Camelia, the Perl 6 bug

IRC log for #cdk, 2009-03-12

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

All times shown according to UTC.

Time Nick Message
00:33 timvdm joined #cdk
01:46 Conrad joined #cdk
01:48 goomba` joined #cdk
01:50 Conrad joined #cdk
02:17 cpyang joined #cdk
03:06 CIA-62 cdk: rajarshi * r14461 /cdk/branches/cdk-1.2.x/src/test/org/openscience​/cdk/smiles/smarts/parser/SMARTSSearchTest.java: Updted the test to check whether c-C occurs in CCc1ccccc1 since though it gives 2 non-unique hits, both of them are identical in that the query c matches the target c and so on.
06:16 carsten joined #cdk
07:20 sneumann joined #cdk
07:27 Gpox joined #cdk
07:43 Gpox joined #cdk
08:26 egonw joined #cdk
08:43 CIA-62 cdk: egonw * r14462 /cdk/branches/cdk-1.2.x/build.props: Release 1.2.0
08:46 CIA-62 cdk: egonw * r14463 /cdk/tags/cdk-1.2.0/: Tagged release 1.2.0
08:46 egonw yeah, yoohoo
08:47 egonw cheers all over the place
08:47 egonw "The Eagle has Landed"
09:13 jonalv joined #cdk
09:14 shk3 joined #cdk
09:32 jbrefort joined #cdk
09:39 Topic for #cdkis now The Chemistry Development Kit (http://cdk.sf.net) | JUnit CDK 1.2.x: 15112, 130 fails + 42 errors (1.14%) | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News Feed: http://beta.friendfeed.com/ro​oms/chemistry-development-kit
09:40 Topic for #cdkis now The Chemistry Development Kit (http://cdk.sf.net) | JUnit CDK 1.2.x: 15112, 130 fails + 42 errors (0.65%) | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News Feed: http://beta.friendfeed.com/ro​oms/chemistry-development-kit
09:40 Topic for #cdkis now The Chemistry Development Kit (http://cdk.sf.net) | JUnit CDK 1.2.x: 13968, 77 fails + 14 errors (0.65%) | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News Feed: http://beta.friendfeed.com/ro​oms/chemistry-development-kit
10:26 CIA-62 cdk: shk3 * r14464 /cdk/branches/jchempaint-primary/src/​main/org/openscience/cdk/controller/ (4 files in 2 dirs): Some minor changes, bug fixes for jcp
10:49 CIA-62 cdk: shk3 * r14465 /jchempaint/trunk/src/main/org/openscience/ (4 files in 3 dirs): Some minor changes, bug fixes for jcp
11:34 CIA-62 cdk: shk3 * r14466 /jchempaint/trunk/src/main/org/openscience/jchempa​int/resources/userhelp_jcp/contain/menuEdit.html: Worked on help
12:20 jpansanel moin
12:20 zarah oh hai jpansanel
12:20 egonw hi jpansanel
12:23 CIA-62 cdk: shk3 * r14467 /cdk/branches/jchempaint-primary/src/​main/org/openscience/cdk/controller/ (ControllerHub.java IChemModelRelay.java): added a setHydrogenCount method to IChemModelRelay
12:28 CIA-62 cdk: shk3 * r14468 /jchempaint/trunk/src/main/​org/openscience/jchempaint/ (2 files in 2 dirs): atom context menu properties settings undoable
12:48 jonalv joined #cdk
12:55 rajarshi joined #cdk
12:56 egonw hi rajarshi!
12:56 rajarshi hi
12:56 zarah oh hai rajarshi
12:56 egonw you don't sleep much, do you?
12:56 rajarshi kthx zarah
12:56 rajarshi :)
12:56 rajarshi i get obsessive sometimes
12:57 rajarshi so it looks like 1.2 is ready to go?
12:57 egonw it's releaed
12:57 egonw on SF
12:57 egonw no email or blog yet though
12:57 rajarshi cool, i'll wait for your blog post
12:57 rajarshi now I can delete my svn dir's and start fresh with git :)
12:59 egonw 'yes :)
12:59 egonw I have set  up two branches
12:59 egonw one for cdk1.2.x bug fixes etc
12:59 egonw and a master
12:59 rajarshi yes, saw the email
12:59 egonw to which I'll port the trunk patches
12:59 rajarshi ok
13:00 rajarshi btw, does all commits to master have to via review?
13:00 egonw yes
13:01 egonw we can't single out trusted developers
13:01 egonw so only the branch master has special powers
13:01 egonw however
13:01 egonw he is also to one people complain to
13:01 rajarshi :)
13:01 egonw and responsible for delivering something sane
13:01 rajarshi btw, i also see that you said that devs can publish their branches on SF
13:02 egonw SF developers have a private web space
13:02 egonw nowadays
13:02 egonw have not used it though
13:02 egonw and limited to 100MB I think
13:02 rajarshi hmm, my plan was to have a web visible guha-master derived from master
13:02 rajarshi with local brnches for of guha-master and then periodically merge these into guha-master and ask for review
13:03 egonw yes, that sounds like the way forward
13:03 egonw oh, btw...
13:03 egonw the branch master will have to do that too
13:03 egonw but I will likely just use pele for that
13:03 rajarshi ok
13:04 egonw because all my patches need review too
13:04 rajarshi so 'branch master' is really only responsible for puting in changes onto SF master
13:04 rajarshi it so happens egonw and 'branch mster' are the same person
13:04 egonw yes
13:05 egonw no special power to me as developer
13:05 rajarshi it would also be useful to send out a notice indicating that people should announce what they're working on - since it's now distributed it'd be difficult to keep track of what's being done without explicit announcements
13:05 egonw yes, maybe
13:05 egonw we'll see how this develops...
13:07 shk3 hi egonw
13:07 egonw hej shk3
13:07 shk3 do you think you will work on the modular hub?
13:08 egonw yes, asap
13:08 egonw arvid will help me
13:08 egonw he had some good ideas
13:08 egonw which actually closely overlaps how the old JCP did it, if I understood correctly
13:08 shk3 sounds good
13:08 egonw with each method from the current hub an action class
13:08 egonw which could be undone too
13:08 shk3 I think this is an essential thing
13:08 egonw yes, it is
13:09 shk3 yes, roughly that't the equivalence
13:09 shk3 the methods in hub should all be undoable
13:09 shk3 I am just making sure they get used everywhere
13:09 shk3 2 or 3 classes left
13:10 shk3 but then everything should be allright in that respect
13:10 egonw ok, good
13:18 CIA-62 cdk: shk3 * r14469 /cdk/branches/jchempaint-primary/src/main/org/​openscience/cdk/controller/ControllerHub.java: setoorder now undoable
13:19 CIA-62 cdk: shk3 * r14470 /jchempaint/trunk/src/main/​org/openscience/jchempaint/ (2 files in 2 dirs): bond context menu properties settings undoable
13:28 shk3 egonw: I am still/again getting ClassCastErrors when trying to write in JCP.
13:29 shk3 Did you work on that?
13:29 egonw yes, I did
13:29 shk3 hm, ok
13:30 shk3 AtomContainer cannot be cast to Molecule
13:30 maclean joined #cdk
13:30 shk3 is what I get now
13:30 shk3 I will have a look
13:30 egonw hi maclean!
13:30 maclean hej
13:30 egonw shk3: please file a bug report with stacktrace
13:30 maclean I wanted to ask you about 'best practice' for setting up a development branch of cdk with git
13:31 egonw hahaha
13:31 egonw yes... what do you want to know?
13:31 maclean well, I want to branch 1.2.x (cutting edge, yes?)
13:31 maclean to do structure generation stuff in
13:32 egonw git checkout -b structureGenerationStuff cdk-1.2.x
13:33 egonw after the git clone call
13:33 maclean ah ok
13:33 maclean but (from your mail) I will only 'check in' locally, not push
13:34 maclean until its at the stage that I can sumbit a patch for review, yes?
13:34 maclean (with git-format-patch?)
13:35 egonw yes
13:35 egonw but meanwhile you can put it up on a local web server
13:35 maclean right.
13:35 egonw so that other can monitor what you do
13:35 egonw and start test things...
13:35 maclean that's  kind of what I was thinking
13:36 maclean I guess I could use my github account
13:36 maclean hmmm
13:36 maclean okay, right I will get on and do that - tack egonw
13:37 egonw shk3: please tell maclean that I'm adding documentation to the wiki as we speak
13:37 egonw and that all questions are justified...
13:39 shk3 egonw: problem is that write(IChemObject) calls writeMoleculeSet
13:39 shk3 which casts to Molecle
13:39 shk3 the code there is missing generics anyway
13:40 shk3 and using them it would naturally become IAtomContainer
13:40 shk3 I can't see why not to use this
13:41 egonw ok, please file a bug report with stacktrace
13:42 egonw shk3: please give gilleain this link: https://apps.sourceforge.net/mediawiki/c​dk/index.php?title=Development_with_Git
13:42 egonw I have updated it regarding his qustions
13:42 egonw afk now
14:43 rajarshi egonw, question: What is the role of IMolecule?
14:51 egonw hahaha...
14:51 egonw back behind keyboard...
14:51 egonw that question keeps poping up all the time
14:52 egonw IAtomCOntainer is somethng really genereal
14:52 egonw an IReaction can be represented in a single IAtomCotnainer
14:52 egonw I PDB file can too...
14:52 egonw I=a
14:52 rajarshi it seems that IMolecule is just syntactic sugar
14:52 rajarshi it provides no extra stuff over IAtomContainer
14:55 egonw no, the idea that IMolecule is an isolated fully connected graph is not enforced
14:55 egonw but implied
14:55 egonw maybe it should enforce it
14:55 rajarshi right now the API says nothing about fully connected graphs etc
14:56 rajarshi I definitely think that if this is the goal it should be enforced
14:56 egonw there should be some discussion on this in the ML archives, I think
14:59 rajarshi btw, would it be OK to include the IDEA project files in the repo? Since it uses relative paths anybody using IDEA can use them (just need to change the compiler path)
15:00 CIA-62 cdk: shk3 * r14471 /jchempaint/trunk/src/main/org/openscien​ce/jchempaint/action/SaveAsAction.java: remembers file name
15:20 egonw rajarshi: yes, I think that's fine
15:23 rajarshi ok
15:23 rajarshi also i did a clone of the repo
15:23 rajarshi does that give me master?
15:23 egonw both, I think
15:23 egonw try:
15:23 rajarshi i also see that I can get origina/master origin/HEAD origin/cdk-1.2.x
15:23 egonw git branch -a
15:24 egonw right
15:24 rajarshi aah, ok I have all of them
15:24 rajarshi so master is trunk, right?
15:24 egonw your local master == origin/master
15:24 egonw yes
15:24 egonw SVN:trunk = Git:master
15:24 egonw SVN:trunk owl:sameAs Git:master
15:24 rajarshi so then what is origin/master ?
15:24 egonw that's the master branch at SF
15:24 rajarshi so origin/master == master
15:25 egonw your 'master' is your local copy if origin/master
15:25 egonw no, it's a local copy :)
15:25 egonw meaning that you can commit stuff locally offline....
15:25 egonw and submit that later
15:25 rajarshi oh ok, so i'd push changes to master and you'd pull from my master
15:25 egonw yes, indeed
15:25 rajarshi ok, got it
15:26 rajarshi is the vflib branch in git? or is that still insvn?
15:26 egonw svn
15:26 rajarshi aah, ok
15:27 rajarshi idea has really improved git handling
15:28 egonw I do not use the git plugin for eclipse yet
15:28 rajarshi good for refactoring
15:28 egonw but still do things from the command line
15:28 rajarshi going to start on a branch now with somesmall change and see how the workflow works out
15:29 egonw sounds good
16:34 shk3 joined #cdk
16:47 egonw joined #cdk
17:57 CIA-62 cdk: shk3 * r14472 /jchempaint/trunk/ (7 files in 4 dirs): more help
19:09 jbrefort joined #cdk
19:24 edrin joined #cdk
20:06 egonw joined #cdk
20:07 timvdm joined #cdk
20:50 bag joined #cdk

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