Camelia, the Perl 6 bug

IRC log for #cdk, 2010-09-20

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

All times shown according to UTC.

Time Nick Message
00:04 s9asad left #cdk
03:11 azeem_ joined #cdk
03:15 azeem left #cdk
05:03 egonw joined #cdk
05:26 sneumann joined #cdk
06:14 egonw left #cdk
06:17 jbrefort joined #cdk
06:58 Gpox joined #cdk
07:10 mgerlich joined #cdk
07:11 jbrefort left #cdk
07:11 s_wolf joined #cdk
08:00 egonw joined #cdk
08:01 mgerlich left #cdk
08:04 mgerlich joined #cdk
08:11 s_wolf left #cdk
08:13 s_wolf joined #cdk
10:34 maclean joined #cdk
10:34 maclean hi
10:34 zarah privet maclean
11:39 egonw hi maclean
11:39 maclean hi
11:39 zarah saluton maclean
11:56 maclean layout is hard
11:59 egonw yes, but that makes the patch only more interesting :)
12:01 maclean http://gilleain.tumblr.com/post/1155550413/cus​tom-axis-layout-ring-bonds-are-badly-drawn-due
12:01 zarah maclean's link is also http://tinyurl.com/358wqxk
12:02 egonw yeah, something fishy with inner bond drawing indeed!
12:02 egonw otherwise nice
12:02 maclean http://gist.github.com/587795
12:02 egonw WaterFallLayout ?
12:02 maclean is the code
12:03 maclean heh :)
12:03 maclean Am working on GridMoleculeSetLayout, which is probably more useful.
12:03 egonw true
12:05 maclean The main difficulty I have is whether layouts for molecule sets should transform (translate) the underlying atomcontainer positions.
12:06 maclean Or just create a list (map?) of AffineTransforms or maybe just boxes.
12:07 egonw affine transforms I would say
12:07 maclean less destructive to the individual models, I agree.
12:10 maclean afk
12:14 egonw indeed
12:19 egonw maclean: oh, and this: http://chem-bla-ics.blogspot.com/20​10/09/noels-chemical-ascii-art.html
12:19 zarah egonw's link is also http://tinyurl.com/25adgc4
12:22 maclean left #cdk
12:31 maclean joined #cdk
12:34 maclean egonw : I saw. ASCII pics are...rarely useful.
12:35 maclean although...http://everything2.com/title/Impossible+triangle
13:32 egonw :)
13:38 egonw left #cdk
14:51 maclean left #cdk
15:22 s_wolf left #cdk
15:23 s_wolf joined #cdk
15:28 slyrus so is there a jcp that works with cdk 1.3.6 yet?
15:30 slyrus (or a non-jcp renderer?)
15:30 slyrus 2d that is
15:37 maclean joined #cdk
15:38 maclean slyrus : I honestly have no idea. Egon may be working on that, I'm not sure.
15:38 slyrus working on jcp w/ 1.3.6?
15:39 maclean One day - one fine day - it will all be integrated (again) and work (again).
15:39 maclean rendering code with 1.3.6, yes.
15:39 slyrus ok. I'd be perfectly happy with a 2d renderer that stayed in sync with cdk that had nothing to do with the rest of the JCP app :)
15:39 maclean yes, that is the most sensible first step.
15:40 maclean then JCP could (if the current developers wanted to) be based on that.
15:50 slyrus that would be nice
15:51 slyrus and if it worked well (in particular if it paid attention to E/Z configurations and didn't get confused w.r.t chiral centers) I could abandon my efforts to write my own :)
15:52 maclean in lisp?
15:52 slyrus clojure
15:52 maclean ah.
15:52 slyrus http://slyrus.github.com/2010/09/1​7/parsing-smiles-with-fnparse.html
15:52 slyrus describes the first part of that effort, the SMILES parser
15:52 slyrus the nice thing about clojure is that I can more easily go back and forth to CDK
15:52 slyrus but, yes, I could probably abandon my lisp-based efforts as well :)
15:52 slyrus CL that is
15:53 maclean yes
15:53 maclean "While this is a nice idea..." :)
15:53 maclean I'm finding the dangers of 'nice ideas' at the moment, with hierarchical renderers, and hierarchical layouts.
15:58 slyrus in the end, I think this particular nice idea forced jchoi to build some nice infrastructure that I was able to use make a parser that is much cleaner than my previous efforts with lots of mutable hash-tables in CL. I didn't mean for that to sound too pejorative...
16:00 maclean it's not so pejorative. It's good to have nice ideas, but it's also good to have working code :)
16:28 s_wolf left #cdk
16:48 jbrefort joined #cdk
17:23 Gpox left #cdk
18:16 egonw joined #cdk
19:03 egonw moin
19:03 zarah privet egonw
19:29 maclean what the hell, blogger? "<imoleculeset>where </imoleculeset>"??
19:29 egonw huh?
19:30 maclean switching from 'compose' to 'edit html' gave me those tags.
19:30 maclean I'm pretty sure that blogger is not using some kind of semantic CDK markup....
19:31 egonw oh, yeah... agreed... that switching is a mess...
19:36 maclean Ohhhh. It's generics. Dammit. It's trying to close the Layout<IMoleculeSet> 'tag' :)
19:40 maclean There you go, egon, blogged.
19:40 egonw tack
19:40 egonw btw, did not get around to cdk-jchempaint
19:40 egonw finishing up two book chapters
19:40 maclean sure.
19:41 maclean Is it possible that I could do it?
19:43 egonw yes, I think so
19:44 maclean Er, how? (Roughly). I swear I won't ask git questions...
19:44 egonw yeah, was just looking up the script I am using...
19:48 egonw sent
19:49 maclean tack
19:50 egonw varsågod
20:21 egonw cu tomorrow
20:22 maclean see you
20:26 egonw left #cdk
20:42 maclean left #cdk
21:05 jbrefort left #cdk
23:30 _ilbot2 joined #cdk
23:30 Topic for #cdk is now The Chemistry Development Kit (http://cdk.sf.net) | JUnit CDK 1.2.7: 15690, 76 fails + 15 errors (0.58%) | Logs at http://moritz.faui2k3.org/​irclog/out.pl?channel=cdk | CDK News Feed: http://friendfeed.com/rooms​/chemistry-development-kit

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