonw otherwise... well done!
20:32 egonw how long have you been using git now?
20:32 sneumann pure git ?
20:32 egonw or git-svn
20:33 sneumann talking hours
20:33 sneumann git-svn longer.
20:33 sneumann I guess you could remember my first git questions ;-)
20:33 egonw honostly? yeah, sure ! :)
20:33 egonw really, can't remember...
20:34 sneumann http://www.kerrybuckley.org/2008/06​/18/avoiding-merge-commits-in-git/
20:34 zarah sneumann's link is also http://tinyurl.com/ye3grum
20:34 * sneumann is reading that
20:35 egonw looks like a good read
20:35 egonw btw, you have discovered gitk yet?
20:35 egonw and tig?
20:36 egonw tig is quite nice actually
20:37 sneumann egit is not recommended (yet) ?
20:38 egonw apparently, it does not pull well when one has local changes...
20:43 sneumann_ joined #cdk
20:46 sneumann_ joined #cdk
20:55 egonw sneumann_: I am cleaning up depedencies
20:58 sneumann_ Yes, I added what even remotely looked like a dependency,
20:58 sneumann_ Is import or bundle-requred preferred ?
20:59 egonw second
20:59 sneumann_ ack
20:59 egonw I'm almost done
20:59 egonw ha, one left
20:59 egonw and that's the famous INChIPlainTextReader I heard complaints about :)
21:03 egonw ok, pushed
21:04 sneumann_ so I pull that
21:04 egonw I pushed to your repos this time
21:07 sneumann_ arrived.
21:07 sneumann_ the plugin is missing jaxb.jar here
21:08 egonw don't see that...
21:11 sneumann_ keep fingers crossed, booting product
21:13 sneumann_ ... second attempt ...
21:17 sneumann_ ... trying to get metware feature back in ...
21:21 egonw ack
21:21 * egonw is trying to rebase the jchempaint patches on cdk 1.3.2
21:51 sneumann_ fighting Bundle 'org.springframework.osgi.aopalliance.osgi' cannot be resolved
21:51 sneumann_ a set target platform ...
21:52 sneumann_ running product, still
21:52 sneumann_ java.lang.NoClassDefFoundError: de/ipbhalle/metfrag/main/MetFlowConvenience
21:53 sneumann_ http://pastebin.com/m12cca4a1
21:54 sneumann_ metfrag_feature includes plugins n.bc.metfrag and d.i.metfrag
21:55 egonw edit the META-INF/MANIFEST.MF of de.ipbhalle.metfrag
21:56 egonw third line
21:56 egonw ah, no, nvm
21:56 sneumann_ Bundle-Name: de.ipbhale.metfrag ?
21:56 egonw yeah, that's fine
21:56 egonw was misreading
21:56 sneumann_ no, that was already the changed version ;-)
21:57 egonw did you add the de.ipbhalle.metfrag plugin to the feature ?
21:58 sneumann_ yes
21:59 egonw OK, now go to this classpath horror page
21:59 sneumann_ yup
21:59 egonw and compare all screenshots with what you have for de.ipbhalle.metfrag
21:59 egonw @pz site:bioclipse.net classpath horror
21:59 zarah Cannot detect feed type
22:00 egonw google site:bioclipse.net classpath horror
22:00 zarah egonw: http://wiki.bioclipse.net/index​.php%3Ftitle%3DClasspath_Horror
22:02 sneumann_ do I have to add that jaxrpc.jar to the runtime classpath ?!
22:02 sneumann_ The . is there
22:03 sneumann_ fixing the build
22:03 sneumann_ fixing the build configuration
22:04 sneumann_ Hm, missing plugin.xml ?!
22:07 sneumann_ horay: java.lang.NoClassDefFoundError: org/apache/axis/client/Service
22:08 sneumann_ do I now add all "referenced libraries" from d.i.metfrag ? Seems so ...
22:11 egonw add to what?
22:15 sneumann_ build.properties of d.i.metfrag ...
22:15 sneumann_ ... next attempt ..
22:15 egonw yes, think so
22:31 sneumann_ Caused by: java.lang.NoClassDefFoundError: javax/activation/DataSource
22:32 egonw ah, I have seen that one
22:32 egonw not that I remember the solution...
22:33 egonw tig
22:34 egonw meanwhile...
22:34 egonw I am making slow progress rebasing cdk-jchempaint on #cdk 1.3.2
22:34 egonw which I must get done today
22:35 sneumann_ 25mins left
22:35 sneumann_ WTF: java.lang.NoClassDefFoundError: javax.mail.internet.MimeMultipart
22:35 sneumann_ who's using that here ???!
22:35 egonw :)
22:37 sneumann_ > metfrag.calculateMetFragScore()
22:37 sneumann_ horay.
22:37 egonw it works ?
22:37 sneumann_ no, but it doesn't fail ;-)
22:38 egonw hehehe
22:38 egonw yes, important step
22:38 sneumann_ mom, console: Search PPM: 10.0
22:38 sneumann_ ;-)
22:38 egonw looks good
22:38 egonw but, I quite agree...
22:39 egonw the plugin does seem to require some splitting up...
22:39 egonw which will also mean clean up of deps
22:39 egonw cause clean ups

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