Camelia, the Perl 6 bug

IRC log for #cdk, 2009-04-07

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

All times shown according to UTC.

Time Nick Message
06:07 jbrefort joined #cdk
06:27 Gpox joined #cdk
06:32 Gpoks joined #cdk
06:32 Gpox joined #cdk
07:04 egonw joined #cdk
07:05 edrin joined #cdk
07:53 jonalv joined #cdk
08:19 shk3 joined #cdk
10:37 maclean joined #cdk
10:39 maclean hi egonw, shk3 : with regards to ola's stereochemistry question, there is an 'AtomParity' class, but it doesn't seem to be used anywhere.
10:39 egonw it's the data model behind stereochemistry
10:39 egonw but AFAIK indeed not used
10:39 egonw not in IO anyway
10:40 shk3 that's one of the tasks I put on this new features brainstorming page
10:40 shk3 I will put olas reqeust there as well
10:40 egonw yeah, that must be urgently fixed nideed
10:40 shk3 somebody else asked for this as well some time ago
10:40 egonw but Todd's patch was for symmetry detection...
10:40 maclean I had talked to Kyrill about using the cdk to determine the stereochemistry of molecules, in order to avoid impossible layouts, which is another use-case
10:40 egonw but no time today to check it out
10:42 maclean http://metallome.blogspot.com/2​009/02/impossible-figures.html
10:42 zarah maclean's link is also http://tinyurl.com/d3ce5q
10:48 shk3 I don't even remember that patch you are referring to
10:49 maclean I thought it was crystal symmetry, though.
10:51 egonw oh, maybe it was...
10:51 egonw then it would certainly not help
10:51 egonw well, a quick and dirty algorithm should work
10:52 egonw particularly if we only support carbons...
10:52 egonw what other atom types can be chiral, egonw wonders
10:52 egonw Pt
10:52 egonw I guess a lot of metals
10:54 maclean yes
10:54 maclean oxygens?
10:54 maclean anything that can have four groups attached, I would suppose
10:54 egonw don't think so...
10:54 egonw LPs are too movable, I think
10:54 maclean warum nicht?
10:54 maclean oh, right.
10:55 maclean of course, there are more complex types of chirality... :)
10:56 maclean but, as you say - a simple solution would be good (and better than no solution)
10:58 egonw I'd be happy with detection of atom based chirality
10:58 egonw I'd be happy with detection of atom based stereochemistry chirality
11:04 shk3 there are tricky cases of course
11:04 shk3 mesoverbindungen and whatever
11:04 maclean mesoverbindungen?
11:06 maclean ah, okay. I googled it
11:07 egonw shk3: true, but they still have stereo centers
11:07 shk3 true
11:07 egonw enumeration of all stereo isomers is the next step
11:07 shk3 so stereo center detection would find them
11:07 shk3 stereo molecule detection not
11:31 shk3 joined #cdk
11:42 egonw shk3: do you know who wrote the canonical SMILES code?
12:07 maclean joined #cdk
12:09 shk3 no
12:09 shk3 was there when I did the chiral stuff
12:09 egonw I will try to recover that information...
12:10 egonw I'll ask Chris
12:12 maclean Well, he's only sitting over there -> so we could ask him :)
12:13 maclean (you would have to know the orientation of my screen, and position of my laptop to interpret that arrow...)
12:13 maclean his desk is actually over there <-
12:13 egonw hahaha...
12:13 egonw but there are walls involved too, not?
12:13 maclean not any more
12:14 maclean we all moved to a portacabin yesterday
12:14 egonw ah, ic...
12:15 egonw shk3 told me something about that indeed
12:22 maclean egonw : he says that he remembers who wrote the code, but it was used for many years in NMRShiftDB without trouble - are we certain that there is a bug?
12:23 egonw tell him, yes I am
12:24 maclean oh, well I guess he'll reply to the mail with the name of the author.
12:25 maclean sorry, that wasn't particularly helpful of me, I suppose :/
14:02 ilbot2 joined #cdk
14:02 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
15:52 egonw joined #cdk
16:05 maclean heh. egonw : when I said 'in general quite hard' I didn't really mean the engineering difficulty of hooking up the code :)
16:05 maclean but I suppose, given the ability to minimize the structure, it could be doable.
16:14 rajarshi joined #cdk
20:13 egonw rajarshi: do you know how to signof commits?
20:14 rajarshi not really
20:14 egonw ha, found it
20:14 egonw git commit -s --amend
20:14 rajarshi doesn't that overwrite the original committer?
20:15 egonw no
20:15 rajarshi i've seen the use of the --signoff arg but I think it applies to mbox style patches
20:16 egonw I thought it is generally applicable
20:16 egonw to indicate that a patch is OK
20:17 rajarshi oh right. -s == --signoff
20:17 rajarshi sheesh, git is the C of VCS's :)
20:18 egonw http://cdk.git.sourceforge.net/git/gitweb.c​gi?p=cdk;a=shortlog;h=refs/heads/cdk-1.2.x
20:18 zarah egonw's link is also http://tinyurl.com/coyy7s
20:18 egonw took me a bit of effort to backport to 1.2.x
20:19 egonw in particular because I kept trying to merge the master one :(
20:19 egonw which gives very many conflicts :)
20:19 rednael [cdk-1.2.x] Added a test to ensure that Integer objects are compared by value rather than reference  http://cdk.git.sourceforge.net/git/g​itweb.cgi?p=cdk;a=commitdiff;h=5134e​180c999e77cb1d5c7fce07644ec847f1b90
20:19 rednael [cdk-1.2.x] Added a test case to check that atom container diffs are correct when using deseriali...   http://cdk.git.sourceforge.net/git/g​itweb.cgi?p=cdk;a=commitdiff;h=b6e67​08bf730ecad40609e076ac7b6ea66f9bf0a
20:19 rednael [cdk-1.2.x] Fixed IntegerDifference so that it actually checks the integer value rather than...   http://cdk.git.sourceforge.net/git/g​itweb.cgi?p=cdk;a=commitdiff;h=cc0ff​c73a13509a9952459061a0f3a8c6fae6010
20:19 zarah rednael's link is also http://tinyurl.com/cfvpvn
20:19 zarah rednael's link is also http://tinyurl.com/coqzpx
20:19 zarah rednael's link is also http://tinyurl.com/cu93jx
20:19 egonw ah, hi rednael
20:19 egonw forgot about you :)
20:19 rajarshi so was this to 1.2.x? or master/
20:21 egonw 1.2.x
20:21 egonw bug fix after all
20:21 rajarshi oh right - how do you set it to automatically merge into master?
20:21 egonw that's manual
20:21 rajarshi oh ok
20:22 egonw but I forgot how I did that...
20:23 rajarshi :)
20:23 egonw because an actual merge gives conflicts...
20:23 egonw ok, going to bed now
20:23 egonw will do that tomorrow
20:23 rajarshi good night
20:52 edrin joined #cdk
22:16 timvdm joined #cdk

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